Effect of Pronoun Resolution on Document Similarity
نویسندگان
چکیده
This paper presents a novel effect of Pronoun Resolution on measurement of document similarity. In this paper we have studied the effect of pronoun resolution within the framework of the Vector Space Model and Probabilistic Latent Semantic Analysis. For this purpose we have developed a Benchmark Corpus consisting of documents whose similarity scores have been given by human beings. We measured the inter-document similarity on these documents using VSM and PLSA. We then performed pronoun resolution on these documents and again calculated the similarity using both methods. Next, the correlation coefficient of the scores was taken with those of the human generated scores. The correlation coefficients clearly demonstrated substantial and consistent improvements of the similarity score after pronoun resolution.
منابع مشابه
A Deep Neural Network for Chinese Zero Pronoun Resolution
This paper investigates the problem of Chinese zero pronoun resolution. Most existing approaches are based on machine learning algorithms, using hand-crafted features, which is labor-intensive. Moreover, semantic information that is essential in the resolution of noun phrases has not been addressed enough by previous approaches on zero pronoun resolution. This is because that zero pronouns have...
متن کاملPronoun Resolution and Summary Extraction From English Documents
This paper presents an approach to generate a precise and meaningful summary of a Document in English Language. Here, we adopt a modified Sidner’s Focusing algorithm to perform pronoun resolution. We devise an algorithm to divide any sort of compound and complex sentences into simple sentences. Anaphora resolution is performed on these simple sentences. Then we find the lexical cohesion between...
متن کاملThe Role of Gender Information in Pronoun Resolution: Evidence from Chinese
Although previous studies have consistently demonstrated that gender information is used to resolve pronouns, the mechanisms underlying the use of gender information continue to be controversial. The present study used event-related potentials (ERPs) to investigate whether working memory modulates the effect of gender information on pronoun resolution. The critical pronoun agreed or disagreed w...
متن کاملA Document-Level SMT System with Integrated Pronoun Prediction
This paper describes one of Uppsala University’s submissions to the pronounfocused machine translation (MT) shared task at DiscoMT 2015. The system is based on phrase-based statistical MT implemented with the document-level decoder Docent. It includes a neural network for pronoun prediction trained with latent anaphora resolution. At translation time, coreference information is obtained from th...
متن کاملDisambiguation of the Neuter Pronoun and Its Effect on Pronominal Coreference Resolution
Coreference resolution, determining the appropriate discourse referent for an anaphoric expression, is an essential but difficult task in natural language processing. It has been observed that an important source of errors in machine-learning based approaches to this task, is the wrong disambiguation of the third person singular neuter pronoun as either referential or non-referential. In this p...
متن کامل