UFRGS: Identifying Categories and Targets in Customer Reviews
نویسندگان
چکیده
This paper reports on our participation in SemEval-2015 Task 12, which was devoted to Aspect-Based Sentiment Analysis. Participants were required to identify the category (entity and attribute), the opinion target, and the polarity of customer reviews. The system we built relies on classification algorithms to identify aspect categories and on a set of rules to identify the opinion target. We propose a two-phase classification approach for category identification and use a simple method for polarity detection. Our results outperform the baseline in many cases, which means our system could be used as an alternative for aspect classification.
منابع مشابه
Identifying Customer Journey Opportunities in 5A Model in Tourism Industry
Purpose: Growing development of technologies helped stronger customers with better relationship with companies. Consequently, marketers should pursue new ways of attracting customers and pathway. Modern customer buying path in the age of communication has been redesigned as 5A model (Aware, Appeal, Ask, Act, Advocate). The purpose is to identify customer opportunities of 5A model in tourism ind...
متن کاملIdentifying Rangeland Development and Restoration Operations Targets (Case Study: Hossein Abad Bazoft Basin, Chaharmahal and Bakhtiari Province, Iran)
The purpose of this study was identifying rangeland development andrestoration operations targets. This study was conducted in Hossein Abad Bazoft basinwith 7162 ha area, located in south east of Farsan County, Chaharmahal and BakhtiariProvince, Iran. The basin has various land uses including rangeland, forest, irrigated farmand garden, dry farm, residential area and rock mass. In this study, d...
متن کاملThe Perception of Usefulness: Iranian Customers’ Evaluation of Customer Reviews
Over the last decade, the retail industry has had a phenomenal growth. All figures show their success and efficiency and many studies have shown the role of customer reviews in encouraging ambivalent purchasers to buy items online. There have been numerous studies on why people read and trust these comments and taking for granted the important role of customer reviews in determining buying deci...
متن کاملMulticlass Classifier Building with Amazon Data to Classify Customer Reviews into Product Categories
As one of the world’s largest online retailers, Amazon.com has a sales volume of tens of billions dollars and as a result huge amounts of customer reviews accumulate. These reviews are great learning samples which could help promote commercial prediction and advertisement. Our project aims to build a multiclass classifier based on the customer reviews which are labeled by product categories in ...
متن کاملIdentifying product features from customer reviews using hybrid patterns
In this paper we have addressed the problem of automatic identification of product features from customer reviews. Costumers, retailors, and manufacturers are popularly using customer reviews on websites for product reputation and sales forecasting. Opinion mining application have been potentially employed to summarize the huge collectionof customer reviews for decision making. In this paper we...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2015