Paper Details

Opinion mining also called as sentiment analysis, has enticed a great deal of attention recently due tomany practical applications and challenging research problems. The important and challenging task in opinion miningis to mine opinion targets and opinion words from online reviews. The key component of mining opinion targets andopinion words is to detecting the relations among the targets and words. To extract opinion targets, opinion words andidentifying the relations between them as an alignment process partially-supervised word alignment model (PSWAM) isused. Then, a graph-based algorithm is used to estimate the confidence of each candidate and the candidates withhigher confidence will be extracted as the opinion targets or opinion words. This model captures opinion relations moreprecisely, especially for long span relations as compared to previous methods based on the nearest-neighbor rules.When dealing with informal online texts, the word alignment model effectively solve the problem of parsing errors.Because of the usage of partial supervision the proposed model obtained better result as compared to unsupervisedalignment model. To decrease the probability of error generation graph-based co- ranking algorithm is used whenestimating candidate confidence. Sentiment analysis is used to get positive negative and neutral reviews. Themanufactures can get the feedback from product reviews to improve the quality of their products in a timely fashion.

Ms. Sunita Patil, Mrs. N. S. Patil, “Mining Opinion Targets and Opinion Words Using Word Alignment Model with Positive, Negative Reviews”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 4, Issue 6, pp. 881-892, June 2017.