Chapter 4 Case Study in Tmall.com & Taobao.com
4.1 Feature of Text information E-Commerce
In this chapter I introduce my own research about how to evaluate customer review data mining and using rough set approach to calculate the decision rules about product. The product I choose is cell phone. The product I choose is Samsung galaxy note III (三星) from taobao.com and tmall.com. With the price rate of 3000-4000 Rmb and take a review from 20 different sellers. This phone itself right now is in so many in the Chinese and worldwide market. First of all I all I select a view seller of Samsung galaxy note III from tmall.com and taobao.com (B2C). After that from the tmall.com and taobao.com seller page, I take some customers review about Samsung galaxy note III. From customer’s review from tmall.com and taobao.com, 20 sellers pick up. The example of the customers review is like example above:
Ex4.1 show customers review from taobao.com
Ex 4.2 show customers review from tmall.com
4.2 Analysis Procedure
Customer’s review data from taobao.com and tmall.com is taken. From collected 20 seller we collect some customer’s review data mining. Separate the noun and adjective from customer review. One by one is divided in two table. After separate noun and adjective, I do some survey to 20 people to find the synonym or similarity. From the similarity, I can eliminate some of the criteria of 26 criteria to 18 category. It can continue to next step of rough set based mining.
Customer’s review Data Mining system smooth software so good system is not smooth pretty good screen delicate screen average screen battery durable battery average pretty good packaging good call quality
Expensive prize
good accessories full accessories accessories genuin...
... middle of paper ...
...omers review from taobao.com and tmall.com, I give marking to each category (table 4.1). After rating it one by one by used the algebra formula it produced table 4.4 and table 4.5.
In the second experiment, is using the 4emkA2 software by input the data from the experiment (table 4.1). The result is shown in Figure 4.4. The use of the software is to fins the decision result. After using the software is found there is 11 generated decision rules (Figure 4.5). Finally the final process of 4emka2 software it is found that Figure 4.6 is reclassification result for the customers review mining.
Future research is needed to compare the classification abilities of this method in various situations with other case-based classification methods is needed to see some other result how to evaluate the customers review by using DRSA method and 4emka2 software decision result.
Healthcare: Sentiment analysis has wide-scale applications in the Healthcare industry. Many patients use internet to post their patient experience in provider facilities. This unbiased feedback from patients is critical for healthcare practices to improve the quality of care. It is not possible for the patient to keep going back to the facility to report post intervention feedback. Extracting patient sentiments from unstructured data in blogs, twitter, Facebook posts help hospitals realize important performance factors like patient satisfaction, staff friendliness, procedure efficiency. Patients also share information regarding their payer experience on the internet. Tweets, posts about insurance benefits, timely service are critical information to the payers to improve their existing services. 94% patients believe hospital’s brand name is important in making a selection. By understanding patient sentiments and taking appropriate action to translate negative feedback into improved care can help a hospital improve its brand image.
Classification Text documents are arranged into groups of pre-labeled class. Learning schemes learn through training text documents and efficiency of these system is tested by using test text documents. Common algorithms include decision tree learning, naive Bayesian classification, nearest neighbor and neural network. This is called supervised learning.
Second, Best buy would renovate its exist stores in different branch based on the more catalogue of customer who living in(Best Buy, 2015). Therefor, some stores is especially designed for a group of customer like A, some became Customer B stores, C stores. For instance, a delicates service stores with exclusive salesman for IT geek like Customer A. If Customer B, a soccer fans buys a headphone, he or she only want to ease his or her demand to listen music instead of the brand. However, Best Buy not only focus on current profit, but also focus on customer group interaction- to analysis customer purchase behavior (Best Buy, 2015). Accordingly, Best Buy would effectively estimate the next purchase and product combination from customer based on this method. For example, normally, a customer will purchase DVD after she or he buys a DVD player but Best buy’s customer profitability analysis indicated that this kinds of customer also buys music accessories in second day after she or he buy DVD. Thus Best Buy would be able to send more relevant e-mail and special combination offer specific customer after customer purchaser behavior
A leading researcher at the University of Washington, Elizabeth Loftus, is specialized in the area of memory. She has recently discovered that when an occurrence is recalled it is not always re-created accurately. Loftus’ research revealed that instead, it is a reconstruction of the actual event. Newly collected information in relation to the topic being re-called can interfere with the memory you’re attempting to recall resulting in inaccurate recollection of the experience. If not be newly collected information it could be from other sources, such as the previous times you’ve told it, experiences from a television episode, a movie, or many other factors. You may have even experienced this yourself when you’ve been in the same place with another person for an event but have two un-matching stories of how the story took place and what occurred.
Business strategy and model: Zappos.com had a differentiation strategy with which they had differentiated themselves from the rest of the market. They had use a unique corporate culture in their company which was one of the major competitive edges of the company. According to the CEO of the company, Tony Hsieh, that everything that they had done at Zappos such as their relationships with 1,200 to 1,500 brands, policies and website style could be copied, however, the only thing that no one could copy from them was their unique culture. Zappos had 10 unique core values as a basis of their company’s culture, employee performance and their overall operations. They were hiring and firing people on the basis of their abilities that whether they were living up to these core values or not.
Base on the case of “Your Choice Furniture”, we marked this system's analysis to formulate solutions in this report; it assisted in evaluating the impact of recent change information technologies of “Your choice furniture” business system for evaluating how well the firm will be performing.
In order to further understand the customer data that has been accumulated through the purchases and transactions that have collected through the Nile online bookstore, three different regression models were completed. These regressions were completed in order to analyze the data in order to help the retailer better understand their customers, regarding which gender purchases more books, given the age of the customers or the day of the week the customer purchased a book. In order to help the retailer understand this information, customer data was gathered and used from over the last year in order to provide interpretations. The first step we week took to complete this process is to run the regressions, which resulted in three different models. We began by running the first regression. The first regression we ran contained only one variable, which was the gender variable. After running the first regression, we were able to successful complete
eBay should continue to focus on growth and continue to fight off competition from rival
Evaluation of Criteria 4. Purchase Decision 5. Post Purchase Evaluation Problem recognition is simply the awareness of a need. The need may be perceived or real. The problem recognition process occurs every time consumers decide they need something whether it is toilet paper or a new home.
In the past decades, e-commerce is growing rapidly and prosperously which has led a number of research studies aimed at understanding how customers use a variety of factors such as pre-purchase factors and post-purchase factors to judge the quality of a website involved in e-commerce activity (Collier and Bienstock, 2006). Product return is a widespread and expensive problem among the whole world. In the online context, product returns are important metric and critical component to online retailers (Yalabik et al., 2005, De et al. 2013, Hong and Pavlou 2014), as they indicate problems in web site content, customer service, consumers’ buying experience or fulfillment operations. Nowadays, the development and innovations of mobile
At least 5 years of the hotel data should be collected in a data warehouse or a data mart. All the old data should be entered into the system and based on ETL (Entry- Transform – Load) method should be loaded into the Data warehouse. If the hotel is not able to provide the historical data then the general data can collected by surveying potential customers, employees and hotel management staff. Mining the large amounts of transaction data allows each restaurant to improve its operations management and product
My thesis contain the identification of accurately classifying the sentiment in text from micro blogs. This addresses the problem by retrieving opinions, performing processing on the data and analyzing the data using machine learning techniques to classify them by sentiment as positive, negative or neutral. I proposed sentimental natural language processing method for processing the text and use various machine learning algorithms and feature selection methods to determine the best approach. The approaches towards sentiment analysis are machine learning based methods, lexicon based methods and linguistic analysis. I proposed sentimental natural language processing Model for processing text to remove irrelevant features that do not affect its orientation. Sentimental natural language processing model carries opinions in natural language process as well as unstructured reviews with pointers, punctuations, emotions, repeated words, symbols, WH questions, URL’s are preprocessed to extract relevant features while sanitizing inputs. Sentimental natural language processing measures the importance of feature...
It has been proven to be one of the most effective management initiatives because it can combine the knowledge, ideas, feelings, feedback from many parties together, especially between an organization and its customers. If this strategy is utilized successfully, it can bring many benefits to businesses. One of the benefits is the development of innovative products. For hospitality industry, two of the main ways to use this strategy is by customer surveying and talking with related parties such as customers, suppliers, business
Data Collection is the process of collecting information that will be utilized in the diagnostic process and eventually used to make business recommendation. In this data collection process, it is critical to ensure the highest quality of data possible. In the data collection component, the information is gathered on the specific department or organization such as inputs, design components, an...
This is the research study which is intended to help to people on the market economy and in the community. It shows the consumer’s behavior that affects their decision making on choosing a right product. And this study how also shows this method can be useful when engaged on the products carefully.