George Manias1, Maria Angeles Sanguino2, Sergio Salmeron2, Argyro Mavrogiorgou1, Athanasios Kiourtis1, Dimosthenis Kyriazis1, 1University of Piraeus, Greece, 2ATOS Research and Innovation, Spain
The tremendous growth and usage of social media in modern societies have led to the production of an enormous real-time volume of social texts and posts, including tweets, that are being produced by users. These collections of social data can be potentially useful, but the extent of meaningful data in these collections is still of high research and business interest. One of the main elements in several application domains, such as policy making, addresses the scope of public opinion analysis. The latter is recently realized through sentiment analysis and Natural Language Processing (NLP), for identifying and extracting subjective information from raw texts. An additional challenge refers to the exploitation and correlation of the sentiment that can be derived for different entities into the same text or even a sentence to analyze the different sentiments that can be expressed for specific products, services, and topics by considering all available information that can be depicted within a text in a holistic way. Hence, this paper evaluates the utilization of an Entity-Level Sentiment Analysis (ELSA) approach on Twitter Data. The latter seeks to enhance the knowledge derived from tweets with the ultimate objective the overall enhancement of the policy making procedures of modern organizations and businesses.
Twitter Sentiment Analysis, Entity-Level Sentiment Analysis, Named Entity Recognition, Policy Making
