KLUE: Simple and robust methods for polarity classification
- Thomas Proisl, P. Greiner, S. Evert, Besim Kabashi
- Computer ScienceInternational Workshop on Semantic Evaluation
- 1 June 2013
This paper uses simple bag-of-words models, a freely available sentiment dictionary automatically extended with distributionally similar terms, as well as lists of emoticons and internet slang abbreviations in conjunction with fast and robust machine learning algorithms to solve the SemEval-2013 sentiment analysis task.
SentiKLUE: Updating a Polarity Classifier in 48 Hours
- S. Evert, Thomas Proisl, P. Greiner, Besim Kabashi
- PhysicsInternational Workshop on Semantic Evaluation
- 1 August 2014
SentiKLUE is an update of the KLUE polarity classifier – which achieved good and robust results in SemEval-2013 with a simple feature set – implemented in 48 hours.
SemantiKLUE: Robust Semantic Similarity at Multiple Levels Using Maximum Weight Matching
- Thomas Proisl, S. Evert, P. Greiner, Besim Kabashi
- Computer ScienceInternational Workshop on Semantic Evaluation
- 1 August 2014
The SemantiKLUE system is a word-to-word alignment of two texts using a maximum weight matching algorithm that combines unsupervised and supervised techniques into a robust system for measuring semantic similarity.
Using High-Quality Resources in NLP: The Valency Dictionary of English as a Resource for Left-Associative Grammars
- Thomas Proisl, Besim Kabashi
- Linguistics, Computer ScienceInternational Conference on Language Resources…
- 1 May 2010
The Valency Dictionary of English can be regarded as being well suited for NLP purposes by being used for accurately parsing natural language with a rule-based approach and by integrating it into a Left-Associative Grammar.
Results of the Translation Inference Across Dictionaries 2019 Shared Task
- J. Gracia, Besim Kabashi, Ilan Kernerman, Marta Lanau-Coronas, Dorielle Lonke
- Linguistics, Computer ScienceTIAD@LDK
- 2019
An overall description of the Translation Inference Across Dictionary shared task, the evaluation data and methodology, and the systems’ results are given.
JSLIM - Computational Morphology in the Framework of the SLIM Theory of Language
- J. Handl, Besim Kabashi, Thomas Proisl, Carsten Weber
- Computer Science, LinguisticsInternational Workshop on Systems and Frameworks…
- 4 September 2009
How the system works, the evolution from previous versions, and how the rules for word form recognition can be used also forword form generation are shown, and the subject of the reversibility of grammar rules is broached with the aim of an automatic word form production without any additional rule system.
A Proposal for a Part-of-Speech Tagset for the Albanian Language
- Besim Kabashi, Thomas Proisl
- LinguisticsInternational Conference on Language Resources…
- 1 May 2016
The Albanian language has some properties that pose difficulties for the creation of a part-of-speech tagset that can adequately represent the underlying linguistic phenomena, and this paper presents a proposal for that tagset.
Albanian Part-of-Speech Tagging: Gold Standard and Evaluation
- Besim Kabashi, Thomas Proisl
- Computer ScienceInternational Conference on Language Resources…
- 1 May 2018
This paper provides mappings from the full tagset to both the original Google Universal Part-of-Speech Tags and the variant used in the Universal Dependencies project and achieves accuracies of up to 95.10%.
Proceedings of TIAD-2019 Shared Task - Translation Inference Across Dictionaries co-located with the 2nd Language, Data and Knowledge Conference (LDK 2019), Leipzig, Germany, May 20, 2019
- J. Gracia, Besim Kabashi, Ilan Kernerman
- Computer Science, LinguisticsTIAD@LDK
- 2019
The main aim of TIAD is to support a coherent experiment framework that enables reliable validation of results and solid comparison of the processes used, to enhance further research on the topic of inferring translations across languages.
EmotiKLUE at IEST 2018: Topic-Informed Classification of Implicit Emotions
- Thomas Proisl, Philipp Heinrich, Besim Kabashi, S. Evert
- Computer ScienceWASSA@EMNLP
- 1 October 2018
EmotiKLUE is a deep learning system that combines independent representations of the left and right contexts of the emotion word with the topic distribution of an LDA topic model that achieves a macro average F₁score of 67.13%, significantly outperforming the baseline produced by a simple ML classifier.
...
...