Method for determining Internet surfing behavior categories of network users

A network user and behavior technology, applied in the Internet field, can solve problems such as infeasibility, difficulty in class probability distribution, and insufficient accuracy

CN105871630AInactive Publication Date: 2016-08-17NAT COMP NETWORK & INFORMATION SECURITY MANAGEMENT CENT
6 Cites 3 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2016-08-17
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention provides a method for determining Internet surfing behavior categories of network users. The method comprises the steps that Internet surfing behavior features of all the network users to be detected are extracted, and a user behavior feature matrix X is formed through a quantitative method of a document vector space model; according to the user behavior feature matrix X, a behavior tendency set T and a user-tendency probability distribution matrix D are obtained through a probability latent semantic analysis PLSA and an EM algorithm; according to the user behavior feature matrix X, a feature word-category probability distribution matrix C is obtained through a support vector machine (SVM) algorithm; T*C runs through matrix multiplication to obtain a tendency-category mapping matrix M; D*M runs through matrix multiplication to obtain a user-category probability distribution matrix Y; according to the probability distribution situation of any network user to be detected in the categories, the network user to be detected is classified into the category with the maximum probability value.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of the Internet, in particular to a method for determining the online behavior category of network users. Background technique

[0002] A large number of cases show that the level of content security management can be effectively improved by using user behavior category information. However, based on the behavior data of web users, it is very difficult to accurately predict the class probability distribution of users by relying on existing techniques. On the one hand, the traditional classification method is to label the sample set, but because it is very difficult or even infeasible to label some users, the classifier based on it is often not accurate enough and the prediction effect is not ideal. On the other hand, traditional clustering methods can only divide users into one cluster, and a user may have multiple types of behavior tendencies. Therefore, traditional clustering-based classification methods ...

Examples

no. 1 example

[0051] The first embodiment of the present invention provides a method for determining the type of online behavior of a network user, such as figure 1 As shown, the method specifically includes the following steps:

[0052] Step S101: within a preset period of time, extract the online behavior characteristics of each network user to be tested, and form a user behavior characteristic matrix X through the quantitative method of the document vector space model according to the online behavior characteristics of all network users to be tested;

[0053] Specifically, the online behavior feature includes: a feature word marked based on the online behavior of the network user to be tested; the online behavior of the network user to be tested includes: the URL link clicked by the network user to be tested and the online search keywords.

[0054] According to the online behavior characteristics of all network users to be tested, the user behavior characteristic matrix X is formed by t...

no. 3 example

[0153] The third embodiment of the present invention provides a method for determining the type of online behavior of a network user, the method specifically includes the following steps:

[0154] Step S301: within a preset period of time, extract the online behavior characteristics of each network user to be tested, and form a user behavior characteristic matrix X through the quantitative method of the document vector space model according to the online behavior characteristics of all network users to be tested;

[0155]

[0156] Step S302: According to the user behavior characteristic matrix X, through the probabilistic latent semantic analysis method PLSA and EM algorithm, the behavior tendency set T and the "user-propensity" probability distribution matrix D are obtained;

[0157]

[0158] Each element vector in the behavior tendency set T represents each behavior tendency;

[0159]

[0160] Each row vector in the "user-propensity" probability distribution matrix ...