Query Intent Inference for Search Response Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search query processing techniques often fail to accurately respond to users when the query is not lexically related to any media title in the catalog, leading to uninformative and frustrating results, as they do not account for the intent behind the search query or provide explanations for missing media titles.
Innovation Solution
A computer-implemented method that computes lexical similarity scores and user engagement scores to infer the intent of search queries, generating responses that include relevant media titles and explanatory messages for out-of-catalog queries, by processing search queries through a system involving a query intent application and a query processing application that utilize entity relevance scores and category relevance scores.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If lexical similarity scoring is used to respond to search queries, then the system can provide automated responses based on text matching, but the system fails to provide informative responses when queries are not lexically related to catalog items
Solution Approach 1:
The patent introduces an intermediary intent classification layer between the lexical similarity scoring and the final query response. This intermediary component analyzes the search query to determine user intent (e.g., searching for a specific title, exploring a genre, finding content by a creator) and uses this intent information to enhance the response. When a query is not lexically related to catalog items, the intent classification provides explanatory information about why no exact matches were found, thereby resolving the contradiction between automated response capability and information completeness.
2Ease of operation
If the system returns lexically similar media titles for unmatched queries, then it maintains response generation, but the returned titles may not be of interest to the user and lack explanatory context
Solution Approach 1:
The patent dynamically adjusts the query response based on the classified user intent. Rather than always returning lexically similar titles, the system adapts its response strategy: for specific title searches, it provides explanatory messages about why the title wasn't found; for exploratory searches, it returns relevant catalog items; for creator-based searches, it filters by creator information. This dynamic adaptation resolves the contradiction by making the response system flexible and context-aware, improving both ease of operation and result reliability.
Solution Approach 2:
The patent implements a feedback mechanism where the classified intent information feeds back into the query response generation process. The intent classification results are used to modify and refine the final response, ensuring that lexically similar titles are only returned when appropriate for the user's intent, and that explanatory messages are provided when matches are not found. This feedback loop ensures that the response is both operationally smooth and reliably relevant to user needs.
Data Source
AI summary
In various embodiments, an intent-based query processing application processes search queries. The intent-based query processing application computes lexical similarity scores between a search query and a set of entities. The intent-based query processing application computes entity relevance scores based on the lexical similarity scores and user engagement scores associated with both the search query and the set of entities. The intent-based query processing application computes a first category relevance score associated with both the search query and a first category based on the entity relevance scores. The intent-based query processing application determines an intent associated with the search query based on the first category relevance score. The intent-based query processing application generates a response to the search query based on the intent.