A5028196038- Natural Language Processing Techniques
- Topic Modeling
- Semantic Web and Ontologies
- Advanced Database Systems and Queries
- Cloud Computing and Resource Management
- Basque language and culture studies
- Speech and dialogue systems
- Text Readability and Simplification
- Algorithms and Data Compression
- Neural Networks and Applications
- Biomedical Text Mining and Ontologies
- Spanish Linguistics and Language Studies
- Music and Audio Processing
- Data Management and Algorithms
- Time Series Analysis and Forecasting
University of the Basque Country
2014-2017
Abstract Computational power needs have greatly increased during the last years, and this is also case in Natural Language Processing (NLP) area, where thousands of documents must be processed, i.e., linguistically analyzed, a reasonable time frame. These computing implied radical change architectures big-scale text processing techniques used NLP. In paper, we present scalable architecture for distributed language processing. The uses Storm to combine diverse NLP modules into chain, which...
Sequence-to-sequence neural architectures are the state of art for addressing task correcting grammatical errors. However, large training datasets required this task. This paper studies use sequence-to-sequence models correction errors in Basque. As there is no data language, we have developed a rule-based method to generate grammatically incorrect sentences from collection correct extracted corpus 500,000 news We built different according strategies combine synthetic examples. From these...
This work describes the formalization of a word structure grammar that represents complex morphological and morphosyntactic information embedded within forms an agglutinative language (Basque), giving comprehensive linguistic description main phenomena, such as affixation, derivation, composition, also taking into account modeling both standard non-standard words. We have identified relevant issues to be addressed in representation grammar. present development Morfeus+, tool for analysis...