Answer Keyword Selection via Relation Value Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for providing answer keywords to user inquiries lack effectiveness in handling user preferences and preventing abuse, leading to unreliable search results.
Innovation Solution
A method and device that calculate a relation value for answer keywords based on search word history and webpage information, selecting and transmitting preferred keywords to user terminals, with the ability to adjust values based on user feedback and prevent abuse by revising relation values using similarity calculations across similar search word groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If answer keywords are provided based on simple search frequency, then the system is easy to operate, but the reliability of search results deteriorates due to inability to handle user preferences and abuse
Solution Approach 1:
The system changes parameters by calculating relation values between inquiry search words and answer candidate keywords based on multiple factors including search history frequency, webpage information frequency, and similarity to other inquiry search words. This multi-parameter approach improves reliability while maintaining ease of operation through automated calculation.
Solution Approach 2:
The system implements feedback mechanisms by using search history data from multiple user terminals to continuously refine and update relation values. This feedback loop allows the system to learn from user behavior patterns and improve search result reliability over time without increasing operational complexity.
2Reliability
If relation values are calculated based on comprehensive search history and webpage information, then the reliability of answer keywords is improved, but the device complexity increases
Solution Approach 1:
The system segments the calculation process into distinct components: extracting inquiry search words from search history, identifying answer candidate keywords, calculating frequencies from different sources (search history and webpage information), computing similarity values, and combining these into final relation values. This segmentation manages complexity by organizing the comprehensive analysis into modular steps.
Solution Approach 2:
The system performs preliminary actions by pre-calculating and storing search history data, webpage information frequencies, and similarity relationships between inquiry search words before they are needed for generating answer keywords. This preparation reduces the computational complexity during actual query processing while maintaining high reliability.
3Measurement precision
If the system processes search history from multiple user terminals, then the accuracy of answer keywords is improved, but the loss of time increases due to extensive data processing
Solution Approach 1:
The system performs preliminary processing of search history data from multiple user terminals by pre-calculating frequencies of inquiry search words and answer candidate keywords, and pre-computing similarity values between related search words. This advance preparation enables fast retrieval and combination of data during actual query processing, improving accuracy without proportionally increasing processing time.
Solution Approach 2:
The system applies local quality by focusing computational resources on calculating relation values for the specific inquiry search word and its most relevant answer candidate keywords, rather than uniformly processing all possible keyword combinations. This targeted approach improves accuracy for the specific query while reducing overall processing time.
Data Source
AI summary
Provided is a method of providing an answer keyword, the method including: obtaining at least one of a search word history including a first inquiry search word of a certain domain pre-received from first user terminals, and webpage information selected by the first user terminals from a search result according to the search word history; extracting answer candidate keywords regarding the first inquiry search word from at least one of the search word history and the webpage information based on keyword lists of the certain domain; calculating a relation between the first inquiry search word and each of the extracted answer candidate keywords; and when the first inquiry search word is received from a second user terminal, transmitting answer keywords for the first inquiry search word, which are selected from the answer candidate keywords based on the relation, to the second user terminal.