"Estudo Comparativo de Formas Lexicais e Estruturas Passivas em Português de Angola e Europeu"

No próximo dia 19 de Abril de 2022, pelas 11h00 na sala S2 do DCC (FC6 1.40),  Timóteo Sumbula Muhongo irá dar uma palestra intitulada "Estudo Comparativo de Formas Lexicais e Estruturas Passivas em Português de Angola e Europeu".


A palestra é organizada pelo DCC-FCUP e pelo grupo de investigação LIAAD-INESCTEC e é aberta a todos os interessados.


Short Bio:


Timóteo Sumbula Muhongo is a PhD student in Human Language Technologies at the University of Porto, a scholarship holder of the Calouste Gulbenkian Foundation and a Master in Terminology and Specialty Information Management at the NOVA University Lisbon. He is a researcher at the Laboratory of Artificial Intelligence and Decision Support (LIAAD) of the institute INESC TEC and at the Linguistics Centre of Porto University (CLUP). He is also a member of the American Organization of Teachers of Portuguese (AOTP) and the Portuguese Association of Linguistics (APL). Research areas: Text Mining, Natural Language Processing, Discourse Analysis, Lexicology and Syntax.



"Estudo Comparativo de Formas Lexicais e Estruturas Passivas em Português de Angola e Europeu"





The study has two aims.  The objective of the first one to identify loanwords originating from Angolan languages and categorize them by lexical class (verbs, nouns, adjectives, and proper nouns). As for computational methods, we have used a set of methods, including part-of-speech tagging, named entity recognition, dependency parsing and regular expressions in R programming language and some specific packages oriented towards Text Mining and NLP. This analysis revealed that a significant proportion of loanwords does not actually appear in existing reference dictionaries. We present a prototype dictionary of Angolan regionalisms accompanied by examples of cotexts, which can be useful for follow-up studies. Our second aim was to analyse the passive structures in Angolan and European Portuguese with the focus on eventive, stative and resultative structures. These were extracted automatically from the available corpora. We have analysed the proportions of passive structures and phrases in Angolan and European Portuguese. Our main finding was that the passive structures were much more common in Angolan Portuguese than in the European one, particularly when considering the oral corpora.


Também lhe pode interessar


"A quantitative understanding of pattern matching"

Prof. Sandra Alves dá palestra no DCC


"Online learning: a quick tour of bandit algorithms"

Sébastien Gerchinovitz dá palestra no DCC


Cameleer: a Deductive Verification Tool for OCaml

Mário Pereira dá palestra no DCC