Provides the different marketing strategy

The users’ log data of visiting website are carried on thorough mining utilizing clustering technology. Then divide customers who have the similar interests into different community and construct different interest models according to each customer community’s different characters, then provide difference marketing strategy.

Reduce the customers’ development cost

Under the electronic commerce environment, the customer mainly may divide into three kinds: The first kind is the low value or the insignificant customer; The second kind is the valuable customer which is cannot easily ring off; The third Class is unceasingly seeking more preferential price in the network and the better service, this kind of customer is the potential valuable customer. The statistical data indicated that developing a new customer’s cost is fifth than retaining an old customer’s cost, but losing an old customer’s price is equal to develop 10 new customers. Through the web usage mining, analyze customer’s behavior; it may help the enterprise to understand the customers’ behavior at the period of turning to the competitors and analyze the main reason which the customers drains. Thus takes the essential measure in view of the existing customer. Moreover, through keeping mining the behavior that customers settle in the website, we can find the customers’ interest, thus change the design of the website’s overall layout to attract the customers.

Shoot for ceiling of using existing customers

Through the web usage mInIng, enterprise may discover the influence factor of customers’ purchase from customers purchase behavior. Then forecast model will be established in which user could be interested in which type commodity is in the future and will be interested in which correlative commodity, so analyze the possibility of purchase some kind of product. After mastering some valuable information when the company promotes the new product in the future, it can achieve with a clear goal and strive for the biggest profit.

Make full use of retrieval function module of website Nearly all electronic commerce websites have the retrieval function, its goal is that let customers find the product or the correlative service ofthe company in the shortest time. Using the web usage mining technology, it can make the effective use of website retrieval function from the following two stratifications: CD Carry on the effective layout to the correlative products. The commodity retrieval module generally can use the precise retrieval technology and the fuzzy retrieval technology. Precise retrieval technology allows users to input many precise key words and search products or services which the users most need. Now a thing the company needs to do is that display some related products or service on the side of the target products or services. But if the user uses fuzzy retrieval (this one is also the most usually used), then it is possible to search several kinds of products or the service, now the company will carry on layout again using associate orders or analysis technology ofthe path and according to the user’s characteristic and other users’ appraise. ® Record the key words of “strangeness”. Each company cannot be all-embracing and have everything expected to find.

When users use the key words which oneself inputs and search, the result the website provides possibly is the empty record? At this time the website frrst needs some extra words to attract these customers; Next record these key words “strangeness” to the key words database; After that analyze these key words as soon as quickly, obtain the frequency of these key words appear, and these key words corresponding products can get how profits are; Finally decide whether needs to purchase this kind of commodity, if it is possible to inform these customers.