Refine
Year of publication
Document Type
- Conference Proceeding (62)
- Part of a Book (19)
- Article (13)
- Working Paper (2)
- Book (1)
- Doctoral Thesis (1)
Keywords
- Automatische Sprachanalyse (29)
- Deutsch (20)
- Korpus <Linguistik> (17)
- Computerlinguistik (16)
- Frame-Semantik (15)
- Annotation (14)
- Semantische Analyse (14)
- Propositionale Einstellung (10)
- Beleidigung (9)
- sentiment analysis (9)
Publicationstate
- Veröffentlichungsversion (75)
- Zweitveröffentlichung (8)
- Postprint (5)
Reviewstate
Publisher
- Association for Computational Linguistics (13)
- European Language Resources Association (10)
- The Association for Computational Linguistics (9)
- German Society for Computational Linguistics & Language Technology und Friedrich-Alexander-Universität Erlangen-Nürnberg (4)
- Springer (4)
- European Language Resources Association (ELRA) (3)
- European language resources association (ELRA) (3)
- Oxford University Press (3)
- Universität Hildesheim (3)
- EACL (2)
Authors like Fillmore 1986 and Goldberg 2006 have made a strong case for regarding argument omission in English as a lexical and construction-based affordance rather than one based on general semantico-pragmatic constraints. They do not, however, address the question of how grammatical restrictions on null complementation might interact with broader narrative conventions, in particular those of genre. In this paper, we attempt to remedy this oversight by presenting a comprehensive overview of genre-based argument omissions and offering a construction-based analysis of genre-based omission conventions. We consider five genre-based omission types: instructional imperatives (Culy 1996, Bender 1999), labelese, diary style (Haegeman 1990), match reports (Ruppenhofer 2004) and quotative clauses. We show that these omission types share important traits; all, for example, have anaphoric rather than indefinite construals. We also show, however, that the omission types differ from each other in idiosyncratic ways. We then address several interrelated representational problems posed by the grammatical treatment of genre-based omissions. For example, the constructions that represent genre-based omission conventions must interact with the lexical entries of verbs, many of which do not generally permit omitted arguments. Accordingly, we offer constructional analyses of genre-based omissions that allow constructions to override lexical valence constraints.
We present a descriptive analysis on the two datasets from the shared task on Source, Subjective Expression and Target Extraction from Political Speeches (STEPS), the only existing German dataset for opinion role extraction of its size. Our analysis discusses the individual properties of the three components, subjective expressions, sources and targets and their relations towards each other. Our observations should help practitioners and researchers when building a system to extract opinion roles from German data.
We present a testsuite for POS tagging German web data. Our testsuite provides the original raw text as well as the gold tokenisations and is annotated for parts-of-speech. The testsuite includes a new dataset for German tweets, with a current size of 3,940 tokens. To increase the size of the data, we harmonised the annotations in already existing web corpora, based on the Stuttgart-Tübingen Tag Set. The current version of the corpus has an overall size of 48,344 tokens of web data, around half of it from Twitter. We also present experiments, showing how different experimental setups (training set size, additional out-of-domain training data, self-training) influence the accuracy of the taggers. All resources and models will be made publicly available to the research community.
We present a new resource for German causal language, with annotations in context for verbs, nouns and adpositions. Our dataset includes 4,390 annotated instances for more than 150 different triggers. The annotation scheme distinguishes three different types of causal events (CONSEQUENCE, MOTIVATION, PURPOSE). We also provide annotations for semantic roles, i.e. of the cause and effect for the causal event as well as the actor and affected party, if present. In the paper, we present inter-annotator agreement scores for our dataset and discuss problems for annotating causal language. Finally, we present experiments where we frame causal annotation as a sequence labelling problem and report baseline results for the prediciton of causal arguments and for predicting different types of causation.
A Supervised learning approach for the extraction of opinion sources and targets from German text
(2019)
We present the first systematic supervised learning approach for the extraction of opinion sources and targets on German language data. A wide choice of different features is presented, particularly syntactic features and generalization features. We point out specific differences between opinion sources and targets. Moreover, we explain why implicit sources can be extracted even with fairly generic features. In order to ensure comparability our classifier is trained and tested on the dataset of the STEPS shared task.
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.
Auf dem Weg zu einer Kartographie: automatische und manuelle Analysen am Beispiel des Korpus ISW
(2021)
Alleviating pain is good and abandoning hope is bad. We instinctively understand how words like alleviate and abandon affect the polarity of a phrase, inverting or weakening it. When these words are content words, such as verbs, nouns, and adjectives, we refer to them as polarity shifters. Shifters are a frequent occurrence in human language and an important part of successfully modeling negation in sentiment analysis; yet research on negation modeling has focused almost exclusively on a small handful of closed-class negation words, such as not, no, and without. A major reason for this is that shifters are far more lexically diverse than negation words, but no resources exist to help identify them. We seek to remedy this lack of shifter resources by introducing a large lexicon of polarity shifters that covers English verbs, nouns, and adjectives. Creating the lexicon entirely by hand would be prohibitively expensive. Instead, we develop a bootstrapping approach that combines automatic classification with human verification to ensure the high quality of our lexicon while reducing annotation costs by over 70%. Our approach leverages a number of linguistic insights; while some features are based on textual patterns, others use semantic resources or syntactic relatedness. The created lexicon is evaluated both on a polarity shifter gold standard and on a polarity classification task.
In this paper we use methods for creating a large lexicon of verbal polarity shifters and apply them to German. Polarity shifters are content words that can move the polarity of a phrase towards its opposite, such as the verb “abandon” in “abandon all hope”. This is similar to how negation words like “not” can influence polarity. Both shifters and negation are required for high precision sentiment analysis. Lists of negation words are available for many languages, but the only language for which a sizable lexicon of verbal polarity shifters exists is English. This lexicon was created by bootstrapping a sample of annotated verbs with a supervised classifier that uses a set of data- and resource-driven features. We reproduce and adapt this approach to create a German lexicon of verbal polarity shifters. Thereby, we confirm that the approach works for multiple languages. We further improve classification by leveraging cross-lingual information from the English shifter lexicon. Using this improved approach, we bootstrap a large number of German verbal polarity shifters, reducing the annotation effort drastically. The resulting German lexicon of verbal polarity shifters is made publicly available.
Alternations play a central role in most current theories of verbal argument structure, wich are devides primarily to model the syntactic flexibility of verbs. Accordingly, these frameworks take verbs, and their projection properties, to be the sole contributors of thematic content to the clause. Approached from this perspective, the German applicative (or be-prefix) construction has puzzling properties. First, while many applicative verbs have transparent base forms, many, including those coined from nouns, do not. Second, applicative verbs are bound by interpretive and argument-realization conditions which cannot be traced to their base forms, if any. These facts suggest that applicative formation is not appropriately modeled as a lexical rule.
Using corpus data from a diverse array of genres, Michaelis and Ruppenhofer propose a unified solution to these two puzzles within the framework of Construction Grammar. Central to this account is the concept of valence augmentation: argument-structure constructions denote event types, and therefore license valence sets which may properly include those of their lexical fillers. As per Panini's Law, resolution of valence mismatch favors the construction over the verb. Like verbs of transfer and location, the applicative construction has a prototype-based event-structure representation: diverse implications of applicative predications--including iteration, transfer, affectedness, intensity and saturation--are shown to derive via regular patterns of semantic extension from the topological concept of coverage.
Semantic role labeling is traditionally viewed as a sentence-level task concerned with identifying semantic arguments that are overtly realized in a fairly local context (i.e., a clause or sentence). However, this local view potentially misses important information that can only be recovered if local argument structures are linked across sentence boundaries. One important link concerns semantic arguments that remain locally unrealized (null instantiations) but can be inferred from the context. In this paper, we report on the SemEval 2010 Task-10 on ‘‘Linking Events and Their Participants in Discourse’’, that addressed this problem. We discuss the corpus that was created for this task, which contains annotations on multiple levels: predicate argument structure (FrameNet and PropBank), null instantiations, and coreference. We also provide an analysis of the task and its difficulties.
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.
German is a language with complex morphological processes. Its long and often ambiguous word forms present a bottleneck problem in natural language processing. As a step towards morphological analyses of high quality, this paper introduces a morphological treebank for German. It is derived from the linguistic database CELEX which is a standard resource for German morphology. We build on its refurbished, modernized and partially revised version. The derivation of the morphological trees is not trivial, especially for such cases of conversions which are morpho-semantically opaque and merely of diachronic interest. We develop solutions and present exemplary analyses. The resulting database comprises about 40,000 morphological trees of a German base vocabulary whose format and grade of detail can be chosen according to the requirements of the applications. The Perl scripts for the generation of the treebank are publicly available on github. In our discussion, we show some future directions for morphological treebanks. In particular, we aim at the combination with other reliable lexical resources such as GermaNet.
Catching the common cause: extraction and annotation of causal relations and their participants
(2017)
In this paper, we present a simple, yet effective method for the automatic identification and extraction of causal relations from text, based on a large English-German parallel corpus. The goal of this effort is to create a lexical resource for German causal relations. The resource will consist of a lexicon that describes constructions that trigger causality as well as the participants of the causal event, and will be augmented by a corpus with annotated instances for each entry, that can be used as training data to develop a system for automatic classification of causal relations. Focusing on verbs, our method harvested a set of 100 different lexical triggers of causality, including support verb constructions. At the moment, our corpus includes over 1,000 annotated instances. The lexicon and the annotated data will be made available to the research community.
The FrameNet lexical database yields information about collocations and multiword expressions in various ways. In some cases phrasal units have been entered from the start as lexical entries (write down). In other cases headword + preposition pairs can be recognized as special collocations Where the preposition in question is a necessary and lexically specified marker of an argument of the headword + fond of, hostile to). Nominal compounds are annotated with respect to noun or (pertinative) adjective modifiers, some of which are analyzable but also entrenched (wheel chair, fiscal year). Nouns that name aggregates, portions, types, etc., sometimes hold lexically specified relations to their dependents (flock of geese). And event nouns frequently Select the support verbs which permit them to enter into predications (file an objection, enter a plea). A subproject aims at extracting, as structured clusters of lexical items, the minimal semantically central kernel dependency graphs from the set of annotations. Such research will yield not only commonplace groupings (eat: dog, bone) but will also yield hitherto unnoticed collocations within such graphs (answer: you, door) where certain dependency links within them are idiomatic or otherwise lexically special, here answer > door. Collocational information can also be retrieved by various types of queries within our MySQL search tool
We compare several different corpus- based and lexicon-based methods for the scalar ordering of adjectives. Among them, we examine for the first time a low- resource approach based on distinctive- collexeme analysis that just requires a small predefined set of adverbial modifiers. While previous work on adjective intensity mostly assumes one single scale for all adjectives, we group adjectives into different scales which is more faithful to human perception. We also apply the methods to both polar and non-polar adjectives, showing that not all methods are equally suitable for both types of adjectives.
In this paper, we report on an effort to develop a gold standard for the intensity ordering of subjective adjectives. Rather than pursue a complete order as produced by paying attention to the mean scores of human ratings only, we take into account to what extent assessors consistently rate pairs of adjectives relative to each other. We show that different available automatic methods for producing polar intensity scores produce results that correlate well with our gold standard, and discuss some conceptual questions surrounding the notion of polar intensity.
We introduce a method for error detection in automatically annotated text, aimed at supporting the creation of high-quality language resources at affordable cost. Our method combines an unsupervised generative model with human supervision from active learning. We test our approach on in-domain and out-of-domain data in two languages, in AL simulations and in a real world setting. For all settings, the results show that our method is able to detect annotation errors with high precision and high recall.
We examine the new task of detecting derogatory compounds (e.g. curry muncher). Derogatory compounds are much more difficult to detect than derogatory unigrams (e.g. idiot) since they are more sparsely represented in lexical resources previously found effective for this task (e.g. Wiktionary). We propose an unsupervised classification approach that incorporates linguistic properties of compounds. It mostly depends on a simple distributional representation. We compare our approach against previously established methods proposed for extracting derogatory unigrams.
Automatic division of spoken language transcripts into sentence-like units is a challenging problem, caused by disfluencies, ungrammatical structures and the lack of punctuation. We present experiments on dividing up German spoken dialogues where we investigate the impact of task setup and data representation, encoding of context information as well as different model architectures for this task.
We discuss the impact of data bias on abusive language detection. We show that classification scores on popular datasets reported in previous work are much lower under realistic settings in which this bias is reduced. Such biases are most notably observed on datasets that are created by focused sampling instead of random sampling. Datasets with a higher proportion of implicit abuse are more affected than datasets with a lower proportion.
Recent work suggests that concreteness and imageability play an important role in the meanings of figurative expressions. We investigate this idea in several ways. First, we try to define more precisely the context within which a figurative expression may occur, by parsing a corpus annotated for metaphor. Next, we add both concreteness and imageability as “features” to the parsed metaphor corpus, by marking up words in this corpus using a psycholinguistic database of scores for concreteness and imageability. Finally, we carry out detailed statistical analyses of the augmented version of the original metaphor corpus, cross-matching the features of concreteness and imageability with others in the corpus such as parts of speech and dependency relations, in order to investigate in detail the use of such features in predicting whether a given expression is metaphorical or not.
Negation is an important contextual phenomenon that needs to be addressed in sentiment analysis. Next to common negation function words, such as not or none, there is also a considerably large class of negation content words, also referred to as shifters, such as the verbs diminish, reduce or reverse. However, many of these shifters are ambiguous. For instance, spoil as in spoil your chance reverses the polarity of the positive polar expression chance while in spoil your loved ones, no negation takes place. We present a supervised learning approach to disambiguating verbal shifters. Our approach takes into consideration various features, particularly generalization features.
This work proposes opinion frames as a representation of discourse-level associations which arise from related opinion topics. We illustrate how opinion frames help gather more information and also assist disambiguation. Finally we present the results of our experiments to detect these associations.
This work proposes opinion frames as a representation of discourse-level associations that arise from related opinion targets and which are common in task-oriented meeting dialogs. We define the opinion frames and explain their interpretation. Additionally we present an annotation scheme that realizes the opinion frames and via human annotation studies, we show that these can be reliably identified.
We study German affixoids, a type of morpheme in between affixes and free stems. Several properties have been associated with them – increased productivity; a bleached semantics, which is often evaluative and/or intensifying and thus of relevance to sentiment analysis; and the existence of a free morpheme counterpart – but not been validated empirically. In experiments on a new data set that we make available, we put these key assumptions from the morphological literature to the test and show that despite the fact that affixoids generate many low-frequency formations, we can classify these as affixoid or non-affixoid instances with a best F1-score of 74%.
Entity framing is the selection of aspects of an entity to promote a particular viewpoint towards that entity. We investigate entity framing of political figures through the use of names and titles in German online discourse, enhancing current research in entity framing through titling and naming that concentrates on English only. We collect tweets that mention prominent German politicians and annotate them for stance. We find that the formality of naming in these tweets correlates positively with their stance. This confirms sociolinguistic observations that naming and titling can have a status-indicating function and suggests that this function is dominant in German tweets mentioning political figures. We also find that this status-indicating function is much weaker in tweets from users that are politically left-leaning than in tweets by right leaning users. This is in line with observations from moral psychology that left-leaning and right-leaning users assign different importance to maintaining social hierarchies.
Editorial
(2020)
Sentiment analysis has so far focused on the detection of explicit opinions. However, of late implicit opinions have received broader attention, the key idea being that the evaluation of an event type by a speaker depends on how the participants in the event are valued and how the event itself affects the participants. We present an annotation scheme for adding relevant information, couched in terms of so-called effect functors, to German lexical items. Our scheme synthesizes and extends previous proposals. We report on an inter-annotator agreement study. We also present results of a crowdsourcing experiment to test the utility of some known and some new functors for opinion inference where, unlike in previous work, subjects are asked to reason from event evaluation to participant evaluation.
The sentiment polarity of an expression (whether it is perceived as positive, negative or neutral) can be influenced by a number of phenomena, foremost among them negation. Apart from closed-class negation words like no, not or without, negation can also be caused by so-called polarity shifters. These are content words, such as verbs, nouns or adjectives, that shift polarities in their opposite direction, e. g. abandoned in “abandoned hope” or alleviate in “alleviate pain”. Many polarity shifters can affect both positive and negative polar expressions, shifting them towards the opposing polarity. However, other shifters are restricted to a single shifting direction. Recoup shifts negative to positive in “recoup your losses”, but does not affect the positive polarity of fortune in “recoup a fortune”. Existing polarity shifter lexica only specify whether a word can, in general, cause shifting, but they do not specify when this is limited to one shifting direction. To address this issue we introduce a supervised classifier that determines the shifting direction of shifters. This classifier uses both resource-driven features, such as WordNet relations, and data-driven features like in-context polarity conflicts. Using this classifier we enhance the largest available polarity shifter lexicon.
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.
Unknown words are a challenge for any NLP task, including sentiment analysis. Here, we evaluate the extent to which sentiment polarity of complex words can be predicted based on their morphological make-up. We do this on German as it has very productive processes of derivation and compounding and many German hapax words, which are likely to bear sentiment, are morphologically complex. We present results of supervised classification experiments on new datasets with morphological parses and polarity annotations.
We propose to use abusive emojis, such as the “middle finger” or “face vomiting”, as a proxy for learning a lexicon of abusive words. Since it represents extralinguistic information, a single emoji can co-occur with different forms of explicitly abusive utterances. We show that our approach generates a lexicon that offers the same performance in cross-domain classification of abusive microposts as the most advanced lexicon induction method. Such an approach, in contrast, is dependent on manually annotated seed words and expensive lexical resources for bootstrapping (e.g. WordNet). We demonstrate that the same emojis can also be effectively used in languages other than English. Finally, we also show that emojis can be exploited for classifying mentions of ambiguous words, such as “fuck” and “bitch”, into generally abusive and just profane usages.
In this contribution, we report on an effort to annotate German data with information relevant to opinion inference. Such information has previously been referred to as effect or couched in terms of eventevaluation functors. We extend the theory and present an extensive scheme that combines both approaches and thus extends the set of inference-relevant predicates. Using these guidelines to annotate 726 German synsets, we achieve good inter-annotator agreement.
As many popular text genres such as blogs or news contain opinions by multiple sources and about multiple targets, finding the sources and targets of subjective expressions becomes an important sub-task for automatic opinion analysis systems. We argue that while automatic semantic role labeling systems (ASRL) have an important contribution to make, they cannot solve the problem for all cases. Based on the experience of manually annotating opinions, sources, and targets in various genres, we present linguistic phenomena that require knowledge beyond that of ASRL systems. In particular, we address issues relating to the attribution of opinions to sources; sources and targets that are realized as zero-forms; and inferred opinions. We also discuss in some depth that for arguing attitudes we need to be able to recover propositions and not only argued-about entities. A recurrent theme of the discussion is that close attention to specific discourse contexts is needed to identify sources and targets correctly.
We present a fine-grained NER annotations scheme with 30 labels and apply it to German data. Building on the OntoNotes 5.0 NER inventory, our scheme is adapted for a corpus of transcripts of biographic interviews by adding categories for AGE and LAN(guage) and also adding label classes for various numeric and temporal expressions. Applying the scheme to the spoken data as well as a collection of teaser tweets from newspaper sites, we can confirm its generality for both domains, also achieving good inter-annotator agreement. We also show empirically how our inventory relates to the well-established 4-category NER inventory by re-annotating a subset of the GermEval 2014 NER coarse-grained dataset with our fine label inventory. Finally, we use a BERT-based system to establish some baselines for NER tagging on our two new datasets. Global results in in-domain testing are quite high on the two datasets, near what was achieved for the coarse inventory on the CoNLLL2003 data. Cross-domain testing produces much lower results due to the severe domain differences.
FrameNet
(2018)
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.
A polarity-sensitive item (PSI), as traditionally defined, is an expression that is restricted to either an affirmative or negative context. PSIs like ‘lift a finger’ and ‘all the time in the world’ sub-serve discourse routines like understatement and emphasis. Lexical–semantic classes are increasingly invoked in descriptions of the properties of PSIs. Here, we use English corpus data and the tools of Frame Semantics (Fillmore, 1982, 1985) to explore Israel’s (2011) observation that the semantic role of a PSI determines how the expression fits into a contextually constructed scalar model. We focus on a class of exceptions implied by Israel’s model: cases in which a given PSI displays two countervailing patterns of polarity sensitivity, with attendant differences in scalar entailments. We offer a set of case studies of polaritysensitive expressions – including verbs of attraction and aversion like ‘can live without’, monetary units like ‘a red cent’, comparative adjectives and time-span adverbials – that demonstrate that the interpretation of a given PSI in a given polar context is based on multiple factors. These factors include the speaker’s perspective on and affective stance towards the described event, available inferences about causality and, perhaps most critically, particulars of the predication, including the verb or adjective’s frame membership, the presence or absence of an ability modal like can, the grammatical construction used and the range of contingencies evoked by the utterance.
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.
We present a quantitative approach to disambiguating flat morphological analyses and producing more deeply structured analyses. Based on existing morphological segmentations, possible combinations of resulting word trees for the next level are filtered first by criteria of linguistic plausibility and then by weighting procedures based on the geometric mean. The frequencies for weighting are derived from three different sources (counts of morphs in a lexicon, counts of largest constituents in a lexicon, counts of token frequencies in a corpus) and can be used either to find the best analysis on the level of morphs or on the next higher constituent level. The evaluation shows that for this task corpus-based frequency counts are slightly superior to counts of lexical data.