Session recommendation system and method based on time information and star map network
A time information and recommendation method technology, applied in neural learning methods, biological neural network models, digital data information retrieval and other directions, can solve the problems of ignoring, discarding time information, dependence on large training data, etc., to reduce the complexity of the network structure, The effect of reducing computation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0063] Example 1:
[0064] In order to effectively utilize the time information of the user browsing items and capture the remote dependency information in the session, this embodiment proposes a session recommendation method based on time information and a star map network. method web framework such as figure 1 shown, first, for each unique item v in the session i Generate a d-dimensional embedding x through the embedding layer i ∈R d , each session sequence is input to a multi-layer star map network. In each layer of the network, the time information of the user browsing items is first used as the edge weight of the session graph, so that the gated graph neural network can aggregate more time-weighted node information. The longer it is, the more interested the user is in the current item, so the user's interest preference can be more accurately captured. At the same time, the star nodes in the star map network are used to obtain the remote dependency information in the ...
Example Embodiment
[0101] Embodiment 2:
[0102] A session recommendation system based on time information and a star map network, which can implement the session recommendation method based on time information and a star map network described in Embodiment 1, including:
[0103] Session sequence acquisition module: used to obtain the session sequence of the user browsing items, and obtain the initialization source node and star node;
[0104] Temporal feature-level information acquisition module: used to obtain the temporal feature-level information of the source node by using the time information of the user browsing items in the session sequence and the gated graph neural network;
[0105] Similarity calculation module: used to calculate the similarity of each source node and star node using attention;
[0106] Module: used to integrate the information of the source node and the star node based on the similarity of the source node and the star node, and obtain a new representation of the sou...
Example Embodiment
[0111] Embodiment three:
[0112] The embodiment of the present invention also provides a session recommendation device based on time information and a star map network, which can implement the session recommendation method based on time information and a star map network described in Embodiment 1, including a processor and a storage medium;
[0113] the storage medium is used for storing instructions;
[0114] The processor is configured to operate in accordance with the instructions to perform the steps of the following methods:
[0115] Obtain the session sequence of the user browsing items, and get the initialization source node and star node;
[0116] Obtain the temporal feature-level information of the source node by using the time information of the user browsing items in the session sequence and the gated graph neural network;
[0117] Calculate the similarity of each source node and star node with attention;
[0118] Integrate the information of the source node and...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap