Neural Network Search Engine Query Vectorization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines face challenges in accurately matching user intents due to the ambiguity of semantemes in query inputs, leading to low accuracy in search results.
Innovation Solution
A neural network-based searching method that acquires user queries and pre-queries, generates a target term vector representation using an MLP, and forecasts optimized search results through a deep neural network semantic model to improve matching accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional search engine matching methods are used, then the search process is simple and fast, but the accuracy of search results is low due to semantic ambiguity
Solution Approach 1:
The patent introduces term vector representations as an intermediary between the query and search results. The query and search results are transformed into vector representations in a semantic space, allowing the system to capture semantic relationships rather than relying on simple keyword matching. This intermediary representation resolves the contradiction by enabling accurate semantic matching without requiring complex rule-based systems.
Solution Approach 2:
The patent changes the parameter representation from discrete keywords to continuous vector embeddings. By transforming the query and search results into vector representations and computing similarity in the continuous semantic space, the system achieves higher accuracy while maintaining computational efficiency, thus resolving the contradiction between accuracy and complexity.
2Measurement precision
If semantic understanding is enhanced to improve search accuracy, then the matching precision improves, but the computational complexity increases
Solution Approach 1:
The patent pre-computes and stores term vector representations for search results in advance. When a query is received, the system only needs to compute the query vector and compare it with pre-computed result vectors, significantly reducing the computational power required at query time while maintaining high intent matching precision.
Data Source
AI summary
A searching method and a searching apparatus based on a neural network and a search engine are disclosed, the searching method including: acquiring a query and a pre-query input by a user; acquiring a plurality of search results according to the query; generating a target term vector representation according to the query, the pre-query and the plurality of search results based on an MLP; and forecasting the target term vector representation based on a semantic model of a deep neural network so as to acquire a plurality of s optimized search results corresponding to the query.