A Text Mining Approach For Lingerie Brand Of Online Word-Of-Mouth Analysis

• Main Authors: Pei-Shan Tsai, 蔡佩珊
• Other Authors: Chao-Chang Chiu
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/64321038417844073291

碩士 === 元智大學 === 資訊管理學系 === 104 === With the rapid development of the Internet to promote industrial development, people's daily lives rely on online shopping. A conventional way of shopping to rely on word of mouth to get people to evaluate. Today, the Internet social patterns, comment, and share experience on the web has become an important basis for later reference product. Global underwear market sales increased by nearly six per cent of female underwear every year about NT $ 25 billion market size, consumer population of nearly 600 million, an average annual per capita consumption of the number of members is 5.5. Female consumer trends increasingly younger. This study analyzes women when buying lingerie brand products eWOM reviews in Taiwan,use the web crawler lingerie brand comment in the forum, and then use text pre-processing, and heuristic N-Phrase Rules algorithm to find positive and negative comments about products. Finally, through a range of topics under the sentiment indicators, to observe the evaluation of the lingerie brand. Use visual correspondence analysis diagram to find all kinds of lingerie brand market segmentation and product strengths and weaknesses. This result can provide manufacturers improve product marketing and service mode, through extraction product advantages and disadvantages comment.