LEXICON BASED EXTRACTION AND OPINION CLASSIFICATION OF ASSOCIATIONS IN TEXT FROM HINDI WEBLOGS

Opinion Mining is a field of Computer Science that is dedicated towards deriving the emotion, attitude and opinion of the author of a content towards an entity, aspect, or a relationship of associated entities. The content may be a text, an image or a multimedia. The backbone of an Opinion Mining System is the classification model that determines the author's sentiment from a digital content that is developed by the author. This paper presents the lexical resources developed, algorithms formalised, experiments conducted and outcomes of those experiments for an ongoing research in the field. In this research an opinion mining system is being developed for Hindi text appearing on Weblogs in the domain of home remedies. The need, objective and scope of this application are also discussed in this paper. From a computing perspective for linguistic processing purposes there is a need to develop more datasets, lexical resources, and computing tools to facilitate the researchers who wish to develop applications for unstructured text in Hindi. As part of this research, domain specific lexical resources and datasets are being developed in Hindi. These resources are expected to be useful to the current and future researchers. During these unprecedented times of Covid-19, there is a need to bring web contents in Hindi related to home remedies under the umbrella of an Opinion Mining System.