Document Search Ranking Using Weighted Related Terms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing document search systems struggle with achieving highly accurate searches, particularly for intellectual property documents, due to the difficulty in selecting appropriate keywords and related terms, leading to inefficiencies and potential oversights in the search process.
Innovation Solution
A document search system that utilizes a processing portion to extract keywords and related terms from text data, assigns weights based on inverse document frequency and similarity degrees, and ranks search results to facilitate easy and accurate document retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a user manually selects keywords and searches patent documents, then the search can be conducted with simple system operation, but the search accuracy and completeness deteriorate due to user skill limitations and difficulty in selecting appropriate keywords
Solution Approach 1:
The system automatically extracts keywords from the input document itself, allowing the document to serve its own search purposes without requiring external keyword selection by the user. The keyword extraction unit identifies important terms directly from the document content, making the search process self-sufficient and eliminating dependency on user keyword selection skills
Solution Approach 2:
The system performs preliminary keyword extraction and related term identification before the actual search process. By pre-processing the document to identify and weight keywords and related terms in advance, the system prepares the search query automatically, eliminating the need for users to manually select keywords and improving both accuracy and ease of operation
2Measurement precision
If the system outputs many patent documents containing the keyword, then the search coverage is comprehensive, but the user must spend excessive time picking up needed documents from the large output
Solution Approach 1:
The system changes the parameter of keyword representation by not only extracting keywords but also identifying related terms and assigning weights to each. This multi-parameter approach (keyword + related term + weight) allows for more precise matching and ranking of search results, reducing the number of irrelevant documents the user must review
Solution Approach 2:
The system provides ranked search results based on the calculated scores from keyword and related term matching. This feedback mechanism prioritizes the most relevant documents at the top of the results, allowing users to quickly identify needed documents without manually reviewing all output documents, thereby reducing document selection time while maintaining search precision
3Measurement precision
If the system uses simple keyword matching, then the system complexity is low, but the search accuracy deteriorates due to inability to capture semantic relationships and related terms
Solution Approach 1:
The system segments the search process into distinct functional units: keyword extraction unit, related term identification unit, weight calculation unit, and search execution unit. This segmentation allows each unit to specialize in a specific task, improving overall search accuracy through coordinated operation of multiple specialized components without requiring the entire system to be overly complex
Solution Approach 2:
The system introduces an intermediary layer between simple keyword matching and final search results. The keyword extraction unit and related term identification unit act as intermediaries that process the input document to generate a enhanced query representation, bridging the gap between simple input and accurate search results while managing system complexity through modular design
Data Source
AI summary
A highly accurate document search, particularly a search for a document relating to intellectual property, is achieved with an easy input method. A document search system includes a processing portion. The processing portion has a function of extracting a keyword included in text data, a function of extracting a related term of the keyword from words included in a plurality of pieces of first reference text analysis data, a function of giving a weight to each of the keyword and the related term, a function of giving a score to each of a plurality of pieces of second reference text analysis data on the basis of the weight, a function of ranking the plurality of pieces of second reference text analysis data on the basis of the score to generate ranking data, and a function of outputting the ranking data.