How useful are similarity word lists for indirect anaphora resolution?
نویسندگان
چکیده
This work tests the use of word similarity lists for anaphora resolution in Portuguese and French corpora. We applied an automatic lexical acquisition technique over parsed texts to identify semantically similar words. We then made use of this lexical knowledge to resolve coreferential definite descriptions where the head noun of the anaphor is different from the head noun of its antecedent, which we call indirect anaphora. We compare the results in both languages with a baseline algorithm. Our proposed technique achieved an Fmeasure of 0.26 for Portuguese and 0.09 for French.
منابع مشابه
Using Word Similarity Lists For Resolving Indirect Anaphora
In this work we test the use of word similarity lists for anaphora resolution in Portuguese corpora. We applied an automatic lexical acquisition technique over parsed texts to identify semantically similar words. After that, we made use of this lexical knowledge to resolve coreferent definite descriptions where the head-noun of the anaphor is different from the head-noun of its antecedent, whic...
متن کاملWord Choice Problem and Anaphora Resolution
A dictionary-based method of detecting of implicit links between words in the texts (so-called indirect anaphora) is discussed. We suggest that the same dictionary and method can be used to partially solve the word choice problem in machine translation. The method consists in using of a dictionary of “scenarios” – lists of words semantically related to the given one, and show that detecting the...
متن کاملEnhanced Word Representations for Bridging Anaphora Resolution
Most current models of word representations (e.g., GloVe) have successfully captured finegrained semantics. However, semantic similarity exhibited in these word embeddings is not suitable for resolving bridging anaphora, which requires the knowledge of associative similarity (i.e., relatedness) instead of semantic similarity information between synonyms or hypernyms. We create word embeddings (...
متن کاملEstimating the Parameters for Linking Unstandardized References with the Matrix Comparator
This paper discusses recent research on methods for estimating configuration parameters for the Matrix Comparator used for linking unstandardized or heterogeneously standardized references. The matrix comparator computes the aggregate similarity between the tokens (words) in a pair of references. The two most critical parameters for the matrix comparator for obtaining the best linking results a...
متن کاملAn Algorithm to Calculate Semantic Distance among Indirect Question Referents
In question answering, anaphora resolution is an important part while extracting an answer in unstructured text. Indirect anaphora is defined as the referent-referee relation between two somehow related entities (e.g. car and garage). These referents of indirect referring expressions are unstated and also not mentioned previously in the text. Distance between these co-refers can be calculated w...
متن کامل