Chapter 4 Case Study in Tmall.com and Taobao.com 4.1 Features of Text Information E-commerce In this chapter I present my research on how to evaluate customer review data mining and use an approximate approach to calculating product decision rules. The product I choose is the cell phone. The product I choose is Samsung Galaxy Note III (三星) from taobao.com and tmall.com. With the price of 3000-4000 Rmb and get a review from 20 different sellers. This phone itself right now is present in so many Chinese and world markets. First of all I select a Samsung Galaxy Note III seller from tmall.com and taobao.com (B2C). Next, from the seller page tmall.com and taobao.com, I take some customers' reviews about Samsung Galaxy Note III. From customer review from tmall.com and taobao.com, 20 sellers collect. The customer review example is like the example above: Ex4.1 shows customer review from taobao.comEx 4.2 shows customer review from tmall.com4.2 Analysis Procedure Customer review data is taken from taobao.com and tmall.com. From the 20 sellers collected we collect the data mining of some customer reviews. Separate the noun and adjective from the customer review. One by one it is divided into two tables. After separating the noun and adjective, I poll 20 people to find the synonym or similarity. From the similarity, I can eliminate some of the criteria from 26 criteria to 18 categories. It can continue with the next stage of rough set based mining. Customer Review Smooth software mining system, so a good system is not smooth good enough screen delicate screen average screen durable battery average battery quite good packaging good call quality Expensive premiumgood accessories complete accessories original accessories... ....half of sheet ......review by omer from taobao.com and tmall.com, I give a rating to each category (table 4.1). After evaluating it one by one using the algebraic formula, it produced table 4.4 and table 4.5. In the second experiment, use the 4emkA2 software by entering the experiment data (table 4.1). The result is shown in Figure 4.4. The use of the software is used to define the outcome of the decision. After using the software, 11 decision rules are generated (Figure 4.5). Finally, in the final process of 4emka2 software it was discovered that Figure 4.6 is the result of reclassification for customer review mining. Future research is needed to compare the classification capabilities of this method in various situations with other case-based classification methods to see other outcome such as rating customer review using DRSA method and decision outcome of 4emka software2.