Papers by Marco A. Sobrevilla Cabezudo
WordNet-Shp: Towards the Building of a Lexical Database for a Peruvian Minority Language (L18-1)
Copied to clipboard
| Challenge: | WordNet-like resources are lexical databases with highly relevance information and data that could be exploited in more complex computational linguistics research and applications. |
| Approach: | They propose to build a WordNet database for a low-resourced and indigenous language in Peru . they propose to use word2vec similarity to compare definition glosses in a dictionary with the content of a Spanish WordNet . |
| Outcome: | The proposed database is based on a bilingual dictionary written in Spanish and an automatic evaluation process using a manually annotated Gold Standard in Shipibo-Koniba. |
Corpus Building and Evaluation of Aspect-based Opinion Summaries from Tweets in Spanish (L18-1)
Copied to clipboard
Daniel Peñaloza, Rodrigo López, Juanjosé Tenorio, Héctor Gómez, Arturo Oncevay-Marcos, Marco A. Sobrevilla Cabezudo
| Challenge: | a corpus of Spanish extractive and abstractive summaries of opinions is presented . the goal is to analyze the summary content and to show how different they are written . |
| Approach: | They present a corpus of Spanish extractive and abstractive summaries of opinions . they analyze the summary agreement between them and their aspect coverage and sentiment orientation . |
| Outcome: | The presented corpus of Spanish extractive and abstractive summaries is a reference for academic research. |