ANALYSIS OF SOCIAL RELATIONSHIPS ON TWITTER DATA FOR ONLINE ANALYTICAL PROCESSING
Sana, Ch. B. N. Lakshmi · Journal of Emerging Technologies and Innovative Research · 2020
Social media platforms, for instance as twitter has become increasingly popular as an emerging platform for messaging and communication. On the other side, online meticulous dealing with multidimensional sorted out data. The purpose of using Hierarchical topic modeling i.e., THLDA to mine the dimension hierarchy of tweets. It can be applied for text OLAP on the tweets. And it uses word2vec to analyze the semantic relationships of words in tweets to obtain more effective dimensions. To improve the model effectiveness the bicliques to calculate the semantic impact of the topic of two tweets. Focusing on how the social impact factors and word semantic similarity influence the experimental results separately. In this we are considering how social relationships impact on the hierarchical topic model. In social relationships we are focusing direct and indirect relationships to follow the unrelated tweeters. We conduct extensive experiments on real twitter data to evaluate to effectiveness of THLDA. We are using hash tags to improve the models.