A Survey of Sentiment Analysis for Journal Citation

摘要

Sentiment analysis approach belongs to the family of machine learning, where the objective is to discover useful patterns stored in a database. Due to wide availability of data, there is an upcoming need for turning such an overwhelming amount of data into useful knowledge. In this paper we recommend different techniques available for high accuracy extraction of citations for academic papers and improve the performance in citation extraction by integration of two techniques. Therefore our aim is to automate the task to determine whether a context is positive or negative. The main goal of sentiment analysis lies in finding the polarity of citation expressed in different research article. In this paper we address the techniques, approaches and methods of the research which are supportive and marked as the essential field of sentiment analysis of citations in research article. This literature survey is done to study the sentiment analysis problem in journal citation to identify different trends and recommend the upcoming research directions.