| Comparison shopping is a new kind of electronic shopping patterns as the development of the e-commerce. With the development of businesses, more and more historical data in shopping website has been accumulated. At the same time the business competition between the same industry enterprises becomes more and more fierce. The decision-making department has to collect valuable information from the massive data to analysis the user behavior so as to find better sales model and to propose a sales plan which can guide the users' buying habits, and then promote the business development of enterprises. In addition each website has a set of separate report system, which makes data scattered, information isolated, and resources redundant. Therefore an integrated, consistent data sources is established to integrate all data sources and to supply an enterprise data platform for effective analysis.Firstly, the current analytics situation for comparison shopping is introduced and then the processing procedure of creating data warehouse and some main issues are discussed. ETL system design for web application system is discussed including the issues of how to extract and translate and load from the unstructured or semi-structured data to structured data. Then the model design of data warehouse is discussed, the theme domain is confirmed, the discussion of dimension's granularity and performance optimization is made, as well as the process of establishing the data warehouse platform is discussed. Finally, it's necessary to prospect the outlook for applications of data warehouse on analyzing the user's behavior in comparison shopping. |
PDF Full Text Request