Refine
Year of publication
Document Type
- Conference Proceeding (236)
- Article (177)
- Part of a Book (147)
- Book (18)
- Part of Periodical (5)
- Other (3)
- Working Paper (3)
- Image (1)
Language
- English (590) (remove)
Keywords
- Korpus <Linguistik> (173)
- Deutsch (126)
- Computerlinguistik (73)
- Interaktion (53)
- Konversationsanalyse (51)
- Annotation (42)
- Forschungsdaten (37)
- Gesprochene Sprache (36)
- Datenmanagement (33)
- Automatische Sprachanalyse (32)
Publicationstate
- Veröffentlichungsversion (590) (remove)
Reviewstate
- Peer-Review (590) (remove)
Publisher
- IDS-Verlag (78)
- Association for Computational Linguistics (34)
- Institut für Deutsche Sprache (20)
- European Language Resources Association (ELRA) (19)
- European Language Resources Association (18)
- Linköping University Electronic Press (15)
- Taylor & Francis (14)
- European language resources association (ELRA) (13)
- Lexical Computing CZ s.r.o. (13)
- Springer Nature (13)
This paper describes the lexical database tool LOLA (Linguistic-Oriented Lexical database Approach) which has been developed for the construction and maintenance of lexicons for the machine translation system LMT. First, the requirements such a tool should meet are discussed, then LMT and the lexical information it requires, and some issues concerning vocabulary acquisition are presented. Afterwards the architecture and the components of the LOLA system are described and it is shown how we tried to meet the requirements worked out earlier. Although LOLA originally has been designed and implemented for the German-English LMT prototype, it aimed from the beginning at a representation of lexical data that can be reused for other LMT or MT prototypes or even other NLP applications. A special point of discussion will therefore be the adaptability of the tool and its components as well as the reusability of the lexical data stored in the database for the lexicon development for LMT or for other applications.
In this paper, we analyze a dramatically aggravated conflict interaction taking place in the course of an association’s meeting in an urban community center. The interaction can be seen as the culmination point of a social conflict developing and increasing over a period of years. In this conflict, one of the crucial points of the sociocultural development in the city under study is to be seen in an exemplary way. Our analysis started with the question, why this conflict is unsolvable although the interest divergences of the opposing parties are not irreconcilable. Our analysis shows that the protagonists practice different communicative social styles. These stylistic differences however, are not the cause for misunderstandings, but the protagonists use stylistic differences and different cultural orientations as a resource for political action. Thereby a process of increasing hardening of perspective divergence emerges together with an interaction modality of drama and of the fundamental grounding of divergent views. Theoretically we are concerned with the explication of a sociolinguistic theory which includes as constitutive components the concepts of communicative social style, of perspectivation and of interaction modality. We want to show, that the analyzed type of sociocultural conflict can be explained by virtue of considering the interplay of features on these three levels.
Online Access Tools for Spoken German: The Resources of the Deutsches Spracharchiv in a Database
(2002)
This paper shows some details of the modernization of the Deutsches Spracharchiv (DSAv). It explores some future possibilities of linguistical documentation and analysis using the Web. The Institut für Deutsche Sprache (IDS) in Mannheim is the central institution for linguistic research in Germany. The DSAv in the IDS is the center for documentation and research of spoken German. These archives include the largest collection of sound recordings of spoken German (dialects and colloquial speech, including e.g. lots of extinct dialects of former German territories in Eastern Europe) - altogether more than 15,000 sound recordings. The lacking clarification and accessibility of this data material has been felt as an essential deficit. The opportunity to edit the sound signal digitally offers a much easier access to spoken language. Through the integration of the already existing information about the corpora and the transcribed texts in an information- and full text databank, as well as the linking of the data with the acoustic signal (alignment), arises a data-pool with considerably better documentation of the materials and a fast direct grasp of the recorded sounds. Thus, the DSAv initiates totally new research questions for the work at the IDS, as well as for linguistics altogether.
The classification of verbs in Levin's (1993) English Verb Classes and Alternations: A preliminary Investigation, on the basis of both intuitive semantic grouping and their participation in valence alternations, is often used by the NLP community as evidence of the semantic similarity of verbs (Jing & McKeown 1998; Lapata & Brew 1999; Kohl et al. 1998). In this paper, we compare the Levin classification with the work of the FrameNet project (Fillmore & Baker 2001), where words (not just verbs) are grouped according to the conceptual structures (frames) that underlie them and their combinatorial patterns are inductively derived from corpus evidence. This means that verbs grouped together in FrameNet (FN) might be semantically similar but have different (or no) alternations, and that verbs which share the same alternation might be represented in two different semantic frames.
In this paper, we investigate the practical applicability of Co-Training for the task of building a classifier for reference resolution. We are concerned with the question if Co-Training can significantly reduce the amount of manual labeling work and still produce a classifier with an acceptable performance.
We describe a simple and efficient Java object model and application programming interface (API) for (possibly multi-modal) annotated natural language corpora. Corpora are represented as elements like Sentences, Turns, Utterances, Words, Gestures and Markables. The API allows linguists to access corpora in terms of these discourse-level elements, i.e. at a conceptual level they are familiar with, with the flexibility offered by a general purpose programming language. It is also a contribution to corpus standardization efforts because it is based on a straightforward and easily extensible data model which can serve as a target for conversion of different corpus formats.
We present a light-weight tool for the annotation of linguistic data on multiple levels. It is based on the simplification of annotations to sets of markables having attributes and standing in certain relations to each other. We describe the main features of the tool, emphasizing its simplicity, customizability and versatility
We apply a decision tree based approach to pronoun resolution in spoken dialogue. Our system deals with pronouns with NP- and non-NP-antecedents. We present a set of features designed for pronoun resolution in spoken dialogue and determine the most promising features. We evaluate the system on twenty Switchboard dialogues and show that it compares well to Byron’s (2002) manually tuned system.
Reframing FrameNet Data
(2004)
The Berkeley FrameNet Project (http://www.icsi.berkeley.edu/~framenet) is building an on-line lexical resource for contemporary English. The database provides information about the semantic and syntactic combinatorial possibilities (valences) of each item analyzed. This paper describes the conceptual basis for what has been called reframing of data in the FrameNet database and exemplifies two new frame-to-frame relations, Causative_of and Inchoative_of, the implementation of which came about as a result of reanalysis of certain frames and lexical units. The new relations are characterized with respect to a triple of frames involving the notion of attaching, and entering them into the database is demonstrated using the Frame Relations Editor. The two relations allow FrameNet to make frame-wise distinctions that capture fairly systematic semantic relationships across sets of lexical units. While the Inheritance and Subframe relations are of particular interest to the NLP research community, Causative_of and Inchoative_of may be more relevant to lexicography.
In this paper, we present the Multiple Annotation approach, which solves two problems: the problem of annotating overlapping structures, and the problem that occurs when documents should be annotated according to different, possibly heterogeneous tag sets. This approach has many advantages: it is based on XML, the modeling of alternative annotations is possible, each level can be viewed separately, and new levels can be added at any time. The files can be regarded as an interrelated unit, with the text serving as the implicit link. Two representations of the information contained in the multiple files (one in Prolog and one in XML) are described. These representations serve as a base for several applications.
We present an implemented XML data model and a new, simplified query language for multi-level annotated corpora. The new query language involves automatic conversion of queries into the underlying, more complicated MMAXQL query language. It supports queries for sequential and hierarchical, but also associative (e.g. coreferential) relations. The simplified query language has been designed with non-expert users in mind.
This article provides an introduction to elexiko, the first German hypertext dictionary to be compiled on a corpus basis, which is currently being developed at the Institut für Deutsche Sprache Mannheim (IDS). First, a brief account of the design is given, followed by a demonstration of the methods and tools that are being employed to compile it. elexiko will provide not only an improved quantity of lexical information, but also a new quality of information which will be explained and illustrated at different levels of the microstructure of the dictionary. The description of word meaning and use in elexiko will be presented in detail, with a particular focus on the treatment of collocations, ambiguity, vagueness, and the presentation of senses. The development of a theoretically grounded procedure for lexicographic disambiguation is also described. This is then followed by a brief account of the treatment of grammatical details. Finally, issues of usability, the progress of the project and its future perspectives will be considered.
We present an implemented machine learning system for the automatic detection of nonreferential it in spoken dialog. The system builds on shallow features extracted from dialog transcripts. Our experiments indicate a level of performance that makes the system usable as a preprocessing filter for a coreference resolution system. We also report results of an annotation study dealing with the classification of it by naive subjects.
Jaw and Order
(2007)
It is well-accepted that the jaw plays an active role in influencing vowel height. The general aim of the current study is to further investigate the extent to which the jaw is active in producing consonantal distinctions, with specific focus on coronal consonants. Therefore, tongue tip and jaw positions are compared for the German coronal consonants Is, J, t, d, n, 1/, that is, consonants having the same active articulators (apical/laminal) but differing in manner of articulation. In order to test the stability of articulatory positions for each of these coronal consonants, a natural perturbation paradigm was introduced by recording two levels of vocal effort: comfortable, and loud without shouting. Tongue and jaw movements of five speakers of German were recorded by means of EMMA during /aCa/ sequences. By analyzing the tongue tip and jaw positions and their spatial variability we found that (1) the jaw's contribution to these consonants varies with manner of articulation, and (2) for all coronal consonants the positions are stable across loudness conditions except for those of the nasal. Results are discussed with respect to the tasks of the jaw, and the possible articulatory adjustments that may accompany louder speech.
We present an implemented system for the resolution of it, this, and that in transcribed multi-party dialog. The system handles NP-anaphoric as well as discourse-deictic anaphors, i.e. pronouns with VP antecedents. Selectional preferences for NP or VP antecedents are determined on the basis of corpus counts. Our results show that the system performs significantly better than a recency-based baseline.
This paper aims to address these problems by dealing with theoretical and methodological questions concerning the national effects of the Bologna Process and the role national factors play in determining the impact of these effects. Altogether the purpose of the paper is to serve as a starting point for future research – both as a guide for systematic and comparative empirical work on higher education, but also for further theoretical and methodological reasoning concerning research on (higher) education policy. As higher education research so far particularly lacks an approach allowing for a competitive and systematic falsification of theoretical arguments by clearly indicating testable and specific hypothesis as well as variables behind the research design (Goedegebuure/Vught 1996) we propose to fall back on neighbouring disciplines, namely social science to improve and enhance the analysis (Slaughter 2001: 398; Altbach 2002: 154; Teichler 1996a: 433, 2005: 448). Several strands of research have to be considered – namely literature on Europeanization as well as insights and approaches of studies dealing with cross-national policy convergence. Taking into account the non-obligatory and mainly intergovernmental character of the Bologna Process the main focus of the paper is on factors related to the effects of transnational communication. The inherent goal is to extend the research agenda on higher education (McLendon 2003: 184ff) and to leave behind the restriction of to analyse only a few cases by striving for a research design that allows for systematic testing and sufficient explanations of cross-national policy convergence at the interface between the Bologna Process and domestic factors.
In our study we use the experimental framework of priming to manipulate our subjects’ expectations of syllable prominence in sentences with a well-defined syntactic and phonological structure. It shows that it is possible to prime prominence patterns and that priming leads to significant differences in the judgment of syllable prominence.
Research on syntactic ambiguity resolution in language comprehension has shown that subjects' processing decisions are influenced by a variety of heterogeneous factors such as e.g., syntactic complexity, semantic fit and the discourse frequency of the competing structures. The present paper investigates a further potentially relevant factor in such processes: effects of syntagmatic lexical chunking (or matching to a complex memorized prefab) whose occurrence would be predicted from usage-based assumptions about linguistic categorisation. Focusing on the widely studied so-called DO/SC-ambiguity in which a post-verbal NP is syntactically ambiguous between a direct object and the subject of an embedded clause, potentially biasing collocational chunks of the relevant type are identified in a number of corpus-linguistic pretests and then investigated in a self-paced reading experiment. The results show a significant increase in processing difficulty from a collocationally neutral over a lexically biasing to a strongly biasing condition. This suggests that syntagmatically complex and partially schematic templates of the kind envisioned in usage-based Construction Grammar may impinge on speakers' online processing decisions during sentence comprehension.
Introduction
(2008)
Current Natural Language Processing (NLP) systems feature high-complexity processing pipelines that require the use of components at different levels of linguistic and application specific processing. These components often have to interface with external e.g. machine learning and information retrieval libraries as well as tools for human annotation and visualization. At the UKP Lab, we are working on the Darmstadt Knowledge Processing Software Repository (DKPro) (Gurevych et al., 2007a; Müller et al., 2008) to create a highly flexible, scalable and easy-to-use toolkit that allows rapid creation of complex NLP pipelines for semantic information processing on demand. The DKPro repository consists of several main parts created to serve the purposes of different NLP application areas
Lexicon schemas and their use are discussed in this paper from the perspective of lexicographers and field linguists. A variety of lexicon schemas have been developed, with goals ranging from computational lexicography (DATR) through archiving (LIFT, TEI) to standardization (LMF, FSR). A number of requirements for lexicon schemas are given. The lexicon schemas are introduced and compared to each other in terms of conversion and usability for this particular user group, using a common lexicon entry and providing examples for each schema under consideration. The formats are assessed and the final recommendation is given for the potential users, namely to request standard compliance from the developers of the tools used. This paper should foster a discussion between authors of standards, lexicographers and field linguists.
In this paper we investigate the coverage of the two knowledge sources WordNet and Wikipedia for the task of bridging resolution. We report on an annotation experiment which yielded pairs of bridging anaphors and their antecedents in spoken multi-party dialog. Manual inspection of the two knowledge sources showed that, with some interesting exceptions, Wikipedia is superior to WordNet when it comes to the coverage of information necessary to resolve the bridging anaphors in our data set. We further describe a simple procedure for the automatic extraction of the required knowledge from Wikipedia by means of an API, and discuss some of the implications of the procedure’s performance.
In this paper, we present a suite of flexible UIMA-based components for information retrieval research which have been successfully used (and re-used) in several projects in different application domains. Implementing the whole system as UIMA components is beneficial for configuration management, component reuse, implementation costs, analysis and visualization.
“Linguistic Landscapes” (LL) is a research method which has become increasingly popular in recent years. In this paper, we will first explain the method itself and discuss some of its fundamental assumptions. We will then recall the basic traits of multilingualism in the Baltic States, before presenting results from our project carried out together with a group of Master students of Philology in several medium-sized towns in the Baltic States, focussing on our home town of Rēzekne in the highly multilingual region of Latgale in Eastern Latvia. In the discussion of some of the results, we will introduce the concept of “Legal Hypercorrection” as a term for the stricter compliance of language laws than necessary. The last part will report on advantages of LL for educational purposes of multilingualism, and for developing discussions on multilingualism among the general public.
In opinion mining, there has been only very little work investigating semi-supervised machine learning on document-level polarity classification. We show that semi-supervised learning performs significantly better than supervised learning when only few labelled data are available. Semi-supervised polarity classifiers rely on a predictive feature set. (Semi-)Manually built polarity lexicons are one option but they are expensive to obtain and do not necessarily work in an unknown domain. We show that extracting frequently occurring adjectives & adverbs of an unlabeled set of in-domain documents is an inexpensive alternative which works equally well throughout different domains.
Though polarity classification has been extensively explored at document level, there has been little work investigating feature design at sentence level. Due to the small number of words within a sentence, polarity classification at sentence level differs substantially from document-level classification in that resulting bag-of-words feature vectors tend to be very sparse resulting in a lower classification accuracy.
In this paper, we show that performance can be improved by adding features specifically designed for sentence-level polarity classification. We consider both explicit polarity information and various linguistic features. A great proportion of the improvement that can be obtained by using polarity information can also be achieved by using a set of simple domain-independent linguistic features.
We present MaJo, a toolkit for supervised Word Sense Disambiguation (WSD), with an interface for Active Learning. Our toolkit combines a flexible plugin architecture which can easily be extended, with a graphical user interface which guides the user through the learning process. MaJo integrates off-the-shelf NLP tools like POS taggers, treebank-trained statistical parsers, as well as linguistic resources like WordNet and GermaNet. It enables the user to systematically explore the benefit gained from different feature types for WSD. In addition, MaJo provides an Active Learning environment, where the
system presents carefully selected instances to a human oracle. The toolkit supports manual annotation of the selected instances and re-trains the system on the extended data set. MaJo also provides the means to evaluate the performance of the system against a gold standard. We illustrate the usefulness of our system by learning the frames (word senses) for three verbs from the SALSA corpus, a version of the TiGer treebank with an additional layer of frame-semantic annotation. We show how MaJo can be used to tune the feature set for specific target words and so improve performance for these targets. We also show that syntactic features, when carefully tuned to the target word, can lead to a substantial increase in performance.
This paper introduces LRTwiki, an improved variant of the Likelihood Ratio Test (LRT). The central idea of LRTwiki is to employ a comprehensive domain specific knowledge source as additional “on-topic” data sets, and to modify the calculation of the LRT algorithm to take advantage of this new information. The knowledge source is created on the basis of Wikipedia articles. We evaluate on the two related tasks product feature extraction and keyphrase extraction, and find LRTwiki to yield a significant improvement over the original LRT in both tasks.
This paper describes the application of probabilistic part of speech taggers to the Dzongkha language. A tag set containing 66 tags is designed, which is based on the Penn Treebank. A training corpus of 40,247 tokens is utilized to train the model. Using the lexicon extracted from the training corpus and lexicon from the available word list, we used two statistical taggers for comparison reasons. The best result achieved was 93.1% accuracy in a 10-fold cross validation on the training set. The winning tagger was thereafter applied to annotate a 570,247 token corpus.
So far, comprehensive grammar descriptions of Northern Sotho have only been available in the form of prescriptive books aiming at teaching the language. This paper describes parts of the first morpho-syntactic description of Northern Sotho from a computational perspective (Faaß, 2010a). Such a description is necessary for implementing rule based, operational grammars. It is also essential for the annotation of training data to be utilised by statistical parsers. The work that we partially present here may hence provide a resource for computational processing of the language in order to proceed with producing linguistic representations beyond tagging, may it be chunking or parsing. The paper begins with describing significant Northern Sotho verbal morpho-syntactics (section 2). It is shown that the topology of the verb can be depicted as a slot system which may form the basis for computational processing (section 3). Note that the implementation of the described rules (section 4) and also coverage tests are ongoing processes upon that we will report in more detail at a later stage.
This paper presents a survey on the role of negation in sentiment analysis. Negation is a very common linguistic construction that affects polarity and, therefore, needs to be taken into consideration in sentiment analysis.
We will present various computational approaches modeling negation in sentiment analysis. We will, in particular, focus on aspects such as level of representation used for sentiment analysis, negation word detection and scope of negation. We will also discuss limits and challenges of negation modeling on that task.
Bootstrapping Supervised Machine-learning Polarity Classifiers with Rule-based Classification
(2010)
In this paper, we explore the effectiveness of bootstrapping supervised machine-learning polarity classifiers using the output of domain-independent rule-based classifiers. The benefit of this method is that no labeled training data are required. Still, this method allows to capture in-domain knowledge by training the supervised classifier on in-domain features, such as bag of words.
We investigate how important the quality of the rule-based classifier is and what features are useful for the supervised classifier. The former addresses the issue in how far relevant constructions for polarity classification, such as word sense disambiguation, negation modeling, or intensification, are important for this self-training approach. We not only compare how this method relates to conventional semi-supervised learning but also examine how it performs under more difficult settings in which classes are not balanced and mixed reviews are included in the dataset.
In the paper we investigate the impact of data size on a Word Sense Disambiguation task (WSD). We question the assumption that the knowledge acquisition bottleneck, which is known as one of the major challenges for WSD, can be solved by simply obtaining more and more training data. Our case study on 1,000 manually annotated instances of the German verb drohen (threaten) shows that the best performance is not obtained when training on the full data set, but by carefully selecting new training instances with regard to their informativeness for the learning process (Active Learning). We present a thorough evaluation of the impact of different sampling methods on the data sets and propose an improved method for uncertainty sampling which dynamically adapts the selection of new instances to the learning progress of the classifier, resulting in more robust results during the initial stages of learning. A qualitative error analysis identifies problems for automatic WSD and discusses the reasons for the great gap in performance between human annotators and our automatic WSD system.
Active learning has been applied to different NLP tasks, with the aim of limiting the amount of time and cost for human annotation. Most studies on active learning have only simulated the annotation scenario, using prelabelled gold standard data. We present the first active learning experiment for Word Sense Disambiguation with human annotators in a realistic environment, using fine-grained sense distinctions, and investigate whether AL can reduce annotation cost and boost classifier performance when applied to a real-world task.
We describe the SemEval-2010 shared task on “Linking Events and Their Participants in Discourse”. This task is an extension to the classical semantic role labeling task. While semantic role labeling is traditionally viewed as a sentence-internal task, local semantic argument structures clearly interact with each other in a larger context, e.g., by sharing references to specific discourse entities or events. In the shared task we looked at one particular aspect of cross-sentence links between argument structures, namely linking locally uninstantiated roles to their co-referents in the wider discourse context (if such co-referents exist). This task is potentially beneficial for a number of NLP applications, such as information extraction, question answering or text summarization.
We present a method and a software tool, the FrameNet Transformer, for deriving customized versions of the FrameNet database based on frame and frame element relations. The FrameNet Transformer allows users to iteratively coarsen the FrameNet sense inventory in two ways. First, the tool can merge entire frames that are related by user-specified relations. Second, it can merge word senses that belong to frames related by specified relations. Both methods can be interleaved. The Transformer automatically outputs format-compliant FrameNet versions, including modified corpus annotation files that can be used for automatic processing. The customized FrameNet versions can be used to determine which granularity is suitable for particular applications. In our evaluation of the tool, we show that our method increases accuracy of statistical semantic parsers by reducing the number of word-senses (frames) per lemma, and increasing the number of annotated sentences per lexical unit and frame. We further show in an experiment on the FATE corpus that by coarsening FrameNet we do not incur a significant loss of information that is relevant to the Recognizing Textual Entailment task.
Opinion holder extraction is one of the important subtasks in sentiment analysis. The effective detection of an opinion holder depends on the consideration of various cues on various levels of representation, though they are hard to formulate explicitly as features. In this work, we propose to use convolution kernels for that task which identify meaningful fragments of sequences or trees by themselves. We not only investigate how different levels of information can be effectively combined in different kernels but also examine how the scope of these kernels should be chosen. In general relation extraction, the two candidate entities thought to be involved in a relation are commonly chosen to be the boundaries of sequences and trees. The definition of boundaries in opinion holder extraction, however, is less straightforward since there might be several expressions beside the candidate opinion holder to be eligible for being a boundary.
This paper describes work directed towards the development of a syllable prominence-based prosody generation functionality for a German unit selection speech synthesis system. A general concept for syllable prominence-based prosody generation in unit selection synthesis is proposed. As a first step towards its implementation, an automated syllable prominence annotation procedure based on acoustic analyses has been performed on the BOSS speech corpus. The prominence labeling has been evaluated against an existing annotation of lexical stress levels and manual prominence labeling on a subset of the corpus. We discuss methods and results and give an outlook on further implementation steps.
Corpus-based identification and disambiguation of reading indicators for German nominalizations
(2010)
Corpus data is often structurally and lexically ambiguous; corpus extraction methodologies thus must be made aware of ambiguities. Therefore, given an extraction task, all relevant ambiguities must be identified. To resolve these ambiguities, contextual data responsible for one or another reading is to be considered. In the context of our present work, German -ung-nominalizations and their sortal readings are under examination. A number of these nominalizations may be read as an event or a result, depending on the semantic group they belong to. Here, we concentrate on nominalizations of verbs of saying (henceforth: "verba dicendi"), identify their context partners and their influence on the sortal reading of the nominalizations in question. We present a tool which calculates the sortal reading of such nominalizations and thus may improve not only corpus extraction, but also e.g. machine translation. Lastly, we describe successful attempts to identify the correct sortal reading, conclusions and future work.
This paper describes general requirements for evaluating and documenting NLP tools with a focus on morphological analysers and the design of a Gold Standard. It is argued that any evaluation must be measurable and documentation thereof must be made accessible for any user of the tool. The documentation must be of a kind that it enables the user to compare different tools offering the same service, hence the descriptions must contain measurable values. A Gold Standard presents a vital part of any measurable evaluation process, therefore, the corpus-based design of a Gold Standard, its creation and problems that occur are reported upon here. Our project concentrates on SMOR, a morphological analyser for German that is to be offered as a web-service. We not only utilize this analyser for designing the Gold Standard, but also evaluate the tool itself at the same time. Note that the project is ongoing, therefore, we cannot present final results.
Streefkerk defines prominence as the perceptually outstanding parts in spoken language. An optimal rating scale for syllable prominence has not been found yet. This paper evaluates a 4-point, an 11-point, a 31-point, and a continuous scale for the rating of syllable prominence and gives support for scales using a higher number of levels. Priming effects found by Arnold, et al., could only be replicated using the 31-point scale.
Prominence has been widely studied on the word level and the syllable level. An extensive study comparing the two approaches is missing in the literature. This study investigates how word and syllable prominence relate to each other in German. We find that perceptual ratings based on the word level are more extreme than those based on the syllable level. The correlations between word prominence and acoustic features are greater than the correlations between syllable prominence and acoustic features.
Between classical symbolic word sense disambiguation (wsd) using explicit deep semantic representations of sentences and texts and statistical wsd using word co-occurrence information, there is a recent tendency towards mediating methods. Similar to so-called lightweight semantics (Marek, 2009) we suggest to only make sparse use of semantic information. We describe an approximation model based upon flat underspecified discourse representation structures (FUDRSs, cf. Eberle, 2004) that weighs knowledge about context structure, lexical semantic restrictions and interpretation preferences. We give a catalogue of guidelines for human annotation of texts by corresponding indicators. Using this, the reliability of an analysis tool that implements the model can be tested with respect to annotation precision and disambiguation prediction and how both can be improved by bootstrapping the knowledge of the system using corpus information. For the balanced test corpus considered the recognition rate of the preferred reading is 80-90% (depending on the smoothing of parse errors).
An interactive, dynamic electronic dictionary aimed at text production should guide the user in innovative ways, especially in respect of difficult, complicated or confusing issues. This paper proposes a design for bilingual dictionaries intended to guide users in text production; we focus on complex phenomena of the interaction between lexis and grammar. It will be argued that a dictionary aimed at guiding the user in lexical selection should implement a type of “decision algorithm”. In addition, it should flag incorrect solutions and should warn against possible wrong generalisations of (foreign) language learners. Our proposals will be illustrated with examples from several languages, as the design principles are generally applicable. The copulative construction which is regarded as the most complicated grammatical structure in Northern Sotho will be analyzed in more detail and presented as a case in point.
In this paper, we explore different linguistic structures encoded as convolution kernels for the detection of subjective expressions. The advantage of convolution kernels is that complex structures can be directly provided to a classifier without deriving explicit features. The feature design for the detection of subjective expressions is fairly difficult and there currently exists no commonly accepted feature set. We consider various structures, such as constituency parse structures, dependency parse structures, and predicate-argument structures. In order to generalize from lexical information, we additionally augment these structures with clustering information and the task-specific knowledge of subjective words. The convolution kernels will be compared with a standard vector kernel.
In order to automatically extract opinion holders, we propose to harness the contexts of prototypical opinion holders, i.e. common nouns, such as experts or analysts, that describe particular groups of people whose profession or occupation is to form and express opinions towards specific items. We assess their effectiveness in supervised learning where these contexts are regarded as labelled training data and in rule-based classification which uses predicates that frequently co-occur with mentions of the prototypical opinion holders. Finally, we also examine in how far knowledge gained from these contexts can compensate the lack of large amounts of labeled training data in supervised learning by considering various amounts of actually labeled training sets.
Semantic argument structures are often incomplete in that core arguments are not locally instantiated. However, many of these implicit arguments can be linked to referents in the wider context. In this paper we explore a number of linguistically motivated strategies for identifying and resolving such null instantiations (NIs). We show that a more sophisticated model for identifying definite NIs can lead to noticeable performance gains over the state-of-the- art for NI resolution.
Active Learning (AL) has been proposed as a technique to reduce the amount of annotated data needed in the context of supervised classification. While various simulation studies for a number of NLP tasks have shown that AL works well on goldstandard data, there is some doubt whether the approach can be successful when applied to noisy, real-world data sets. This paper presents a thorough evaluation of the impact of annotation noise on AL and shows that systematic noise resulting from biased coder decisions can seriously harm the AL process. We present a method to filter out inconsistent annotations during AL and show that this makes AL far more robust when applied to noisy data.
We introduce a system that learns the participants of arbitrary given scripts. This system processes data from web experiments, in which each participant can be realized with different expressions. It computes participants by encoding semantic similarity and global structural information into an Integer Linear Program. An evaluation against a gold standard shows that we significantly outperform two informed baselines.
This paper aims at contributing to the analysis of overlaps in turns-at-talk from both a sequential and a multimodal perspective. Overlaps have been studied within Conversation Analysis by focusing mainly on verbal and vocal resources; taking into account multimodal resources such as gesture, bodily posture, and gaze contributes to a better understanding of participants’ orientations to the sequential organization of overlapping talk and their management of speakership. First, we introduce the way in which overlaps have been studied in Conversation Analysis, mainly by Jefferson (1973, 1983, 2004) and Schegloff (2000); then we propose possible implications of their multimodal analysis. In order to demonstrate that speakers systematically orient to the overlap onset and resolution we analyze the multimodal conduct of overlapped speakers. Findings show methodical variations in trajectories of overlap resolution: speakers’ gestures in overlap display themselves as maintaining or withdrawing their turn, thereby exhibiting the speakership achieved and negotiated during overlap.
Mechanism-based thinking on policy diffusion. A review of current approaches in political science
(2011)
Despite theoretical and methodological progress in what is now coined as the third generation of diffusion studies, explicitly dealing with the causal mechanisms underlying diffusion processes and comparatively analyzing them is only of recent date. As a matter of fact, diffusion research has ended up in a diverse and often unconnected array of theoretical assumptions relying both on rational as well as constructivist reasoning – a circumstance calling for more theoretical coherence and consistency. Against this backdrop, this paper reviews and streamlines diffusion literature in political science. Diffusion mechanisms largely cluster around two causal arguments determining the desires and preferences of actors for choosing alternative policies. First, existing diffusion mechanisms accounts can be grouped according to the rationality for policy adoption, this means that government behavior is based on the instrumental considerations of actors or on constructivist arguments like norms and rule-driven actors. Second, diffusion mechanisms can either directly impact on the beliefs of actors or they might influence the structural conditions for decision-making. Following this logic, four basic diffusion mechanisms can be identified in mechanism-based thinking on policy diffusion: emulation, socialization, learning, and externalities.
This paper uses a devil’s advocate position to highlight the benefits of metadata creation for linguistic resources. It provides an overview of the required metadata infrastructure and shows that this infrastructure is in the meantime developed by various projects and hence can be deployed by those working with linguistic resources and archiving. Possible caveats of metadata creation are mentioned starting with user requirements and backgrounds, contribution to academic merits of researchers and standardisation. These are answered with existing technologies and procedures, referring to the Component Metadata Infrastructure (CMDI). CMDI provides an infrastructure and methods for adapting metadata to the requirements of specific classes of resources, using central registries for data categories, and metadata schemas. These registries allow for the definition of metadata schemas per resource type while reusing groups of data categories also used by other schemas. In summary, rules of best practice for the creation of metadata are given.
XML has been designed for creating structured documents, but the information that is encoded in these structures are, by definition, out of scope for XML. Additional sources, normally not easily interpretable by computers, such as documentation are needed to determine the intention of specific tags in a tag-set. The Component Metadata Infrastructure (CMDI) takes a rather pragmatic approach to foster interoperability between XML instances in the domain of metadata descriptions for language resources. This paper gives an overview of this approach.
Linguistics is facing the challenge of many other sciences as it continues to grow into increasingly complex subfields, each with its own separate or overarching branches. While linguists are certainly aware of the overall structure of the research field, they cannot follow all developments other than those of their subfields. It is thus important to help specialists but also newcomers alike to bushwhack through evolved or unknown territory of linguistic data. A considerable amount of research data in linguistics is described with metadata. While studies described and published in archived journals and conference proceedings receive a quite homogeneous set of metadata tags — e.g., author, title, publisher —, this does not hold for the empirical data and analyses that underlie such studies. Moreover, lexicons, grammars, experimental data, and other types of resources come in different forms; and to make things worse, their description in terms of metadata is also not uniform, if existing at all. These problems are well-known and there are now a number of international initiatives — e.g., CLARIN, FlareNet, MetaNet, DARIAH — to build infrastructures for managing linguistic resources. The NaLiDa project, funded by the German Research Foundation, aims at facilitating the management and access to linguistic resources originating from German research institutions. In cooperation with the German SFB 833 research center, we are developing a combination of faceted and full-text search to give integrated access through heterogeneous metadata sets. Our approach is supported by a central registry for metadata field descriptors, and a component repository for structured groups of data categories as larger building blocks.
This article looks at Latgalian from a perspective of a classification of languages. It starts by discussing relevant terms relating to sociolinguistic language types. It argues that Latgalian and its speakers show considerable similarities with many languages in Europe which are considered to be regional languages – hence, also Latgalian should be classified as such. In a second part, the article uses sociolinguistic data to indicate that the perceptions of speakers confirm this classification. Therefore, Latgalian should also officially be treated with the respect that other regional languages in Europe enjoy.
The perception of syllable prominence depends to a limited extent on the acoustic properties of the speech signal in question. Psychoacoustic factors are involved as well. Thus, research often relies on two types of data: subjective prominence ratings collected in perception experiments and acoustic measures. A problem with the rating data is noise resulting from individual approaches to the rating task. This paper addresses the question of how this noise can be reduced by normalization, evaluating 12 normalization methods. In a perception experiment, prominence ratings concerning German read speech were collected. From the raw rating data 12 different ‘mirror’ data-sets were computed according to the 12 methods. Each mirror data-set was correlated with the same set of underlying acoustic data. The multiple regression setup included raw syllable duration as well as within-syllable maximum F0 and intensity. Adjusted r2-values could beraised considerably with selected methods.
This article discusses the situation of the Latgalian language in Latvia today. It first provides an overview of languages in Latvia, followed by a historical and contemporary sketch of the societal position of Latgalian and by an account of current Latgalian language activism. On this basis, the article then applies schemes of language functions and of evaluations of the societal position of minority languages to Latgalian. Given the range of functions that Latgalian fulfils today and the wishes and attempts by activists to expand these functions, the article argues that it is surprising that so little attention is given to Latgalian in mainstream Latvian and international sociolinguistic publications. In this light, the fate of the language is difficult to prognose, but a lot depends on whether the Latvian state will clarify its own unclear perception of policies towards Latgalian and on how much attention it will receive in the future.
Over the past decades, problems related to linguistic minorities and their well-being, as well as to minority languages and their maintenance, have developed as an independent branch of minority studies. Studies of language in society and sociolinguistics, strategies of minority language survival and the empowerment of their speakers have produced a considerable output of case studies and theoretical writings.In this multifaceted field of investigation, language use, language practices, language policies and language politics represent interrelated aspects of social and linguistic relations that cannot be meaningfully addressed from a point of view of one scientific discipline only. This is specially the case when one wants to understand processes of language loss and maintenance, or the revitalization and empowerment of a language community. Such processes are linguistic expressions of complex social settings, and reflect group and individual identities that in turn express changing systems of collective values, human networks, fashions and social practices.
Electronic dictionaries should support dictionary users by giving them guidance in text production and text reception, alongside a user-definable offer of lexicographic data for cognitive purposes. In this article, we sketch the principles of an interactive and dynamic electronic dictionary aimed at text production and text reception guiding users in innovative ways, especially with respect to difficult, complicated or confusing issues. The lexicographer has to do a very careful analysis of the nature of the possible problems to suggest an optimal solution for a specific problem. We are of the opinion that there are numerous complex situations where users need more detailed support than currently available in e-dictionaries, enabling them to make valid and correct choices. For highly complex situations, we suggest guidance through a decision tree-like device. We assume that the solutions proposed here are not specific to one language only but can, after careful analysis, be applied to e-dictionaries in different languages across the world.
Knowledge Acquisition with Natural Language Processing in the Food Domain: Potential and Challenges
(2012)
In this paper, we present an outlook on the effectiveness of natural language processing (NLP) in extracting knowledge for the food domain. We identify potential scenarios that we think are particularly suitable for NLP techniques. As a source for extracting knowledge we will highlight the benefits of textual content from social media. Typical methods that we think would be suitable will be discussed. We will also address potential problems and limits that the application of NLP methods may yield.
In this paper, we examine methods to automatically extract domain-specific knowledge from the food domain from unlabeled natural language text. We employ different extraction methods ranging from surface patterns to co-occurrence measures applied on different parts of a document. We show that the effectiveness of a particular method depends very much on the relation type considered and that there is no single method that works equally well for every relation type. We also examine a combination of extraction methods and also consider relationships between different relation types. The extraction methods are applied both on a domain-specific corpus and the domain-independent factual knowledge base Wikipedia. Moreover, we examine an open-domain lexical ontology for suitability.
We present a gold standard for semantic relation extraction in the food domain for German. The relation types that we address are motivated by scenarios for which IT applications present a commercial potential, such as virtual customer advice in which a virtual agent assists a customer in a supermarket in finding those products that satisfy their needs best. Moreover, we focus on those relation types that can be extracted from natural language text corpora, ideally content from the internet, such as web forums, that are easy to retrieve. A typical relation type that meets these requirements are pairs of food items that are usually consumed together. Such a relation type could be used by a virtual agent to suggest additional products available in a shop that would potentially complement the items a customer has already in their shopping cart. Our gold standard comprises structural data, i.e. relation tables, which encode relation instances. These tables are vital in order to evaluate natural language processing systems that extract those relations.
This paper presents Release 2.0 of the SALSA corpus, a German resource for lexical semantics. The new corpus release provides new annotations for German nouns, complementing the existing annotations of German verbs in Release 1.0. The corpus now includes around 24,000 sentences with more than 36,000 annotated instances. It was designed with an eye towards NLP applications such as semantic role labeling but will also be a useful resource for linguistic studies in lexical semantics.
Current work on sentiment analysis is characterized by approaches with a pragmatic focus, which use shallow techniques in the interest of robustness but often rely on ad-hoc creation of data sets and methods. We argue that progress towards deep analysis depends on a) enriching shallow representations with linguistically motivated, rich information, and b) focussing different branches of research and combining ressources to create synergies with related work in NLP. In the paper, we propose SentiFrameNet, an extension to FrameNet, as a novel representation for sentiment analysis that is tailored to these aims.
In this paper, we describe MLSA, a publicly available multi-layered reference corpus for German-language sentiment analysis. The construction of the corpus is based on the manual annotation of 270 German-language sentences considering three different layers of granularity. The sentence-layer annotation, as the most coarse-grained annotation, focuses on aspects of objectivity, subjectivity and the overall polarity of the respective sentences. Layer 2 is concerned with polarity on the word- and phrase-level, annotating both subjective and factual language. The annotations on Layer 3 focus on the expression-level, denoting frames of private states such as objective and direct speech events. These three layers and their respective annotations are intended to be fully independent of each other. At the same time, exploring for and discovering interactions that may exist between different layers should also be possible. The reliability of the respective annotations was assessed using the average pairwise agreement and Fleiss’ multi-rater measures. We believe that MLSA is a beneficial resource for sentiment analysis research, algorithms and applications that focus on the German language.
This paper presents an annotation scheme for English modal verbs together with sense-annotated data from the news domain. We describe our annotation scheme and discuss problematic cases for modality annotation based on the inter-annotator agreement during the annotation. Furthermore, we present experiments on automatic sense tagging, showing that our annotations do provide a valuable training resource for NLP systems.
In two eye-tracking experiments, we investigated the relationship between the subject preference in the resolution of subject-object ambiguities in German embedded clauses and semantic word order constraints (i.e., prominence hierarchies relating to the specificity/referentiality of noun phrases, case assignment and thematic role assignment). Our central research question concerned the timecourse with which prominence information is used and particularly whether it modulates the subject preference. In both experiments, we replicated previous findings of reanalysis effects for object-initial structures. Our findings further suggest that noun phrase prominence does not alter initial parsing strategies (viz., the subject preference), but rather modulates the ease of later reanalysis processes. In Experiment 1, the object case assigned by the verb did not affect the ease of reanalysis. However, the syntactic reanalysis was rendered more difficult when the order of the two arguments violated the specificity/referentiality hierarchy. Experiment 2 revealed that the initial subject preference also holds for verbs favoring an object-initial base order (i.e., dative object-experiencer verbs). However, the advantage for subject-initial sentences is neutralized in relatively late processing stages when the thematic role hierarchy and the specificity hierarchy converge to promote scrambling.
The current state of the art for metadata provision allows for a very flexible approach, catering for the needs of different archives and communities, referring to common data category registries that describe the meaning of a data category at least to authors of metadata. Component models for metadata provisions are for example used by CLARIN and META-SHARE, but there is also an increased flexibility in other metadata schemas such as Dublin Core, which is usually not seen as appropriate for meaningful description of language resources.
Making resources available for others and putting this to a second use in other projects has never been more widely accepted as a sensible efficient way to avoid a waste of efforts and resources. However, when it comes to the details, there is still a vast number of problems. This workshop has aimed at being a forum to address issues and challenges in the concrete work with metadata for LRs, not restricted to a single initiative for archiving LRs. It has allowed for exchange and discussion and we hope that the reader finds the articles here compiled interesting and useful.
This paper presents the system architecture as well as the underlying workflow of the Extensible Repository System of Digital Objects (ERDO) which has been developed for the sustainable archiving of language resources within the Tübingen CLARIN-D project. In contrast to other approaches focusing on archiving experts, the described workflow can be used by researchers without required knowledge in the field of long-term storage for transferring data from their local file systems into a persistent repository.
Creating and maintaining metadata for various kinds of resources requires appropriate tools to assist the user. The paper presents the metadata editor ProFormA for the creation and editing of CMDI (Component Metadata Infrastructure) metadata in web forms. This editor supports a number of CMDI profiles currently being provided for different types of resources. Since the editor is based on XForms and server-side processing, users can create and modify CMDI files in their standard browser without the need for further processing. Large parts of ProFormA are implemented as web services in order to reuse them in other contexts and programs.
The paper’s purpose is to give an overview of the work on the Component Metadata Infrastructure (CMDI) that was implemented in the CLARIN research infrastructure. It explains, the underlying schema, the accompanying tools and services. It also describes the status and impact of the CMDI developments done within the CLARIN project and past and future collaborations with other projects.
This paper describes the status of the standardization efforts of a Component Metadata approach for describing Language Resources with metadata. Different linguistic and Language & Technology communities as CLARIN, META-SHARE and NaLiDa use this component approach and see its standardization of as a matter for cooperation that has the possibility to create a large interoperable domain of joint metadata. Starting with an overview of the component metadata approach together with the related semantic interoperability tools and services as the ISOcat data category registry and the relation registry we explain the standardization plan and efforts for component metadata within ISO TC37/SC4. Finally, we present information about uptake and plans of the use of component metadata within the three mentioned linguistic and L&T communities.
The Component Metadata Infrastructure (CMDI) in a project on sustainable linguistic resources
(2012)
The sustainable archiving of research data for predefined time spans has become increasingly important to researchers and is stipulated by funding organizations with the obligatory task of being observed by researchers. An important aspect in view of such a sustainable archiving of language resources is the creation of metadata, which can be used for describing, finding and citing resources. In the present paper, these aspects are dealt with from the perspectives of two projects: the German project for Sustainability of Linguistic Data at the University of Tubingen (NaLiDa, cf. http://www.sfs.uni-tuebingen.de/nalida) and the Dutch-Flemish HLT Agency hosted at the Institute for Dutch Lexicology (TST-Centrale, cf.http://www.inl.nl/tst-centrale). Both projects unfold their approaches to the creation of components and profiles using the Component Metadata Infrastructure (CMDI) as underlying metadata schema for resource descriptions, highlighting their experiences as well as advantages and disadvantages in using CMDI.
This paper describes the ongoing work to integrate WebLicht into the CLARIN infrastructure. It introduces the CLARIN infrastructure for scholars in the humanities and social sciences as well as WebLicht - an orchestration and execution environment that is built upon Service Oriented Architecture principles. The integration of WebLicht into the CLARIN infrastructure involves adapting it to the standards and practices used within CLARIN, including distributed repositories, CMDI metadata, and persistent identifiers.
The instructions under which raters quantify syllable prominence perception need to be simple in order to maintain immediate reactions. This leads to noise in the rating data that can be dealt with by normalization, e.g. setting central tendency = 0 and dispersion = 1 (as in Z-score normalization). Questions arise such as: Which parameter is adequate here to capture central tendency? Which reference distribution should the normalization be based on? In this paper 16 different normalization methods are evaluated. In a perception experiment using German read speech (prose and poetry), syllable prominence ratings were collected. From the rating data 16 complete “mirror” data-sets were computed according to the 16 methods. Each mirror data-set was correlated with the same set of measures from the underlying acoustic data, focusing on raw syllable duration which is seen as a rather straightforward acoustic aspect of syllable prominence. Correlation coefficients could be raised considerably by selected methods.
Towards a part-of-speech ontology: encoding morphemic units of two South African Bantu languages
(2012)
This article describes the design of an electronic knowledge base, namely a morpho-syntactic database structured as an ontology of linguistic categories, containing linguistic units of two related languages of the South African Bantu group: Northern Sotho and Zulu. These languages differ significantly in their surface orthographies, but are very similar on the lexical and sub-lexical levels. It is therefore our goal to describe the morphemes of these languages in a single common database in order to outline and interpret commonalities and differences in more detail. Moreover, the relational database which is developed defines the underlying morphemic units (morphs) for both languages. It will be shown that the electronic part-of-speech ontology goes hand in hand with part-of-speech tagsets that label morphemic units. This database is designed as part of a forthcoming system providing lexicographic and linguistic knowledge on the official South African Bantu languages.
This special issue of the Journal on Ethnopolitics and Minority Issues in Europe (JEMIE) brings together some of the participants of the symposium Political and Economic Resources and Obstacles of Minority Language Maintenance organized by the Language Survival Network ‘POGA’ at Tallinn University, Estonia, in December 2010. More than 20 scholars representing linguistics, anthropology, social sciences and law participated in the symposium, to present papers and discuss questions related to minority language loss, maintenance and revitalization. The six case studies contained in this special issue look at different minorities and regions in the European Union, Russia and the US. The linguistic communities discussed are the Russian-, Võru/Seto- and Latgalian-speaking minorities of Estonia and Latvia; the Welsh- and Breton-speaking communities of the Celtic language; the Russian Finno-Ugrian people with regional autonomies; and the native American groups of the Delaware/Cherokee and the Oneida. The reader will find articles relating to interdisciplinary research approaches in and on minority languages and minority language communities.
Interested in formally modelling similarity between narratives, we investigate judgements of similarity between narratives in a small corpus of film reviews and book–film comparisons. A main finding is that judgements tend to concern multiple levels of story representation at once. As these texts are pragmatically related to reception contexts, we find many references to reception quality and optimality. We conclude that current formal models of narrative can not capture the task of naturalistic narrative comparisons given in the analysed reviews, but that the development of models containing a more reception-oriented point of view will be necessary.
The understanding of story variation, whether motivated by cultural currents or other factors, is important for applications of formal models of narrative such as story generation or story retrieval. We present the first stage of an experiment to elicit natural narrative variation data suitable for evaluation with respect to story similarity, to qualitative and quantitative analysis of story variation, and also for data processing. We also present few preliminary results from the first stage of the experiment, using Red Riding Hood and Romeo and Juliet as base texts.
We examine predicative adjectives as an unsupervised criterion to extract subjective adjectives. We do not only compare this criterion with a weakly supervised extraction method but also with gradable adjectives, i.e. another highly subjective subset of adjectives that can be extracted in an unsupervised fashion. In order to prove the robustness of this extraction method, we will evaluate the extraction with the help of two different state-of-the-art sentiment lexicons (as a gold standard).
In the rapidly changing circumstances of our increasingly digital world, reading is also becoming an increasingly digital experience: electronic books (e-books) are now outselling print books in the United States and the United Kingdom. Nevertheless, many readers still view e-books as less readable than print books. The present study thus used combined EEG and eyetracking measures in order to test whether reading from digital media requires higher cognitive effort than reading conventional books. Young and elderly adults read short texts on three different reading devices: a paper page, an e-reader and a tablet computer and answered comprehension questions about them while their eye movements and EEG were recorded. The results of a debriefing questionnaire replicated previous findings in that participants overwhelmingly chose the paper page over the two electronic devices as their preferred reading medium. Online measures, by contrast, showed shorter mean fixation durations and lower EEG theta band voltage density – known to covary with memory encoding and retrieval – for the older adults when reading from a tablet computer in comparison to the other two devices. Young adults showed comparable fixation durations and theta activity for all three devices. Comprehension accuracy did not differ across the three media for either group. We argue that these results can be explained in terms of the better text discriminability (higher contrast) produced by the backlit display of the tablet computer. Contrast sensitivity decreases with age and degraded contrast conditions lead to longer reading times, thus supporting the conclusion that older readers may benefit particularly from the enhanced contrast of the tablet. Our findings thus indicate that people’s subjective evaluation of digital reading media must be dissociated from the cognitive and neural effort expended in online information processing while reading from such devices.
We investigate the task of detecting reliable statements about food-health relationships from natural language texts. For that purpose, we created a specially annotated web corpus from forum entries discussing the healthiness of certain food items. We examine a set of task-specific features (mostly) based on linguistic insights that are instrumental in finding utterances that are commonly perceived as reliable. These features are incorporated in a supervised classifier and compared against standard features that are widely used for various tasks in natural language processing, such as bag of words, part-of speech and syntactic parse information.
We investigate how the granularity of POS tags influences POS tagging, and furthermore, how POS tagging performance relates to parsing results. For this, we use the standard “pipeline” approach, in which a parser builds its output on previously tagged input. The experiments are performed on two German treebanks, using three POS tagsets of different granularity, and six different POS taggers, together with the Berkeley parser. Our findings show that less granularity of the POS tagset leads to better tagging results. However, both too coarse-grained and too fine-grained distinctions on POS level decrease parsing performance.
“My Curiosity was Satisfied, but not in a Good Way”: Predicting User Ratings for Online Recipes
(2014)
In this paper, we develop an approach to automatically predict user ratings for recipes at Epicurious.com, based on the recipes’ reviews. We investigate two distributional methods for feature selection, Information Gain and Bi-Normal Separation; we also compare distributionally selected features to linguistically motivated features and two types of frameworks: a one-layer system where we aggregate all reviews and predict the rating vs. a two-layer system where ratings of individual reviews are predicted and then aggregated. We obtain our best results by using the two-layer architecture, in combination with 5 000 features selected by Information Gain. This setup reaches an overall accuracy of 65.60%, given an upper bound of 82.57%.
The annotation of parts of speech (POS) in linguistically annotated corpora is a fundamental annotation layer which provides the basis for further syntactic analyses, and many NLP tools rely on POS information as input. However, most POS annotation schemes have been developed with written (newspaper) text in mind and thus do not carry over well to text from other domains and genres. Recent discussions have concentrated on the shortcomings of present POS annotation schemes with regard to their applicability to data from domains other than newspaper text.
This paper describes a first version of an integrated e-dictionary translating possessive constructions from English to Zulu. Zulu possessive constructions are difficult to learn for non-mother tongue speakers. When translating from English into Zulu, a speaker needs to be acquainted with the nominal classification of nouns indicating possession and possessor. Furthermore, (s)he needs to be informed about the morpho-syntactic rules associated with certain combinations of noun classes. Lastly, knowledge of morpho-phonetic changes is also required, because these influence the orthography of the output word forms. Our approach is a novel one in that we combine e-lexicography and natural language processing by developing a (web) interface supporting learners, as well as other users of the dictionary to produce Zulu possessive constructions. The final dictionary that we intend to develop will contain several thousand nouns which users can combine as they wish. It will also translate single words and frequently used multiword expressions, and allow users to test their own translations. On request, information about the morpho-syntactic and morpho-phonetic rules applied by the system are displayed together with the translation. Our approach follows the function theory: the dictionary supports users in text production, at the same time fulfilling a cognitive function.
So far, there have been few descriptions on creating structures capable of storing lexicographic data, ISO 24613:2008 being one of the latest. Another one is by Spohr (2012), who designs a multifunctional lexical resource which is able to store data of different types of dictionaries in a user-oriented way. Technically, his design is based on the principle of a hierarchical XML/OWL (eXtensible Markup Language/Web Ontology Language) representation model. This article follows another route in describing a model based on entities and relations between them; MySQL (usually referred to as: Structured Query Language) describes a database system of tables containing data and definitions of relations between them. The model was developed in the context of the project "Scientific eLexicography for Africa" and the lexicographic database to be built thereof will be implemented with MySQL. The principles of the ISO model and of Spohr's model are adhered to with one major difference in the implementation strategy: we do not place the lemma in the centre of attention, but the sense description — all other elements, including the lemma, depend on the sense description. This article also describes the contained lexicographic data sets and how they have been collected from different sources. As our aim is to compile several prototypical internet dictionaries (a monolingual Northern Sotho dictionary, a bilingual learners' Xhosa–English dictionary and a bilingual Zulu–English dictionary), we describe the necessary microstructural elements for each of them and which principles we adhere to when designing different ways of accessing them. We plan to make the model and the (empty) database with all graphical user interfaces that have been developed, freely available by mid-2015.
Self-Regulated Learning (SRL) is a term that can be used to describe an individual’s ability to develop a skill set allowing him or her to learn in a number of different ways. SRL can also relate to new pedagogical theories that encourage teachers in formal education to motivate and support their students into achieving a high level of self-regulation. This paper reports on the findings of a number of surveys conducted with a wide variety of teachers in different countries, regarding their perceptions of SRL. The results and analysis of these surveys help inform not only the perceptions of SRL amongst teachers but also examine the challenges and opportunities that arise from taking this approach.
We examine the task of separating types from brands in the food domain. Framing the problem as a ranking task, we convert simple textual features extracted from a domain-specific corpus into a ranker without the need of labeled training data. Such method should rank brands (e.g. sprite) higher than types (e.g. lemonade). Apart from that, we also exploit knowledge induced by semi-supervised graph-based clustering for two different purposes. On the one hand, we produce an auxiliary categorization of food items according to the Food Guide Pyramid, and assume that a food item is a type when it belongs to a category unlikely to contain brands. On the other hand, we directly model the task of brand detection using seeds provided by the output of the textual ranking features. We also harness Wikipedia articles as an additional knowledge source.
We report on the two systems we built for Task 1 of the German Sentiment Analysis Shared Task, the task on Source, Subjective Expression and Target Extraction from Political Speeches (STEPS). The first system is a rule-based system relying on a predicate lexicon specifying extraction rules for verbs, nouns and adjectives, while the second is a translation-based system that has been obtained with the help of the (English) MPQA corpus.
Accurate opinion mining requires the exact identification of the source and target of an opinion. To evaluate diverse tools, the research community relies on the existence of a gold standard corpus covering this need. Since such a corpus is currently not available for German, the Interest Group on German Sentiment Analysis decided to create such a resource and make it available to the research community in the context of a shared task. In this paper, we describe the selection of textual sources, development of annotation guidelines, and first evaluation results in the creation of a gold standard corpus for the German language.
Automatic Food Categorization from Large Unlabeled Corpora and Its Impact on Relation Extraction
(2014)
We present a weakly-supervised induction method to assign semantic information to food items. We consider two tasks of categorizations being food-type classification and the distinction of whether a food item is composite or not. The categorizations are induced by a graph-based algorithm applied on a large unlabeled domain-specific corpus. We show that the usage of a domain-specific corpus is vital. We do not only outperform a manually designed open-domain ontology but also prove the usefulness of these categorizations in relation extraction, outperforming state-of-the-art features that include syntactic information and Brown clustering.
Measuring the quality of metadata is only possible by assessing the quality of the underlying schema and the metadata instance. We propose some factors that are measurable automatically for metadata according to the CMD framework, taking into account the variability of schemas that can be defined in this framework. The factors include among others the number of elements, the (re-)use of reusable components, the number of filled in elements. The resulting score can serve as an indicator of the overall quality of the CMD instance, used for feedback to metadata providers or to provide an overview of the overall quality of metadata within a repository. The score is independent of specific schemas and generalizable. An overall assessment of harvested metadata is provided in form of statistical summaries and the distribution, based on a corpus of harvested metadata. The score is implemented in XQuery and can be used in tools, editors and repositories.