A5033287456- Natural Language Processing Techniques
- Translation Studies and Practices
- Topic Modeling
- Semantic Web and Ontologies
- Linguistics and language evolution
- linguistics and terminology studies
- Speech and dialogue systems
- Linguistic Studies and Language Acquisition
- Genetics, Bioinformatics, and Biomedical Research
- Lexicography and Language Studies
- Mathematics, Computing, and Information Processing
- Literature, Language, and Rhetoric Studies
- Computational Drug Discovery Methods
- Library Science and Information Systems
- Historical and Linguistic Studies
- Immunodeficiency and Autoimmune Disorders
- Text and Document Classification Technologies
- Medieval Philosophy and Theology
- Historical Linguistics and Language Studies
- Text Readability and Simplification
- Medieval Literature and History
- Digital Humanities and Scholarship
- Web Data Mining and Analysis
- Geophysical Methods and Applications
- Building materials and conservation
Università Cattolica del Sacro Cuore
2018-2023
University of Milan
2013-2020
University of Pavia
2018
University of Milano-Bicocca
2015-2018
This paper describes the changes applied to original process used convert Index Thomisticus Treebank, a corpus including texts in Medieval Latin by Thomas Aquinas, into annotation style of Universal Dependencies. The are made both harmonise Dependencies version Treebank with two other available treebanks and fix errors inconsistencies resulting from process. details treatment different issues PoS tagging, lemmatisation assignment dependency relations. Finally, it assesses quality new...
This paper presents the structure of LiLa Knowledge Base, i.e. a collection multifarious linguistic resources for Latin described with same vocabulary knowledge description and interlinked according to principles so-called Linked Data paradigm. Following its highly lexically based nature, core Base consists large lemmas, serving as backbone achieve interoperability between resources, by linking all those entries in lexical tokens corpora that point lemma. After detailing architecture...
In this paper we are going to detail an unsupervised, graph-based approach for word sense discrimination on tweets. We deal with problem by constructing a graph of co-occurrences. By defining distance graph, obtain metric space, which can apply aggregative algorithm clustering. As result, will get clusters representing contexts that discriminate the possible senses term. present some experimental results both data set consisting tweets collected and task 14 at SemEval-2010.