A Relative Study of factors and approaches for Hindi Anaphora Resolution

A literature survey on theoretical approaches of anaphora resolution system and various factors among which these approaches are based is presented in this paper. From last two decades, the significant amount of work on Natural language processing application such as anaphora resolution has been gradually increased but with respect to Indian language, especially Hindi, there is a long way to go. The success rate of anaphora resolution system also improves the performance of machine translation system, automatic text summarization and question answering system, etc. Many researchers implemented different factors suitable to their approach for anaphora resolution among which are gender agreement, number agreement, semantic analysis, salience measurement, word sense disambiguation, etc. The paper reviewed the factors useful for resolving anaphora and incorporation these factors for building the strategies for resolution and computational behavioral of anaphora resolution approaches. The authors finally put forward the lessons learnt from it, which may prove to be important knowledge sources for different information retrieval tasks.