Alibaba Group Holding Limited

Ningjun Su

Haijie Gu

Zhixiong Yang

Jinyin Zhang

Qifeng Dai

Haiping Ma

Enhong Chen

Zhonghua Deng

The present disclosure discloses a method and a system of displaying cross-website information. The method includes obtaining, by a first server of a first website, product information under a category of at least one desired product of a first user. Next, the first server obtains information of similar product units under a product category of a second server, where a similar product unit includes a set of displayable products. The first server then computes degrees of similarity between the product information under the category of the at least one desired product and respective product information of the similar product units. For the product information under the category of the at least one desired product, the first server selects a predetermined number of top similar product units in a descending order of the degrees of similarity, and renders products included in the predetermined number of top similar product units as products to be displayed. The first server displays the products on a web page of the first website that is viewed by the first user.

Method and system of displaying cross-website information

Data processing includes accessing a product category tree, the product category tree comprising a plurality of hierarchical levels. A leaf category level of the plurality of hierarchical levels comprises a leaf category node. The leaf category node includes product information. The product information comprises a plurality of product attribute parameters. The plurality of product attribute parameters comprises standard product information. Data processing further includes selecting, among the plurality of product attribute parameters that correspond to the leaf category node, a representative product attribute parameter that is representative of the product information; and partitioning standard product information of the leaf category node to obtain a plurality of sets using the representative product attribute parameter, wherein each of the plurality of sets includes at least some of the standard product information.

Processing of categorized product information

Classifying a user into a user group is described, including: obtaining action record information associated with a user with respect to a plurality of information categories; determining a preference value for at least one of the plurality of information categories based at least in part on the action record information; determining a demand preference dispersion value for the user based at least in part on the preference value, wherein the demand preference dispersion value represents a degree to which the user has dispersed interest across the plurality of information categories; and determining whether the user is to be associated with a demand-focused user group based at least in part on the demand preference dispersion value.

Providing information recommendations based on determined user groups

Embodiments of the present application relate to a personalized information pushing method, a personalized information pushing device, and a computer program product for pushing personalized information. A personalized information pushing method is provided. The method includes retrieving network behavior data related to access operations performed by user terminals, determining a numerical value of a degree of correlation between a user terminal and a plurality of information providing terminals having a correlation within a set time window based on the network behavior data, retrieving information providing terminals corresponding to a first user terminal to form a first data set based on the numerical values, retrieving information of the information providing terminals from the first data set to generate a first information, and pushing the first information to the first user terminal.

Growing community of inventors

Hangzhou, China

Ningjun Su