期刊名称：Journal of Automation, Mobile Robotics & Intelligent Systems (JAMRIS)
印刷版ISSN：1897-8649
电子版ISSN：2080-2145
出版年度：2018
卷号：12
期号：4
DOI：10.14313/JAMRIS_4-2018/28
出版社：Industrial Research Inst. for Automation and Measurements, Warsaw
摘要：Nowadays, Customer's product reviews can be widely found on the Web, be it in personal blogs, forums, or e- commerce websites. They contain important products' in- forma.on and therefore became a new data source for compe..ve intelligence. .n that account, these reviews need to be analyzed and summarized in order to help the leader of an en.ty (company, brand, etc.) to make ap- propriate decisions in an e.ec.ve way. .owever, most previous review summariza.on studies focus on summa- rizing sen.ment distribu.on toward di.erent product fe- atures without taking into account that the real advan- tages and disadvantages of a product clarify over .me. For this reason, in this work we aim to propose a new sy- stem for product opinion summariza.on which depends on the .me when reviews are e.pressed and that covers the sen.ments change about product features. The pro- posed system firstly, generates a summary based on pro- duct features in order to give more accurate and e.cient informa.on about di.erent features. .econdly, classify the product based on its features in its appropriate class (good, medium or bad product) using a fuzzy logic sy- stem. The e.perimental results demonstrate the e.ec.- veness of the proposed system to generate the real image of a product and its features in reviews.
