DKPro (Darmstadt Knowledge Processing) 2017


Notes:

  • Story understanding
  • Story understanding system
  • Story understanding systems

Resources:

  • clarino .. language analysis portal
  • dkpro.github.io .. projects focusing on re-usable natural language processing software
  • rocstories .. story cloze test and rocstories corpora

Wikipedia:

References:

See also:

Apache OpenNLP 2016 | Sentence Splitter 2017 | Text Classification & Dialog Systems 2017 | WEKA & Dialog Systems 2017


An Arranged Marriage: Integrating DKPro Core in the Language Analysis Portal
M Kouylekov, E Lapponi, S Oepen, RE de Castilho – clarin.eu
Abstract In this paper we describe an ongoing effort to create an inter-operation framework that makes accessible the DKPro Core repository of Natural Language Processing (NLP) tools with the CLARINO Language Analysis Portal (LAP). The effort includes creating a …

Paraphrase detection using machine translation and textual similarity algorithms
D Kravchenko – Conference on Artificial Intelligence and Natural …, 2017 – Springer
… Table 2. Toolkits results. Toolkit. Accuracy. F1 score. SEMILAR. 62.26. 60.15. DKPro Similarity. 61.14. 59.30. Python difflib. 57.07. 53.51. Algorithm of [1]. 60.02 … F1 score. SEMILAR. 78.95. 75.19. 78.41. 74.71. 78.40. 74.67. DKPro Similarity. 78.38. 74.30. 77.84. 73.83. 78.74. 74.79 …

Representation and Interchange of Linguistic Annotation. An In-Depth, Side-by-Side Comparison of Three Designs
RE de Castilho, N Ide, E Lapponi, S Oepen… – Proceedings of the 11th …, 2017 – aclweb.org
… Specifically, we will take an in-depth look at how the results of morpho- syntactic analysis are represented in (a) the DKPro Core component collection1 (Eckart de Castilho and Gurevych, 2014), (b) the Language Analysis Portal2 (LAP; Lapponi et al … Figure 2: DKPro Core Zoom …

Neural, non-neural and hybrid stance detection in tweets on catalan independence
M Wojatzki, T Zesch – Proceedings of the Second …, 2017 – pdfs.semanticscholar.org
… Since both approaches require tokenized texts, we apply the Twitter-specific Ark- Tokenizer [4] from the DKPro Core framework (v1.9.0) [2] beforehand … For training the model we rely on an SVM with a linear kernel provided by the DKPro-TC framework …

Using Argumentative Structure to Grade Persuasive Essays
A Stiegelmayr, M Mieskes – … Conference of the German Society for …, 2017 – Springer
… (1977) shows a moderate agreement. All values were calculated using DKPro Statistics 4 (Meyer et al. 2014). 4 Experimental Setup. We use DKPro Components 5 , such as DKPro Core, DKPro TC and Uby for our experiments …

Automatic Coding of PISA Short Text Responses Across Multiple Languages
F Zehner, C Sälzer, F Goldhammer – researchgate.net
Page 1. Automatic Coding of PISA Short Text Responses Across Multiple Languages Fabian Zehner, Christine Sälzer, Frank Goldhammer German Institute for International Educational Research (DIPF), Centre for Technology …

The Concept of Lexical Platform
M Piasecki, T Walkowiak, E Rudnicka, T Naskr?t… – pdfs.semanticscholar.org
… 2.2 Platforms UBY LMF2 platform ([7]) was built as a solution for integrating LRs on both structural and semantic level. 12 LRs3 have been linked into a complex system. 2 https://dkpro.github.io/ dkpro-uby/ 3 https://www.ukp.tu-darmstadt.de/data/lexical-resources/uby/ Page 3 …

New Frontiers in Supervised Word Sense Disambiguation: Building Multilingual Resources and Neural Models on a large scale
A Raganato, R Navigli – 2017 – iris.uniroma1.it
… differ in format, construction guidelines and underlying sense inventory. In fact, also a general-purpose framework for word sense disambiguation, ie, DKPro WSD [Miller et al., 2013], that is designed to support the needs of WSD researchers, has a …

Clear-cut methodology for Arabic OCR and post-correction with low technical skilled annotators
AG Martínez, T Feige, T Eich – … of the 2nd International Conference on …, 2017 – dl.acm.org
… To carry out the xmi conversion, we built a java program that uses the DKPro Core toolkit, a collection of software components for natural language processing based on the Apache UIMA framework [11] 4. Just before dumping the data into xmi format, we pass the texts through …

Watset: automatic induction of synsets from a graph of synonyms
D Ustalov, A Panchenko, C Biemann – arXiv preprint arXiv:1704.07157, 2017 – arxiv.org
… readable and not for- mally structured. While semantic relations can be automatically extracted using tools such as DKPro JWKTL4 and Wikokit5, words in these re- lations are not disambiguated. For instance, the synonymy pairs …

Evaluation of uncertainty in the measurement of sense of natural language constructions
OV Bisikalo, OM Vasilevskyi – International Journal of …, 2017 – metrology-journal.org
… The practical significance of the results is to obtain software technology to produce tools based on the DKPro Core linguistics … Muller, J. Steimle, M. Weimer, T. Zesch, Darmstadt knowledge processing repository based on UIMA [Electronic resource], February 9, 2007 (2007 …

Mining Social Science Publications for Survey Variables
A Zielinski, P Mutschke – Proceedings of the Second Workshop on NLP …, 2017 – aclweb.org
… 4.1 Feature Extraction For pre-processing, we use a pipeline of tools from DKPro (de Castilho and Gurevych, 2014) that sup- ports tokenization, lemmatization, part-of-speech tagging and Named Entity Recognition … Dkpro simi- larity: An open source framework for text similarity …

RelTextRank: An Open Source Framework for Building Relational Syntactic-Semantic Text Pair Representations
K Tymoshenko, A Moschitti, M Nicosia… – Proceedings of ACL 2017 …, 2017 – aclweb.org
… Currently, we use a custom UIMA type system defined for our pipeline, however, in future we plan to use the type systems used in other widely used UIMA pipelines, eg, DKPro (de Castilho and Gurevych, 2014) … Dkpro similarity: An open source framework for text similarity …

Flexible NLP Pipelines for Digital Humanities Research
JM van der Zwaan, W Smink, A Sools… – 4th Digital Humanities …, 2017 – dh2017.adho.org
… command line tool. This flexibility is missing from existing frameworks for creating NLP pipelines, such as DKPro (Eckart de Castilho, and Gurevych, 2015) using the UIMA framework (Ferrucci, and Lally, 2004). In addition to …

Building a Web-Scale Dependency-Parsed Corpus from CommonCrawl
A Panchenko, E Ruppert, S Faralli, SP Ponzetto… – arXiv preprint arXiv …, 2017 – arxiv.org
… 3.3. Linguistic Analysis of Texts Linguistic analysis consists of four stages presented in Fig- ure 1 and is implemented using the Apache Hadoop frame- work9 for parallelization and the Apache UIMA frame- work10 for integration of linguistic analysers via the DKPro Core library …

Semantic integration of conceptual models
LM Costa, C Sousa, C Pereira – 2017 – repositorio.inesctec.pt
… 352 L. Costa et al. 8050120@estg.ipp.pt Page 5. The similarity value between two concepts, in this second phase, is obtained through the application of the following syntactic measures: Levenshtein, Jaro or MongeElkan (available in SimPack7 and DKPro8 APIs) …

Extending Scientific Literature Search by Including the Author’s Writing Style.
A Rexha, M Kröll, H Ziak, R Kern – BIR@ ECIR, 2017 – pdfs.semanticscholar.org
… Table 2: List of stylometric features used (Tweedie & Baayen, 2002). 4 https://dkpro.github.io/ dkpro-statistics/ BIR 2017 Workshop on Bibliometric-enhanced Information Retrieval 97 Page 6. First, we select a list of stylometric features to extract from the source and the target texts …

Recognizing insufficiently supported arguments in argumentative essays
C Stab, I Gurevych – Proceedings of the 15th Conference of the …, 2017 – aclweb.org
… All three annotators independently annotated an evaluation set of 433 arguments. We evaluated the agreement between the annotators using sev- eral inter-annotator agreement measures imple- mented in DKPro Agreement (Meyer et al., 2014) …

” let’s Talk About Talks”-Supporting Knowledge Exchange Processes on Wiki Discussion Pages
S Heimbuch, D Bodemer – 2017 – psyarxiv.com
… inevitable. Computational advancements and research on NLP In recent years yielded into more sophisticated li- braries, tools and frameworks (eg DKPro TC) for analyses in wiki-related contexts (Daxenberger and Gurevych, 2014) …

Investigating neural architectures for short answer scoring
B Riordan, A Horbach, A Cahill, T Zesch… – Proceedings of the 12th …, 2017 – aclweb.org
… This baseline is based on DkPro TC (Daxenberger et al., 2014) and relies on support vector classification using Weka (Hall et al., 2009). We preprocess the data using the ClearNlp Segmenter 2 via DKPro Core (Eckart de Castilho and Gurevych, 2014). The …

Unsupervised, knowledge-free, and interpretable word sense disambiguation
A Panchenko, F Marten, E Ruppert, S Faralli… – arXiv preprint arXiv …, 2017 – arxiv.org
… These representations were successfully used to perform WSD using the IMS. DKPro WSD (Miller et al., 2013) is a modu- lar, extensible Java framework for word sense dis- ambiguation … DKPro WSD: A Generalized UIMA-based Framework for Word Sense Disam- biguation …

Semantic Technologies for Re-Use of Clinical Routine Data.
M Kreuzthaler, C Martínez-Costa… – Studies in health …, 2017 – books.google.com
… HITEx [12] is based on GATE [13], an NLP engine. The DKPro Core collection also has to be mentioned [14] in the context of NLP functionalities … Bär, D., Zesch, T., & Gurevych, I.(2013, August). DKPro Similarity: An Open Source Framework for Text Similarity …

LSDSem 2017: Exploring data generation methods for the story cloze test
M Bugert, Y Puzikov, A Rücklé, J Eckle-Kohler… – Proceedings of the 2nd …, 2017 – aclweb.org
… story. All features were extracted using the DKPro TC Frame- work (Daxenberger et al., 2014). 3.2 Neural Network The … Zesch. 2014. DKPro TC: A Java-based Framework for Supervised Learning Experiments on Textual Data. In …

Russian Tagging and Dependency Parsing Models for Stanford CoreNLP Natural Language Toolkit
A Putintseva – Knowledge Engineering and Semantic Web: 8th …, 2017 – books.google.com
… An Annotation is stored as a typesafe heterogeneous map [1]. This architecture provides a light weight and easy to understand and use solution (in comparison to UIMA, DKPro Core and several other packages), which also benefits from a vast documentation …

Making topic modeling easy: A programming library in Python
F Jannidis, S Pielström, C Schöch, T Vitt – dh2017.adho.org
… A convenient tool for NLP analysis during prepro- cessing does exist in the DARIAH-DKPro- Wrapper (DDW) that covers a wide range of NLP tasks for many different languages and generates annotations in a Py- thon-Pandas compatible format easily usable within our library …

Russian Tagging and Dependency Parsing Models for Stanford CoreNLP Natural Language Toolkit
L Kovriguina, I Shilin, A Shipilo, A Putintseva – International Conference on …, 2017 – Springer
… An Annotation is stored as a typesafe heterogeneous map [1]. This architecture provides a light weight and easy to understand and use solution (in comparison to UIMA, DKPro Core and several other packages), which also benefits from a vast documentation …

nutIE—A modern open source natural language processing toolkit
S Žitnik, D Draskovic, B Nikoli?… – … Forum (TELFOR), 2017 …, 2017 – ieeexplore.ieee.org
… UIMA is developed more as a framework for large-scale data processing and by itself does not include many NLP components. Though, an open source project DKPro Core [5], which is based on UIMA, solves the problem by providing UIMA components …

Docforia: A Multilayer Document Model
M Klang, P Nugues – Proceedings of the 21st Nordic Conference on …, 2017 – aclweb.org
… 2010), and the oft cited UIMA project (Fer- rucci and Lally, 2004). Some tools also extend UIMA such as DKPro Core (Eckart de Castilho and Gurevych, 2014). In contrast to the UIMA project (Ferrucci and Lally, 2004), which …

Lexical-semantic resources: yet powerful resources for automatic personality classification
XS Vu, L Flekova, L Jiang, I Gurevych – arXiv preprint arXiv:1711.09824, 2017 – arxiv.org
… categorized into noun.plant). In this paper, DKPro Uby (Gurevych et al., 2012) is further employed to extract all above required information to rep- resent in pos and sux from given texts. Sentiment Features For sentiment features …

The Influence of Spelling Errors on Content Scoring Performance
A Horbach, Y Ding, T Zesch – Proceedings of the 4th Workshop on …, 2017 – aclweb.org
… and “pub- lic leaderbord” data for testing. We preprocess the data using the ClearNLP segmenter and POS tagger provided through DKPro Core (Eckart de Castilho and Gurevych, 2014). We use a standard feature set often …

Fine-grained essay scoring of a complex writing task for native speakers
A Horbach, D Scholten-Akoun, Y Ding… – Proceedings of the 12th …, 2017 – aclweb.org
… All data is preprocessed using a DKPro pipeline (Eckart de Castilho and Gurevych, 2014) consisting of segmentation, POS-tagging (both OpenNLP2), lemmatization using the MateLem- matizer (Anders et al., 2010) and parsing using the StanfordParser (Rafferty and Manning …

Developing A Web-based Workbook for English
B Rudzewitz, R Ziai, K De Kuthy, D Meurers – spraakbanken.gu.se
… NLP steps: segmentation, POS tagging, lemmatization, morphological analysis and dependency parsing using DKPro (de Castilho and Gurevych 2014) – Manual correction of the automatically obtained linguistic information • When teacher selects Feedback interface …

A tool for extracting sense-disambiguated example sentences through user feedback
B Boullosa, RE de Castilho, A Geyken… – Proceedings of the …, 2017 – aclweb.org
… As for the technologies, the tool was developed using Java Wicket and relying on Solr1, Mallet 2, DKPro Core (Eckart de Castilho and Gurevych, 2014), Tomcat and MySQL. The next subsections describe the system in more detail. 3.1 Searching …

Mnogoznal: an unsupervised system for word sense disambiguation
D Ustalov, D Teslenko, A Panchenko… – … Multi-Conference on, 2017 – ieeexplore.ieee.org
… the WordNet and similar lexical ontologies. DKPro WSD [6] is general-purpose framework for WSD that uses a lexical ontology as the sense inventory and offers the variety of WordNet-based algorithms. Babelfy [7] is a WSD …

Trust, but Verify! Better Entity Linking through Automatic Verification
B Heinzerling, M Strube, CY Lin – Proceedings of the 15th Conference of …, 2017 – aclweb.org
… Feature extraction is implemented as a UIMA pipeline (Ferrucci and Lally, 2004); using the Stanford CoreNLP (Manning et al., 2014) UIMA components provided by DKPro (Eckart de Castilho and Gurevych, 2014) for text segmen- tation, POS tagging, and named entity recogni …

Combining dependency parsing and a lexical network based on lexical functions for the identification of collocations
A Fonseca, F Sadat, F Lareau – International Conference on Computational …, 2017 – Springer
… Then the lemmatization is performed with DKPRO LanguageToolLemmatizer 9 . MaltParser 10 is used as a dependency parser. Finally, we use DKPRO to generate a file containing all phrases in the CONLL-X format. We perform two types of analysis …

Developing A Web-based Workbook for English Supporting the Interaction of Students and Teachers
B Rudzewitz, R Ziai, K De Kuthy, D Meurers – … of the Joint 6th Workshop on …, 2017 – ep.liu.se
… We use the Unstructured Information Management Architecture (UIMA, Ferrucci and Lally, 2004) along with the DKPro component repository (de Castilho and Gurevych, 2014), which provides ac- cess to many state-of-the-art NLP tools within an industry-strength processing …

Semantic Integration of Conceptual Models
L Costa, C Sousa, C Pereira – World Conference on Information Systems …, 2017 – Springer
… syntactically. The similarity value between two concepts, in this second phase, is obtained through the application of the following syntactic measures: Levenshtein, Jaro or MongeElkan (available in SimPack 7 and DKPro 8 APIs) …

Improving Semantic Textual Similarity with Phrase Entity Alignment
V Sowmya, BV Vardhan, MSVSB Raju – International Journal of Intelligent …, 2017 – inass.org
… The system NTNU-CORE [20] used TakeLab features, DKPro features in addition with GateWordMatch feature and trained the system using Support Vector Regression(SVR) which attained maximum correlation 0.4035 for SMT dataset …

Design of an Information Extraction Pipeline for German Clinical Texts
M Kittner, B Bajwa, D Rieke, M Lamping… – … for Health Care …, 2017 – banzai-deim.urv.cat
… Alternatively, generation of synonyms, translation of terms to English or edit distance measures such as Levenshtein or Jaccard could be helpful. Addition- ally, we will consider application of NLP methods available within GATE or the DKPro Core collection …

Reflex-SMAS, a Complex Adaptive System: An Empirical Evaluation
H Assoudi, H Lounis – International Conference on E-Technologies, 2017 – Springer
… As depicted in the diagram above, our Relflex-SMAS prototype core was implemented in Java using the open source ABMS framework Repast Simphony (2.1) [19, 20], the open source framework for Text Similarity DKPro Similarity (2.1.0) [21], the open source R language (R 3.1 …

Lsdsem 2017 shared task: The story cloze test
N Mostafazadeh, M Roth, A Louis… – Proceedings of the 2nd …, 2017 – aclweb.org
… 3 tbmihaylov LSTM ? Word2Vec ? 0.728 4 ukp BiLSTM Spring 2016, Winter 2017 GloVe Stanford CoreNLP, DKPro TC 0.717 5 acoli SVM ? GloVe, Word2Vec ? 0.700 6 roemmele RNN Spring 2016, Winter 2017 Skip-Thought ? 0.672 …

Enriching Argumentative Texts with Implicit Knowledge
M Becker, M Staniek, V Nastase, A Frank – International Conference on …, 2017 – Springer
… 3. Spearman correlation results on the German version of the MC30: 0.76; RG65: 0.79; wordsim353: 0.69; ZG222: 0.42. https://dkpro.github.io/dkpro-similarity/wordpairsimilarity …

Same same, but different: Compositionality of Paraphrase Granularity Levels
D Benikova, T Zesch – … of the International Conference Recent Advances …, 2017 – acl-bg.org
… Finding Events We pre-annotate the first sen- tence of each pair with the Stanford Dependency Parser in the version provided by DKPro Core (Eckart de Castilho and Gurevych, 2014), using the RNN model with collapsed dependencies, as this parser worked best for our …

To GUI or not to GUI?
N Reiter, J Kuhn, M Willand – INFORMATIK 2017, 2017 – dl.gi.de
… steps). We employ the pipeline framework Apache UIMA for this purpose and develop drama-specific components as UIMA components. Existing NLP components (as in DKpro [EdCG14]) are reused as much as possible. New …

A Survey and Comparative Study of Arabic Diacritization Tools
O Hamed, T Zesch – jlcl.org
… In this section, we present our experimental setup: used data, baselines, diacritization tools, and evaluation metrics. The experiments were carried out using DKPro TC, the open-source UIMA-based framework for supervised text classification (Daxenberger et al., 2014) …

Classification of telicity using cross-linguistic annotation projection
A Friedrich, D Gateva – Proceedings of the 2017 Conference on …, 2017 – aclweb.org
… Features. We model each instance by means of a variety of syntactic-semantic features, using the toolkit provided by Friedrich et al. (2016).6 Pre- processing is done using Stanford CoreNLP (Chen and Manning, 2014) based on dkpro (Eckart de Castilho and Gurevych, 2014) …

Measurement of the quality of structured and unstructured data accumulating in the product life cycle in a data quality dashboard
S Chellathurai Saroja – 2017 – elib.uni-stuttgart.de
Page 1. Institute of Parallel and Distributed Systems Department of Applications of Parallel and Distributed Systems Universität Stuttgart IPVS Universitätsstraße 38 D-70569 Stuttgart Master Thesis Nr. 0990-0004 Measurement of the quality of structured and unstructured data …

GraphDocExplore: A Framework for the Experimental Comparison of Graph-based Document Exploration Techniques
T Falke, I Gurevych – Proceedings of the 2017 Conference on Empirical …, 2017 – aclweb.org
… The first is a file- based generator that simply reads a static graph for a document collection from a file. The second integrates DKPro Core (Eckart de Castilho and Gurevych, 2014) to create co-occurrence graphs of automatically recognized entities …

Utilizing Automatic Predicate-Argument Analysis for Concept Map Mining
T Falke, I Gurevych – IWCS 2017—12th International Conference on …, 2017 – aclweb.org
… For all experiments, we preprocessed the documents with components of DKPro Core (Eckart de Castilho and Gurevych, 2014), using tokenization, part-of-speech tagging and constituency parsing from the Stanford NLP tools and Snowball for stemming …

Distantly Supervised POS Tagging of Low-Resource Languages under Extreme Data Sparsity: The Case of Hittite
M Sukhareva, F Fuscagni, J Daxenberger… – Proceedings of the …, 2017 – aclweb.org
… Colons as well as paragraphs are aligned to German translations. The NLP pipeline for Hittite was built on the ba- sis of the uimaFIT library and, more specifically, DKPro Core libraries (Eckart de Castilho and Gurevych, 2014) …

An extensible multilingual open source lemmatizer
A Aker, J Petrak, F Sabbah – Proceedings of the International …, 2017 – eprints.whiterose.ac.uk
… 611223. References Daniel Bär, Torsten Zesch, and Iryna Gurevych. 2013. Dkpro similarity: An open source framework for text similarity. In ACL (Conference System Demon- strations). pages 121–126. Ann Bies, Justin Mott, Colin Warner, and Seth Kulick. 2012 …

Building an argument search engine for the web
H Wachsmuth, M Potthast, K Al Khatib, Y Ajjour… – Proceedings of the 4th …, 2017 – aclweb.org
… argument units. It is a de facto standard for natural language processing (Fer- rucci and Lally, 2004), and it also forms the basis of other text analysis frameworks, such as DKPro (Eckart de Castilho and Gurevych, 2014). UIMA …

Germeval 2017: Shared task on aspect-based sentiment in social media customer feedback
M Wojatzki, E Ruppert, S Holschneider… – Shared Task on …, 2017 – researchgate.net
… The CRF classifier uses the surface token without processing (lemmatization, standardization, lowercasing) and the POS tag. For tokenization and POS tagging, we use the DKPro tools (Eckart de Castilho and Gurevych, 2014) in the UIMA framework (Ferrucci and Lally, 2004) …

Financial decision-making process based on unstructured data sources and domain ontologies
MK Radzimski – 2017 – e-archivo.uc3m.es
Page 1. Universidad Carlos III de Madrid Repositorio institucional e-Archivo http://e-archivo.uc3m. es Tesis Tesis Doctorales 2017-06-12 Financial decision-making process based on unstructured data sources and domain ontologies Radzimski, Mateusz Karol …

Computational models for semantic textual similarity
A González Aguirre – 2017 – addi.ehu.es
… 26 ix Page 12. CONTENTS 2.7 Evaluation . . . . . 28 2.8 Best systems for Semantic Textual Similarity . . . . . 30 2.8.1 DKPro . . . . . 31 2.8.2 Takelab . . . . . 32 2.8.3 Ebiquity-Core …

Parsing argumentation structures in persuasive essays
C Stab, I Gurevych – Computational Linguistics, 2017 – MIT Press
Create a new account. Email. Returning user. Can’t sign in? Forgot your password? Enter your email address below and we will send you the reset instructions. Email. Cancel. If the address matches an existing account you will …

Collaborative Web-based Tools for Multi-layer Text Annotation
C Biemann, K Bontcheva, RE de Castilho… – Handbook of Linguistic …, 2017 – Springer
… entity annotation. 3.2.2 Back-End. The back-end of WebAnno was implemented using Java (Wicket [15], Spring Framework [43], DKPro Core [20]). Hibernate and JPA [1] are used for persisting objects in a MySQL database. We …

Beyond Similarity and Accuracy-A New Take on Automating Scientific Paper Recommendations
L Steinert – 2017 – researchgate.net
… mit dir reden konnte. Torsten Danke auch für deine Betreuung. Zu Anfang meiner Arbeit habe ich dich mit Emails zu DKPro überschüttet. Dennoch hast du jede Email geduldig und schnell beantwortet. Ich habe unsere Zusammenarbeit …

Improving hypernymy extraction with distributional semantic classes
A Panchenko, D Ustalov, S Faralli, SP Ponzetto… – arXiv preprint arXiv …, 2017 – arxiv.org
… 317–322, San Francisco, CA, USA. ACM. Meyer, CM, Mieskes, M., Stab, C., and Gurevych, I. (2014). DKPro Agreement: An Open-Source Java Li- brary for Measuring Inter-Rater Agreement. In Proceed- ings of COLING 2014 …

Common Crawled Web Corpora: Constructing corpora from large amounts of web data
KB Kristoffersen – 2017 – duo.uio.no
Page 1. Common Crawled web corpora Constructing corpora from large amounts of web data Kjetil Bugge Kristoffersen Thesis submitted for the degree of Master in Informatics: Programming and Networks (Language Technology group) 60 credits Department of Informatics …

Empty links and crossing lines: querying multi-layer annotation and alignment in parallel corpora
O ?ulo, S Hansen-Schirra, K Maksymski… – … and evaluation of …, 2017 – books.google.com
… The data structure is used within the CroCoAPI to communicate between the interface and the input- output (IO) level; it can, furthermore, be used as data connector to applications such as the lexical chainer embedded in DKPro (Gurevych et al. 2007, see be- low) …

Text similarity by using GloVe word vector representations
I SÁNCHEZ RODRÍGUEZ – 2017 – riunet.upv.es
… Thanks to the work of these groups, a lot of techniques, ideas, and even frameworks have emerged to help develop text similarity systems. One example of an open source framework for text similarity is DKPro Similarity [3]. In …

D 4.1-Gap Analysis of DARIAH Research Infrastructure
S Buddenbohm, R Barthauer – 2017 – hal.archives-ouvertes.fr
Page 1. HAL Id: hal-01663594 https://hal.archives-ouvertes.fr/hal-01663594 Submitted on 14 Dec 2017 HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not …

A Heuristic Approach of Text Summarization for Bengali Documentation
S Abujar, M Hasan, MSI Shahin… – … and Networking (8th …, 2017 – researchgate.net
Page 1. A Heuristic Approach of Text Summarization for Bengali Documentation Sheikh Abujar Dept. of CSE Jahangirnagar University Savar, Dhaka,Bangladesh sheikhabujar@gmail.com Mahmudul Hasan Dept. of CSE Comilla …

Designing annotation schemes: from model to representation
N Ide, C Chiarcos, M Stede, S Cassidy – Handbook of Linguistic …, 2017 – Springer
… of formats via mapping, and, as a result, transducers among formats are increasingly available that allow for the processing of annotated language resources by different tools and for different purposes (eg, ANC2Go [34], Pepper [50], and transducers available with DKPro 13 ) …

Argumentation mining in user-generated web discourse
I Habernal, I Gurevych – Computational Linguistics, 2017 – MIT Press
Create a new account. Email. Returning user. Can’t sign in? Forgot your password? Enter your email address below and we will send you the reset instructions. Email. Cancel. If the address matches an existing account you will …

Knowledge-based Supervision for Domain-adaptive Semantic Role Labeling
S Hartmann – 2017 – tuprints.ulb.tu-darmstadt.de
Page 1. Knowledge-based Supervision for Domain-adaptive Semantic Role Labeling Vom Fachbereich Informatik der Technischen Universität Darmstadt genehmigte Dissertation zur Erlangung des akademischen Grades Dr. rer. nat. vorgelegt von Silvana Hartmann, Dipl. Ling …

Post-Debugging in Large Scale Big Data Analytic Systems
EBS Edlich – GI-Edition – pdfs.semanticscholar.org
Page 66. Einreichung für: Big Data Management Systems in Business and Industrial Applications (BigBIA), Geplant als Veröffentlichung innerhalb der Lecture Notes in Informatics (LNI) Post-Debugging in Large Scale Big Data …

Learning Knowledge Graph Embeddings for Natural Language Processing
M Chen – 2017 – pdfs.semanticscholar.org
Page 1. Learning Knowledge Graph Embeddings for Natural Language Processing Muhao Chen Department of Computer Science University of California, Los Angeles Winter 2017 Page 2. Abstract Knowledge graph embeddings …

Controversy awareness on evidence-led discussions as guidance for students in wiki-based learning
S Heimbuch, D Bodemer – The Internet and Higher Education, 2017 – Elsevier
Skip to main content …

Controversy Analysis and Detection
S Dori-Hacohen – 2017 – scholarworks.umass.edu
Page 1. University of Massachusetts Amherst ScholarWorks@UMass Amherst Doctoral Dissertations Dissertations and Theses 2017 Controversy Analysis and Detection Shiri Dori-Hacohen Follow this and additional works at: http://scholarworks.umass.edu/dissertations_2 …

Analysing debates on climate change with textual entailment and ontologies
R Szabo, A Groza – … and Processing (ICCP), 2017 13th IEEE …, 2017 – ieeexplore.ieee.org
Page 1. Analysing debates on climate change with textual entailment and ontologies Roxana Szabo and Adrian Groza Intelligent System Group, Department of Computer Science, Technical University of Cluj-Napoca, Romania …

Interpretable neural models for natural language processing
T Lei – 2017 – dspace.mit.edu
Page 1. Interpretable Neural Models for Natural Language Processing by Tao Lei BS, Peking University (2010) MS, Massachusetts Institute of Technology (2013) OF TEQHNGY MAR 13 2017 LIBRARIES ARCHNES Submitted …

Core techniques of question answering systems over knowledge bases: a survey
D Diefenbach, V Lopez, K Singh, P Maret – Knowledge and Information …, 2017 – Springer
Page 1. Knowl Inf Syst DOI 10.1007/s10115-017-1100-y SURVEY PAPER Core techniques of question answering systems over knowledge bases: a survey Dennis Diefenbach1 · Vanessa Lopez2 · Kamal Singh1 · Pierre Maret1 …

1. Accelerating Linguistic Research through Advanced Digital Methods
J Odijk – dhbenelux2017.eu
Page 1. 1 Abstracts DHBenelux 2017 conference Posters 1. Accelerating Linguistic Research through Advanced Digital Methods Jan Odijk, Utrecht University In this poster + demo I demonstrate two web applications that have …

Requirements engineering artifact quality: definition and control
H Femmer – 2017 – mediatum.ub.tum.de
Page 1. Page 2. Page 3. Institut für Informatik der Technischen Universität München Requirements Engineering Artifact Quality: Definition and Control Henning Femmer Vollständiger Abdruck der von der Fakultät für Informatik …

From Event to Story Understanding
N Mostafazadeh – 2017 – search.proquest.com
From Event to Story Understanding. Abstract. Building systems that have natural language understanding capabilities has been one of the oldest and the most challenging pursuits in AI. In this thesis, we present our research …

(Visited 107 times, 1 visits today)