Recommendation Embedding Update Using Natural Language Semantics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional recommendation models struggle to model semantic information and utilize external semantic knowledge, leading to poor recommendation precision due to reliance solely on collaborative information.
Innovation Solution
A data processing method that combines collaborative information with semantic information by using an embedding network and natural language processing model to generate and update feature representations, incorporating natural language text to enhance modeling capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a conventional recommendation model uses only collaborative information from a data set, then the model architecture remains simple and online inference latency is short, but the model cannot model semantic information and recommendation precision is poor
Solution Approach 1:
The patent merges collaborative filtering information with semantic information from a pre-trained language model into a unified recommendation model. The embedding network integrates both types of information through shared layers and combined loss functions, allowing the model to leverage both collaborative patterns and semantic understanding simultaneously, thereby improving recommendation precision while managing complexity through architectural integration
2Adaptability or versatility
If a conventional recommendation model uses only collaborative information from a data set, then the model architecture remains simple, but the model cannot obtain external semantic knowledge
Solution Approach 1:
The patent introduces a pre-trained language model as an intermediary component that provides semantic knowledge to the recommendation system. This language model acts as a mediator between external semantic knowledge and the collaborative filtering mechanism, enabling the system to access external knowledge without completely redesigning the recommendation architecture, thus balancing adaptability improvement with manageable complexity increase
3Measurement precision
If semantic information is incorporated into the recommendation model, then recommendation precision is improved, but online inference latency increases
Solution Approach 1:
The patent performs preliminary processing by pre-training a language model on large-scale text data before deployment in the recommendation system. This pre-training phase captures semantic knowledge in advance, allowing the model to leverage pre-learned representations during online inference without requiring computationally intensive processing at query time, thereby reducing inference latency while maintaining improved recommendation precision
Data Source
AI summary
A data processing method is provided. The method includes: obtaining first data, where the first data includes attribute information of a user or attribute information of an item in a plurality of dimensions; generating second data based on the first data, where the second data is a natural language text used to describe the attribute information; obtaining a first feature representation based on the first data through an embedding network, where the first feature representation includes a plurality of embedding vectors, and each embedding vector corresponds to one dimension; obtaining a second feature representation based on the second data through a natural language processing model, where the second feature representation includes a semantic feature of the second data; and updating the embedding network based on a difference between the second feature representation and the first feature representation, to obtain an updated embedding network.