Menü Überspringen
Kontakt
Presse
Deutsch
English
Not track
Datenverarbeitung
Suche
Anmelden
DIPF aktuell
Forschung
Infrastrukturen
Institut
Zurück
Kontakt
Presse
Deutsch
English
Not track
Datenverarbeitung
Suche
Startseite
>
Forschung
>
Publikationen
>
Publikationendatenbank
Ergebnis der Suche in der DIPF Publikationendatenbank
Ihre Abfrage:
(Schlagwörter: "Linguistik")
zur erweiterten Suche
Suchbegriff
Nur Open Access
Suchen
Markierungen aufheben
Alle Treffer markieren
Export
115
Inhalte gefunden
Alle Details anzeigen
Which argument is more convincing? Analyzing and predicting convincingness of Web arguments using […]
Habernal, Ivan; Gurevych, Iryna
Sammelbandbeitrag
| Aus: Association for Computational Linguistics (Hrsg.): Proceedings of the 54th annual meeting of the Association for Computational Linguistics (ACL 2016): Long papers | Stroudsburg; PA: Association for Computational Linguistics | 2016
36970 Endnote
Autor*innen:
Habernal, Ivan; Gurevych, Iryna
Titel:
Which argument is more convincing? Analyzing and predicting convincingness of Web arguments using bidirectional LSTM
Aus:
Association for Computational Linguistics (Hrsg.): Proceedings of the 54th annual meeting of the Association for Computational Linguistics (ACL 2016): Long papers, Stroudsburg; PA: Association for Computational Linguistics, 2016 , S. 1589-1599
URL:
http://www.aclweb.org/anthology/P16-1150
Dokumenttyp:
4. Beiträge in Sammelbänden; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Algorithmus; Argumentation; Automatisierung; Computerlinguistik; Kommunikation; Online; Prognose; Qualität; Rhetorik; Soziale Software; Textanalyse; Überzeugung; World wide web 2.0
Abstract (english):
We propose a new task in the field of computational argumentation in which we investigate qualitative properties of Web arguments, namely their convincingness. We cast the problem as relation classification, where a pair of arguments having the same stance to the same prompt is judged. We annotate a large datasets of 16k pairs of arguments over 32 topics and investigate whether the relation "A is more convincing than B" exhibits properties of total ordering; these findings are used as global constraints for cleaning the crowdsourced data. We propose two tasks: (1) predicting which argument from an argument pair is more convincing and (2) ranking all arguments to the topic based on their convincingness. We experiment with feature-rich SVM and bidirectional LSTM and obtain 0.76-0.78 accuracy and 0.35-0.40 Spearman's correlation in a cross-topic evaluation. We release the newly created corpus UKPConvArg1 and the experimental software under open licenses. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
What makes word-level neural machine translation hard. A case study on English-German translation
Hirschmann, Fabian; Nam, Jinseok; Fürnkranz, Johannes
Sammelbandbeitrag
| Aus: The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING) | Osaka: The COLING 2016 Organizing Committee | 2016
36983 Endnote
Autor*innen:
Hirschmann, Fabian; Nam, Jinseok; Fürnkranz, Johannes
Titel:
What makes word-level neural machine translation hard. A case study on English-German translation
Aus:
The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING), Osaka: The COLING 2016 Organizing Committee, 2016 , S. 3199-3208
URL:
http://aclweb.org/anthology/C/C16/C16-1301.pdf
Dokumenttyp:
4. Beiträge in Sammelbänden; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Automatisierung; Computerlinguistik; Computerunterstütztes Verfahren; Deutsch; Englisch; Syntax; Übersetzung; Wort; Wörterbuch
Abstract (english):
Traditional machine translation systems often require heavy feature engineering and the combination of multiple techniques for solving different subproblems. In recent years, several end-to-end learning architectures based on recurrent neural networks have been proposed. Unlike traditional systems, Neural Machine Translation (NMT) systems learn the parameters of the model and require only minimal preprocessing. Memory and time constraints allow to take only a fixed number of words into account, which leads to the out-of-vocabulary (OOV) problem. In this work, we analyze why the OOV problem arises and why it is considered a serious problem in German. We study the effectiveness of compound word splitters for alleviating the OOV problem, resulting in a 2.5+ BLEU points improvement over a baseline on the WMT'14 German-to-English translation task. For English-to-German translation, we use target-side compound splitting through a special syntax during training that allows the model to merge compound words and gain 0.2 BLEU points. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
Modeling extractive sentence intersection via subtree entailment
Levy, Omer; Dagan, Ido; Stanovsky, Gabriel; Eckle-Kohler, Judith; Gurevych, Iryna
Sammelbandbeitrag
| Aus: The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING) | Osaka: The COLING 2016 Organizing Committee | 2016
36987 Endnote
Autor*innen:
Levy, Omer; Dagan, Ido; Stanovsky, Gabriel; Eckle-Kohler, Judith; Gurevych, Iryna
Titel:
Modeling extractive sentence intersection via subtree entailment
Aus:
The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING), Osaka: The COLING 2016 Organizing Committee, 2016 , S. 2891-2901
URL:
http://www.aclweb.org/anthology/C/C16/C16-1272.pdf
Dokumenttyp:
4. Beiträge in Sammelbänden; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Algorithmus; Computerlinguistik; Daten; Klassifikation; Semantik; Struktur; Syntax; Text
Abstract (english):
Sentence intersection captures the semantic overlap of two texts, generalizing over paradigms such as textual entailment and semantic text similarity. Despite its modeling power, it has received little attention because it is difficult for non-experts to annotate. We analyze 200 pairs of similar sentences and identify several underlying properties of sentence intersection. We leverage these insights to design an algorithm that decomposes the sentence intersection task into several simpler annotation tasks, facilitating the construction of a high quality dataset via crowdsourcing. We implement this approach and provide an annotated dataset of 1,764 sentence intersections. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
Semi-automatic detection of cross-lingual marketing blunders based on pragmatic label propagation […]
Meyer, Christian M.; Eckle-Kohler, Judith; Gurevych, Iryna
Sammelbandbeitrag
| Aus: The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING) | Osaka: The COLING 2016 Organizing Committee | 2016
36985 Endnote
Autor*innen:
Meyer, Christian M.; Eckle-Kohler, Judith; Gurevych, Iryna
Titel:
Semi-automatic detection of cross-lingual marketing blunders based on pragmatic label propagation in Wiktionary
Aus:
The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING), Osaka: The COLING 2016 Organizing Committee, 2016 , S. 2071-2081
URL:
http://www.aclweb.org/anthology/C/C16/C16-1195.pdf
Dokumenttyp:
4. Beiträge in Sammelbänden; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Begriff; Computerlinguistik; Computerunterstütztes Verfahren; Fremdsprache; Marketing; Problem
Abstract (english):
We introduce the task of detecting cross-lingual marketing blunders, which occur if a trade name resembles an inappropriate or negatively connotated word in a target language. To this end, we suggest a formal task definition and a semi-automatic method based the propagation of pragmatic labels from Wiktionary across sense-disambiguated translations. Our final tool assists users by providing clues for problematic names in any language, which we simulate in two experiments on detecting previously occurred marketing blunders and identifying relevant clues for established international brands. We conclude the paper with a suggested research roadmap for this new task. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
TAXI. A taxonomy induction method based on lexico-syntactic patterns, substrings and focused […]
Panchenko, Alexander; Faralli, Stefano; Ruppert, Eugen; Remus, Steffen; Naets, Hubert; […]
Sammelbandbeitrag
| Aus: Association for Computational Linguistics (Hrsg.): Proceedings of the 10th International Workshop on Semantic Evaluation co-located with NAACL 2016 | Stroudsburg; PA: Association for Computational Linguistics | 2016
37068 Endnote
Autor*innen:
Panchenko, Alexander; Faralli, Stefano; Ruppert, Eugen; Remus, Steffen; Naets, Hubert; Fairon, Cédrick; Ponzetto, Simone Paolo; Biemann, Chris
Titel:
TAXI. A taxonomy induction method based on lexico-syntactic patterns, substrings and focused crawling
Aus:
Association for Computational Linguistics (Hrsg.): Proceedings of the 10th International Workshop on Semantic Evaluation co-located with NAACL 2016, Stroudsburg; PA: Association for Computational Linguistics, 2016 , S. 1320-1327
URL:
http://www.aclweb.org/anthology/S16-1206
Dokumenttyp:
4. Beiträge in Sammelbänden; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Computerlinguistik; Taxonomie; Methode; Sprache; Englisch; Niederländisch; Französisch; Italienisch; Text; Begriff; Struktur; Evaluation
Abstract:
We present a system for taxonomy construction that reached the first place in all sub-tasks of the SemEval 2016 challenge on Taxonomy Extraction Evaluation. Our simple yet effective approach harvests hypernyms with substring inclusion and Hearst-style lexico-syntactic patterns from domain-specific texts obtained via language model based focused crawling. Extracted taxonomies are evaluated on English, Dutch, French and Italian for three domains each (Food, Environment and Science). Evaluations against a gold standard and by human judgment show that our method outperforms more complex and knowledge-rich approaches on most domains and languages. Furthermore, to adapt the method to a new domain or language, only a small amount of manual labour is needed. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
Task-oriented intrinsic evaluation of semantic textual similarity
Reimers, Nils; Beyer, Philip; Gurevych, Iryna
Sammelbandbeitrag
| Aus: The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING) | Osaka: The COLING 2016 Organizing Committee | 2016
36988 Endnote
Autor*innen:
Reimers, Nils; Beyer, Philip; Gurevych, Iryna
Titel:
Task-oriented intrinsic evaluation of semantic textual similarity
Aus:
The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING), Osaka: The COLING 2016 Organizing Committee, 2016 , S. 87-96
URL:
https://www.aclweb.org/anthology/C/C16/C16-1009.pdf
Dokumenttyp:
4. Beiträge in Sammelbänden; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Computerlinguistik; Evaluation; Korrelation; Messverfahren; Semantik; Systemvergleich; Text
Abstract (english):
Semantic Textual Similarity (STS) is a foundational NLP task and can be used in a wide range of tasks. To determine the STS of two texts, hundreds of different STS systems exist, however, for an NLP system designer, it is hard to decide which system is the best on. To answer this question, an intrinsic evaluation of the STS systems is conducted by comparing the output of the system to human judgments on semantic similarity. The comparison is usually done using Pearson cor- relation. In this work, we show that relying on intrinsic evaluations with Pearson correlation can be misleading. In three common STS based tasks we could observe that the Pearson correlation was especially ill-suited to detect the best STS system for the task and other evaluation measures were much better suited. In this work we define how the validity of an intrinsic evaluation can be assessed and compare different intrinsic evaluation methods. Understanding of the properties of the targeted task is crucial and we propose a framework for conducting the intrinsic evaluation which takes the properties of the targeted task into account. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
Still not there? Comparing traditional sequence-to-sequence models to encoder-decoder neural […]
Schnober, Carsten; Eger, Steffen; Do Dinh, Erik-Lân; Gurevych, Iryna
Sammelbandbeitrag
| Aus: The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING) | Osaka: The COLING 2016 Organizing Committee | 2016
36984 Endnote
Autor*innen:
Schnober, Carsten; Eger, Steffen; Do Dinh, Erik-Lân; Gurevych, Iryna
Titel:
Still not there? Comparing traditional sequence-to-sequence models to encoder-decoder neural networks on monotone string translation tasks
Aus:
The COLING 2016 Organizing Committee (Hrsg.): Proceedings of the 26th International Conference on Computational Linguistics (COLING), Osaka: The COLING 2016 Organizing Committee, 2016 , S. 1703-1714
URL:
http://aclweb.org/anthology/C16-1160
Dokumenttyp:
4. Beiträge in Sammelbänden; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Computerlinguistik; Computerunterstütztes Verfahren; Korrektur; Rechtschreibung; Text
Abstract (english):
We analyze the performance of encoder-decoder neural models and compare them with well-known established methods. The latter represent different classes of traditional approaches that are applied to the monotone sequence-to-sequence tasks OCR post-correction, spelling correction, grapheme-to-phoneme conversion, and lemmatization. Such tasks are of practical relevance for various higher-level research fields including digital humanities, automatic text correction, and speech recognition. We investigate how well generic deep-learning approaches adapt to these tasks, and how they perform in comparison with established and more specialized methods, including our own adaptation of pruned CRFs. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
Recognizing the absence of opposing arguments in persuasive essays
Stab, Christian; Gurevych, Iryna
Sammelbandbeitrag
| Aus: Association for Computational Linguistics (Hrsg.): Proceedings of the 3rd Workshop on Argument Mining held in conjunction with the 2016 Annual Meeting of the Association for Computational Linguistics (ACL 2016) | Stroudsburg; PA: Association for Computational Linguistics | 2016
36976 Endnote
Autor*innen:
Stab, Christian; Gurevych, Iryna
Titel:
Recognizing the absence of opposing arguments in persuasive essays
Aus:
Association for Computational Linguistics (Hrsg.): Proceedings of the 3rd Workshop on Argument Mining held in conjunction with the 2016 Annual Meeting of the Association for Computational Linguistics (ACL 2016), Stroudsburg; PA: Association for Computational Linguistics, 2016 , S. 113-118
URL:
http://aclweb.org/anthology/W/W16/W16-2813.pdf
Dokumenttyp:
4. Beiträge in Sammelbänden; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Argumentation; Aufsatz; Computerlinguistik; Dokument; Gegensatz; Klassifikation; Modell
Abstract (english):
In this paper, we introduce an approach for recognizing the absence of opposing arguments in persuasive essays. We model this task as a binary document classification and show that adversative transitions in combination with unigrams and syntactic production rules significantly outperform a challenging heuristic baseline. Our approach yields an accuracy of 75.6% and 84% of human performance in a persuasive essay corpus with various topics. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
Predicting the spelling difficulty of words for language learners
Beinborn, Lisa; Zesch, Torsten; Gurevych, Iryna
Sammelbandbeitrag
| Aus: Association for Computational Linguistic (Hrsg.): Proceedings of the 11th workshop on innovative use of NLP for building educational applications held in conjunction with NAACL 2016 | Stroudsburg; PA: Association for Computational Linguistics | 2016
36973 Endnote
Autor*innen:
Beinborn, Lisa; Zesch, Torsten; Gurevych, Iryna
Titel:
Predicting the spelling difficulty of words for language learners
Aus:
Association for Computational Linguistic (Hrsg.): Proceedings of the 11th workshop on innovative use of NLP for building educational applications held in conjunction with NAACL 2016, Stroudsburg; PA: Association for Computational Linguistics, 2016 , S. 73-83
URL:
https://www.ukp.tu-darmstadt.de/fileadmin/user_upload/Group_UKP/wall/BEA2016_SpellingDifficulty.pdf
Dokumenttyp:
4. Beiträge in Sammelwerken; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Computerlinguistik; Deutsch; Englisch; Fehler; Fremdsprache; Italienisch; Modell; Muttersprache; Phonetik; Psycholinguistik; Rechtschreibung
Abstract (english):
In many language learning scenarios, it is important to anticipate spelling errors. We model the spelling difficulty of words with new features that capture phonetic phenomena and are based on psycholinguistic findings. To train our model, we extract more than 140,000 spelling errors from three learner corpora covering English, German and Italian essays. The evaluation shows that our model can predict spelling difficulty with an accuracy of over 80% and yields a stable quality across corpora and languages. In addition, we provide a thorough error analysis that takes the native language of the learners into account and provides insights into cross-lingual transfer effects. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
Token-level metaphor detection using neural networks
Do Dinh, Erik-Lân; Gurevych, Iryna
Sammelbandbeitrag
| Aus: Association for Computational Linguistics (Hrsg.): Proceedings of the fourth workshop on metaphor in NLP held in conjunction with NAACL 2016 | Stroudsburg; PA: Association for Computational Linguistics | 2016
36978 Endnote
Autor*innen:
Do Dinh, Erik-Lân; Gurevych, Iryna
Titel:
Token-level metaphor detection using neural networks
Aus:
Association for Computational Linguistics (Hrsg.): Proceedings of the fourth workshop on metaphor in NLP held in conjunction with NAACL 2016, Stroudsburg; PA: Association for Computational Linguistics, 2016 , S. 28-33
URL:
https://www.ukp.tu-darmstadt.de/fileadmin/user_upload/Group_UKP/publikationen/2016/2016_DoDinh_NAACL_pages.pdf
Dokumenttyp:
4. Beiträge in Sammelwerken; Tagungsband/Konferenzbeitrag/Proceedings
Sprache:
Englisch
Schlagwörter:
Automatisierung; Computerlinguistik; Netzwerk; Semantik; Textanalyse
Abstract (english):
Automatic metaphor detection usually relies on various features, incorporating e.g. selectional preference violations or concreteness ratings to detect metaphors in text. These features rely on background corpora, hand-coded rules or additional, manually created resources, all specific to the language the system is being used on. We present a novel approach to metaphor detection using a neural network in combination with word embeddings, a method that has already proven to yield promising results for other natural language processing tasks. We show that foregoing manual feature engineering by solely relying on word embeddings trained on large corpora produces comparable results to other systems, while removing the need for additional resources. (DIPF/Orig.)
DIPF-Abteilung:
Informationszentrum Bildung
Markierungen aufheben
Alle Treffer markieren
Export
<
1
2
3
(aktuell)
4
...
12
>
Alle anzeigen
(115)