Query Correction via Reattempt Learning for Search Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face frustration when searching for media content due to typographical errors, incorrect phrasing, or misinterpretation of voice inputs, leading to repeated queries and potential overcorrection by search systems, which can hinder the retrieval of desired results.
Innovation Solution
A system that identifies reattempt query pairs by determining if a second query is a reattempt of a first query, matches it with a similar query pair from a database of reattempt query pairs associated with acceptable results, and performs a search based on the identified pair, thereby streamlining the search process and avoiding overcorrection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system analyzes each query individually and outputs suggestions for alternative queries, then users can correct typographical errors and improve query accuracy, but users experience frustration and time loss when entering repeated queries to find desirable content
Solution Approach 1:
The system performs preliminary analysis by identifying reattempt queries and matching them with previously successful query pairs before the user completes their search sequence. When a reattempt query is detected, the system proactively retrieves successful alternative queries from the database and presents them to the user, eliminating the need for the user to manually enter multiple variations of the same query.
2Productivity
If the system automatically corrects user queries without user selection, then the search process is streamlined and users avoid manual correction, but the system overcorrects queries and may change the user's intended search meaning
Solution Approach 1:
The system implements feedback by presenting multiple corrected query options to the user rather than automatically selecting one correction. The user can review the suggested corrections and select the one that best matches their intent, or indicate that none are appropriate. This feedback loop ensures that automatic correction does not override the user's intended meaning while still streamlining the search process.
3Measurement precision
If the system presents multiple correction options to the user, then users can select the most appropriate correction and maintain query accuracy, but users experience additional time loss reviewing and selecting from multiple options
Solution Approach 1:
The system performs preliminary filtering by analyzing the user's query sequence and context to pre-rank correction options based on likelihood of matching user intent. The most probable corrections are presented first, and the system can automatically apply the top-ranked correction if confidence thresholds are met, reducing the number of options the user must review while maintaining high correction accuracy.
Data Source
AI summary
Systems and methods are described to access a set of reattempt query pairs, where each respective pair comprises an initial query and a reattempt of the initial query, and is associated with an indication of whether a reply generated for output based on the respective query pair was acceptable. In response to determining that a second query received after a first query constitutes a reattempt of the first query, a query pair in the set of reattempt query pairs may be identified that matches at least one of the first query and the second query, and is associated with an indication that a reply generated for output based on the query pair was acceptable. A search may be performed based on the identified query pair in the set of reattempt query pairs, and a reply may be generated for output based on the performed search.