Refine
Year of publication
Document Type
- Conference Proceeding (27)
- Article (4)
- Part of a Book (3)
- Doctoral Thesis (1)
- Other (1)
- Report (1)
Has Fulltext
- yes (37)
Keywords
- Maschinelles Lernen (37) (remove)
Publicationstate
- Veröffentlichungsversion (18)
- Zweitveröffentlichung (15)
- Postprint (5)
Reviewstate
Publisher
- Association for Computational Linguistics (6)
- European Language Resources Association (6)
- Springer (5)
- Asian Federation of Natural Language Processing (2)
- Incoma Ltd. (2)
- Northern European Association for Language Technology (2)
- ACTA Press (1)
- Association for Computing Machinery (1)
- Cambridge University Press (1)
- Dublin City University (1)
This contribution summarizes the lessons learned from the organization of a joint conference on text analytics research by the Business, Economic, and Related Data (BERD@NFDI) and Text+ consortia within the National Research Data Infrastructure (NFDI) in Germany. The collaboration aimed to identify common ground and foster interdisciplinary dialogue between scholars in the humanities and in the business domain. The lessons learned include the importance of presenting research questions using textual data to establish common ground, similarities in methodology for processing textual data between the consortia, similarities in research data management, and the need for regular interconsortial discussions on textual analysis methods and data. The collaboration proved valuable for interdisciplinary dialogue within the NFDI, and further collaboration between the consortia is planned.
Computational language models (LMs), most notably exemplified by the widespread success of OpenAI's ChatGPT chatbot, show impressive performance on a wide range of linguistic tasks, thus providing cognitive science and linguistics with a computational working model to empirically study different aspects of human language. Here, we use LMs to test the hypothesis that languages with more speakers tend to be easier to learn. In two experiments, we train several LMs—ranging from very simple n-gram models to state-of-the-art deep neural networks—on written cross-linguistic corpus data covering 1293 different languages and statistically estimate learning difficulty. Using a variety of quantitative methods and machine learning techniques to account for phylogenetic relatedness and geographical proximity of languages, we show that there is robust evidence for a relationship between learning difficulty and speaker population size. However, contrary to expectations derived from previous research, our results suggest that languages with more speakers tend to be harder to learn.
Linguistische Studien arbeiten häufig mit einer Differenzierung zwischen gesprochener und geschriebener Sprache bzw. zwischen Kommunikation der Nähe und Distanz. Die Annahme eines Kontinuums zwischen diesen Polen bietet sich für eine Verortung unterschiedlichster Äußerungsformen an, inklusive unkonventioneller Textsorten wie etwa Popsongs. Wir konzipieren, implementieren und evaluieren ein automatisiertes Verfahren, das mithilfe unkorrelierter Entscheidungsbäume entsprechende Vorhersagen auf Textebene durchführt. Für die Identifizierung der Pole definieren wir einen Merkmalskatalog aus Sprachphänomenen, die als Markierer für Nähe/Mündlichkeit bzw. Distanz/Schriftlichkeit diskutiert werden, und wenden diesen auf prototypische Nähe-/Mündlichkeitstexte sowie prototypische Distanz-/Schrifttexte an. Basierend auf der sehr guten Klassifikationsgüte verorten wir anschließend eine Reihe weiterer Textsorten mithilfe der trainierten Klassifikatoren. Dabei erscheinen Popsongs als „mittige Textsorte“, die linguistisch motivierte Merkmale unterschiedlicher Kontinuumsstufen vereint. Weiterhin weisen wir nach, dass unsere Modelle mündlich kommunizierte, aber vorab oder nachträglich verschriftlichte Äußerungen wie Reden oder Interviews vollkommen anders verorten als prototypische Gesprächsdaten und decken Klassifikationsunterschiede für Social-Media-Varianten auf. Ziel ist dabei nicht eine systematisch-verbindliche Einordung im Kontinuum, sondern eine empirische Annäherung an die Frage, welche maschinell vergleichsweise einfach bestimmbaren Merkmale („shallow features“) nachweisbar Einfluss auf die Verortung haben.
Der vorliegende Beitrag erörtert am Beispiel des aktuell im Aufbau befindlichen Korpus GiesKaNe (= Gie[ßen]Ka[ssel]Ne[uhochdeutsch]) grundlegende Fragen nach dem Verhältnis von Standard und Innovation bei der Erweiterung der Korpuslandschaft durch neue Korpora. Bei jedem neu zu erstellenden Korpus stellt sich die Frage, inwieweit man den bereits etablierten Standards folgt, oder ob es legitim oder vielleicht sogar notwendig ist, neue Modelle der Annotation linguistischer Kategorien zu entwickeln. In diesem Sinne bespricht der Beitrag die Grenzen einer reinen Modellübernahme mit Bezug auf das POS-Tagging in anderen historischen Referenzkorpora und mit Bezug auf TIGER als Baumbank für das Gegenwartsdeutsche. Um trotz der Arbeit mit einer innovativen Alternative dem Prinzip der Interoperabilität gerecht zu werden, wird im Beitrag die Arbeit mit maschinellem Lernen ins Spiel gebracht. Dieses ermöglicht es, aus den vorhandenen Textoberflächenmerkmalen und den vorliegenden Annotationen auch alternative Annotationsmodelle abzuleiten und mittels einer Mehrebenenannotation anzubieten, sodass ein Korpus den Anforderungen an interoperable Nutzbarkeit und wissenschaftlichen Erkenntnisfortschritt gleichermaßen gerecht werden kann.
Automatic summarization systems usually are trained and evaluated in a particular domain with fixed data sets. When such a system is to be applied to slightly different input, labor- and cost-intensive annotations have to be created to retrain the system. We deal with this problem by providing users with a GUI which allows them to correct automatically produced imperfect summaries. The corrected summary in turn is added to the pool of training data. The performance of the system is expected to improve as it adapts to the new domain.
Content
1 Predicting learner knowledge of individual words using machine learning
Drilon Avdiu, Vanessa Bui, Klára Ptacinová Klimci´ková
2 Automatic Generation and Semantic Grading of Esperanto Sentences in a Teaching Context
Eckhard Bick
3 Toward automatic improvement of language produced by non-native language learners
Mathias Creutz, Eetu Sjöblom
4 Linguistic features and proficiency classification in L2 Spanish and L2 Portuguese
Iria del Ri´o
5 Integrating large-scale web data and curated corpus data in a search engine supporting German literacy education
Sabrina Dittrich, Zarah Weiss, Hannes Schröter, Detmar Meurers
6 Formalism for a language agnostic language learning game and productive grid generation
Sylvain Hatier, Arnaud Bey, Mathieu Loiseau
7 Understanding Vocabulary Growth Through An Adaptive Language Learning System
Elma Kerz, Andreas Burgdorf, Daniel Wiechmann, Stefan Meeger,Yu Qiao, Christian Kohlschein, Tobias Meisen
8 Summarization Evaluation meets Short-Answer Grading
Margot Mieskes, Ulrike Padó
9 Experiments on Non-native Speech Assessment and its Consistency
Ziwei Zhou, Sowmya Vajjala, Seyed Vahid Mirnezami
10 The Impact of Spelling Correction and Task Context on Short Answer Assessment for Intelligent Tutoring Systems
Ramon Ziai, Florian Nuxoll, Kordula De Kuthy, Björn Rudzewitz, Detmar Meurers
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
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.
We present a supervised machine learning AND system which tackles semantic similarity between publication titles by means of word embeddings. Word embeddings are integrated as external components, which keeps the model small and efficient, while allowing for easy extensibility and domain adaptation. Initial experiments show that word embeddings can improve the Recall and F score of the binary classification sub-task of AND. Results for the clustering sub-task are less clear, but also promising and overall show the feasibility of the approach.
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.
Die zentrale Aufgabenstellung des Verbundprojektes TextTransfer (Pilot) war eine Machbarkeitsprüfung für die Entwicklung eines Text-Mining-Verfahrens, mit dem Forschungsergebnisse automatisiert auf Hinweise zu Transfer- und Impactpotenzialen untersucht werden können. Das vom Projektkoordinator IDS verantwortete Teilprojekt konzentrierte sich dabei auf die Entwicklung der methodischen Grundlagen, während der Projektpartner TIB vornehmlich für die Bereitstellung eines geeigneten Datensatzes verantwortlich war. Solchen automatisierten Verfahren liegen zumeist textbasierte Daten als physisches Manifest wissenschaftlicher Erkenntnisse zugrunde, die im Falle von TextTransfer (Pilot) als empirische Grundlage herangezogen wurden. Das im Verbund zur Anwendung gebrachte maschinelle Lernverfahren stützte sich ausschließlich auf deutschsprachige Projektendberichte öffentlich geförderter Forschung. Diese Textgattung eignet sich insbesondere hinsichtlich ihrer öffentlichen Verfügbarkeit bei zuständigen Gedächtnisorganisationen und aufgrund ihrer im Vergleich zu anderen Formaten wissenschaftlicher Publikation relativen strukturellen wie sprachlichen Homogenität. TextTransfer (Pilot) ging daher grundsätzlich von der Annahme struktureller bzw. sprachlicher Ähnlichkeit in Berichtstexten aus, bei denen der Nachweis tatsächlich erfolgten Transfers zu erbringen war. Im Folgenden wird in diesen Fällen von Texten bzw. textgebundenen Forschungsergebnissen mit Transfer- und Impactpotenzial gesprochen werden. Es wurde ferner postuliert, dass sich diese Indizien von sprachlichen Eigenschaften in Texten zu Projekten ohne nachzuweisenden bzw. ggf. auch niemals erfolgtem, aber potenziell möglichem Transfer oder Impact unterscheiden lassen. Mit einer Verifizierung dieser Annahmen war es möglich, Transfer- oder Impactwahrscheinlichkeiten in großen Mengen von Berichtsdaten ohne eingehende Lektüre zu prognostizieren.
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.
Corpus REDEWIEDERGABE
(2020)
This article presents the corpus REDEWIEDERGABE, a German-language historical corpus with detailed annotations for speech, thought and writing representation (ST&WR). With approximately 490,000 tokens, it is the largest resource of its kind. It can be used to answer literary and linguistic research questions and serve as training material for machine learning. This paper describes the composition of the corpus and the annotation structure, discusses some methodological decisions and gives basic statistics about the forms of ST&WR found in this corpus.
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.
This paper presents experiments on sentence boundary detection in transcripts of spoken dialogues. Segmenting spoken language into sentence-like units is a challenging task, due to disfluencies, ungrammatical or fragmented structures and the lack of punctuation. In addition, one of the main bottlenecks for many NLP applications for spoken language is the small size of the training data, as the transcription and annotation of spoken language is by far more time-consuming and labour-intensive than processing written language. We therefore investigate the benefits of data expansion and transfer learning and test different ML architectures for this task. Our results show that data expansion is not straightforward and even data from the same domain does not always improve results. They also highlight the importance of modelling, i.e. of finding the best architecture and data representation for the task at hand. For the detection of boundaries in spoken language transcripts, we achieve a substantial improvement when framing the boundary detection problem as a sentence pair classification task, as compared to a sequence tagging approach.
Beyond Citations: Corpus-based Methods for Detecting the Impact of Research Outcomes on Society
(2020)
This paper proposes, implements and evaluates a novel, corpus-based approach for identifying categories indicative of the impact of research via a deductive (top-down, from theory to data) and an inductive (bottom-up, from data to theory) approach. The resulting categorization schemes differ in substance. Research outcomes are typically assessed by using bibliometric methods, such as citation counts and patterns, or alternative metrics, such as references to research in the media. Shortcomings with these methods are their inability to identify impact of research beyond academia (bibliometrics) and considering text-based impact indicators beyond those that capture attention (altmetrics). We address these limitations by leveraging a mixed-methods approach for eliciting impact categories from experts, project personnel (deductive) and texts (inductive). Using these categories, we label a corpus of project reports per category schema, and apply supervised machine learning to infer these categories from project reports. The classification results show that we can predict deductively and inductively derived impact categories with 76.39% and 78.81% accuracy (F1-score), respectively. Our approach can complement solutions from bibliometrics and scientometrics for assessing the impact of research and studying the scope and types of advancements transferred from academia to society.
In this article, we examine the effectiveness of bootstrapping supervised machine-learning polarity classifiers with the help of a domain-independent rule-based classifier that relies on a lexical resource, i.e., a polarity lexicon and a set of linguistic rules. The benefit of this method is that though no labeled training data are required, it allows a classifier to capture in-domain knowledge by training a supervised classifier with in-domain features, such as bag of words, on instances labeled by a rule-based classifier. Thus, this approach can be considered as a simple and effective method for domain adaptation. Among the list of components of this approach, we investigate how important the quality of the rule-based classifier is and what features are useful for the supervised classifier. In particular, the former addresses the issue in how far linguistic modeling is relevant for this task. We not only examine how this method performs under more difficult settings in which classes are not balanced and mixed reviews are included in the data set but also compare how this linguistically-driven method relates to state-of-the-art statistical domain adaptation.
Sentiment Analysis is the task of extracting and classifying opinionated content in natural language texts. Common subtasks are the distinction between opinionated and factual texts, the classification of polarity in opinionated texts, and the extraction of the participating entities of an opinion(-event), i.e. the source from which an opinion emanates and the target towards which it is directed. With the emerging Web 2.0 which describes the shift towards a highly user-interactive communication medium, the amount of subjective content on the World Wide Web is steadily increasing. Thus, there is a growing need for automatically processing this type of content which is provided by sentiment analysis. Both natural language processing, which is the task of providing computational methods for the analysis and representation of natural language, and machine learning, which is the task of building task-specific classification models on the basis of empirical data, may be instrumental in mastering the challenges of the automatic sentiment analysis of written text. Many problems in sentiment analysis have been proposed to be solved with machine learning methods exclusively using a fairly low-level feature design, such as bag of words, containing little linguistic information. In this thesis, we examine the effectiveness of linguistic features in various subtasks of sentiment analysis. Thus, we heavily draw from the insights gained by natural language processing. The application of linguistic features can be applied on various classification methods, be it in rule-based classification, where the linguistic features are directly encoded as a classifier, in supervised machine learning, where these features complement basic low-level features, or in bootstrapping methods, where these features form a rule-based classifier generating a labeled training set from which a supervised classifier can be trained. In this thesis, we will in particular focus on scenarios where the combination of linguistic features and machine learning methods is effective. We will look at common text classification tasks, both coarse-grained and fine-grained, and extraction tasks.
While good results have been achieved for named entity recognition (NER) in supervised settings, it remains a problem that for low resource languages and less studied domains little or no labelled data is available. As NER is a crucial preprocessing step for many natural language processing tasks, finding a way to overcome this deficit in data remains of great interest. We propose a distant supervision approach to NER that is both language and domain independent where we automatically generate labelled training data using gazetteers that we previously extracted from Wikipedia. We test our approach on English, German and Estonian data sets and contribute further by introducing several successful methods to reduce the noise in the generated training data. The tested models beat baseline systems and our results show that distant supervision can be a promising approach for NER when no labelled data is available. For the English model we also show that the distant supervision model is better at generalizing within the same domain of news texts by comparing it against a supervised model on a different test set.
We examine the combination of pattern-based and distributional similarity for the induction of semantic categories. Pattern-based methods are precise and sparse while distributional methods have a higher recall. Given these particular properties we use the prediction of distributional methods as a back-off to pattern-based similarity. Since our pattern-based approach is embedded into a semi-supervised graph clustering algorithm, we also examine how distributional information is best added to that classifier. Our experiments are carried out on 5 different food categorization tasks.
We examine the task of relation extraction in the food domain by employing distant supervision. We focus on the extraction of two relations that are not only relevant to product recommendation in the food domain, but that also have significance in other domains, such as the fashion or electronics domain. In order to select suitable training data, we investigate various degrees of freedom. We consider three processing levels being argument level, sentence level and feature level. As external resources, we employ manually created surface patterns and semantic types on all these levels. We also explore in how far rule-based methods employing the same information are competitive.
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.
We explore the feasibility of contextual healthiness classification of food items. We present a detailed analysis of the linguistic phenomena that need to be taken into consideration for this task based on a specially annotated corpus extracted from web forum entries. For automatic classification, we compare a supervised classifier and rule-based classification. Beyond linguistically motivated features that include sentiment information we also consider the prior healthiness of food items.
One problem of data-driven answer extraction in open-domain factoid question answering is that the class distribution of labeled training data is fairly imbalanced. In an ordinary training set, there are far more incorrect answers than correct answers. The class-imbalance is, thus, inherent to the classification task. It has a deteriorating effect on the performance of classifiers trained by standard machine learning algorithms. They usually have a heavy bias towards the majority class, i.e. the class which occurs most often in the training set. In this paper, we propose a method to tackle class imbalance by applying some form of cost-sensitive learning which is preferable to sampling. We present a simple but effective way of estimating the misclassification costs on the basis of class distribution. This approach offers three benefits. Firstly, it maintains the distribution of the classes of the labeled training data. Secondly, this form of meta-learning can be applied to a wide range of common learning algorithms. Thirdly, this approach can be easily implemented with the help of state-of-the-art machine learning software.
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.
We present a major step towards the creation of the first high-coverage lexicon of polarity shifters. In this work, we bootstrap a lexicon of verbs by exploiting various linguistic features. Polarity shifters, such as ‘abandon’, are similar to negations (e.g. ‘not’) in that they move the polarity of a phrase towards its inverse, as in ‘abandon all hope’. While there exist lists of negation words, creating comprehensive lists of polarity shifters is far more challenging due to their sheer number. On a sample of manually annotated verbs we examine a variety of linguistic features for this task. Then we build a supervised classifier to increase coverage. We show that this approach drastically reduces the annotation effort while ensuring a high-precision lexicon. We also show that our acquired knowledge of verbal polarity shifters improves phrase-level sentiment analysis.
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.
In recent years, text classification in sentiment analysis has mostly focused on two types of classification, the distinction between objective and subjective text, i.e. subjectivity detection, and the distinction between positive and negative subjective text, i.e. polarity classification. So far, there has been little work examining the distinction between definite polar subjectivity and indefinite polar subjectivity. While the former are utterances which can be categorized as either positive or negative, the latter cannot be categorized as either of these two categories. This paper presents a small set of domain independent features to detect indefinite polar sentences. The features reflect the linguistic structure underlying these types of utterances. We give evidence for the effectiveness of these features by incorporating them into an unsupervised rule-based classifier for sentence-level analysis and compare its performance with supervised machine learning classifiers, i.e. Support Vector Machines (SVMs) and Nearest Neighbor Classifier (kNN). The data used for the experiments are web-reviews collected from three different domains.
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.
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.
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.
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.
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.
In this paper, we investigate the role of predicates in opinion holder extraction. We will examine the shape of these predicates, investigate what relationship they bear towards opinion holders, determine what resources are potentially useful for acquiring them, and point out limitations of an opinion holder extraction system based on these predicates. For this study, we will carry out an evaluation on a corpus annotated with opinion holders. Our insights are, in particular, important for situations in which no labelled training data are available and only rule-based methods can be applied.
In this paper, we compare three different generalization methods for in-domain and cross-domain opinion holder extraction being simple unsupervised word clustering, an induction method inspired by distant supervision and the usage of lexical resources. The generalization methods are incorporated into diverse classifiers. We show that generalization causes significant improvements and that the impact of improvement depends on the type of classifier and on how much training and test data differ from each other. We also address the less common case of opinion holders being realized in patient position and suggest approaches including a novel (linguistically-informed) extraction method how to detect those opinion holders without labeled training data as standard datasets contain too few instances of this type.
Researchers in many disciplines, sometimes working in close cooperation, have been concerned with modeling textual data in order to account for texts as the prime information unit of written communication. The list of disciplines includes computer science and linguistics as well as more specialized disciplines like computational linguistics and text technology. What many of these efforts have in common is the aim to model textual data by means of abstract data types or data structures that support at least the semi-automatic processing of texts in any area of written communication.