Generating domain specific sentiment lexicons using the Web Directory

based. There has been a demand for the construction of generated and labeled sentiment lexicon. Fordata on the social web (E.g., tweets), methods which make use of the synonymy relation don't work well,as we completely ignore the significance of terms belonging to specific domains. Here we propose togenerate a sentiment lexicon for any domain specified, using a twofold method. First we build sentimentscores using the micro-blogging data, and then we use these scores on the ontological structure providedby Open Directory Project [1], to build a custom sentiment lexicon for analyzing domain specific microbloggingdata.