Search Reusability Appraisal via Past Request Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for searching and retrieving reusable software assets in software development systems do not effectively appraise reusability over time and often return irrelevant results due to lack of information on past search requests, even when semantically similar search requests are made.
Innovation Solution
A method that correlates new search requests with past requests by adjusting the relevance of search results based on semantic similarity, using a search server to store and retrieve search history from a database, and employing traditional search methods like Boolean or form-based searches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional search methods process each search request separately, then the search process is simple and fast, but the reusability appraisal of assets over time is meaningless and irrelevant results are returned
Solution Approach 1:
The system performs preliminary actions by storing and indexing past search requests and their results in a database before new searches occur. This historical data is prepared in advance to enable correlation with future search requests, improving reusability appraisal without adding significant complexity to the search execution process
Solution Approach 2:
The system implements feedback by analyzing correlations between past search requests and current search requests. Results from past searches are fed back into the system to adjust and improve the appraisal of asset reusability for current searches, creating a continuous learning mechanism that enhances reliability over time
2Measurement precision
If past search history is incorporated to improve reusability appraisal, then search result relevance improves, but system complexity and data storage requirements increase
Solution Approach 1:
The system extracts only the essential elements from past search requests - specifically the search terms, results, and correlation metrics - and stores them in a structured index. This selective extraction reduces the volume of stored data while maintaining the precision needed for accurate reusability appraisal and search result relevance
Solution Approach 2:
The system changes parameters by transforming raw search history data into indexed, correlated data structures with specific parameters such as search terms, result identifiers, and correlation scores. This parameter transformation reduces storage requirements while enabling precise measurement of search result relevance through efficient data retrieval
Data Source
AI summary
A method and system for searching and retrieving reusable assets is disclosed. An index file in a database stores reusable assets. A search request history in the database stores past search requests and corresponding past search results. A search program performs a search upon receiving a new search request within the index file and generates a new search result. The search program correlates past search requests to the new search request by comparing search terms in past search requests to search terms in the new search request. Search terms in past search requests are replaced with semantically similar terms based on taxonomy level and synonyms and results of highly correlated past search requests are added to the new search result. Reusability of each reusable asset in the new search result is appraised.