Semantic Search via Dynamic Word Vector Space
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional information search methods using relational databases struggle with providing accurate responses to user queries, especially when queries are not in the same format as stored information or require semantic inference, and managing ontologies for real-time information updates is challenging.
Innovation Solution
A method that generates basic and expanded word vectors from user queries and linguistic analysis results, using a vector space database to search for similar vectors, and adjusts weighting values based on the origin of these vectors to enhance search scores, allowing for improved information retrieval through web searching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a relational database is used for information search, then information can be stored and retrieved systematically, but appropriate information cannot be searched if the query format does not match the stored information format
Solution Approach 1:
The patent transforms the search mechanism from format-matching to semantic-similarity-based by changing the parameter from exact format correspondence to vector space similarity measurement. User queries and database information are both converted to word vectors, allowing searches based on semantic meaning rather than format matching.
Solution Approach 2:
The patent replaces the mechanical database query system with a semantic vector space system. Instead of using traditional database query mechanisms that require format matching, the system uses vector space similarity calculations to retrieve information based on semantic meaning.
2Adaptability or versatility
If ontology is used to handle semantic matters in queries, then semantic inference can be performed, but it is very difficult to construct and manage the ontology and impossible to construct so many ontologies for inferring queries
Solution Approach 1:
The patent extracts the semantic processing function from complex ontology construction and embeds it directly into the search mechanism through word vector representation. This eliminates the need for separate ontology construction and management systems while retaining semantic inference capabilities.
Solution Approach 2:
The patent introduces word vectors as an intermediary between user queries and database information. These vectors serve as a mediator that captures semantic meaning without requiring explicit ontology construction, enabling semantic inference through vector similarity calculations.
3Productivity
If conventional vector space models are used for information searching, then search operations can be performed, but recent information and large amounts of data cannot be effectively reflected
Solution Approach 1:
The patent makes the vector space dynamic by continuously updating it with new information from web sources. The system periodically crawls web data, processes it into word vectors, and updates the vector space database, ensuring recent information is reflected while maintaining search efficiency.
Data Source
AI summary
According to the present invention, a method for searching for information using the Web and a method for voice conversation using same involve generating a basic word vector for at least one of a provided user query and a language analysis result; using the basic word vector, searching a vector space database for a vector space corresponding to the basic word vector; searching, if a similarity between the basic word vector and the found vector space is lower than a preset reference, a vector space database for a vector space corresponding to an expanded word vector using the expanded word vector generated from the result of the Web search performed using at least one of the user query and the language analysis result; and searching for knowledge information on the basis of the vector space found in the basic searching step or the expanded searching step. Thus, improved research results may be provided in response to the user query.