Fine-grained sentiment analysis of online reviews

A huge number of online reviews which are valuable voice of customer benefit consumers and product designers. However, with increasing growth of reviews and commodity attributes becoming more diverse, the traditional sentiment analysis of overall emotional tendency already can´t satisfy the needs of buyers and manufacturers. Consumers hope to get analysis results of target attributes and manufacturers hope to master strengths and weaknesses of products and also consumers´ taste. This study conducts a fine-grained sentiment analysis research to solve this problem and we increase generality of the application by the general methods we use to crawl reviews, produce general category of product attributes in women´ dress and find implicit features based on POS rules.