Dynamic Case Tagging for Prioritized Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Caseworkers face challenges in identifying relevant cases when they do not know what to search for or what is needed to work on a case, as existing systems require parameter and content matching, which can be inefficient in dynamic situations.
Innovation Solution
A method that dynamically generates and compares tags for a work-in-progress case with completed cases, creating a prioritized list of relevant cases based on matching tags and case flow histories, using a tag engine to extract and identify tags from unstructured content, and updating the list in real-time as the case progresses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional parameter and content matching is used to search for relevant cases, then the system can retrieve cases based on predefined search criteria, but the system becomes inefficient when caseworkers do not know what to search for or what is needed
Solution Approach 1:
The system performs self-service by automatically analyzing the current case content and flow history to generate relevant case recommendations without requiring manual search input from the caseworker. The system extracts tags from unstructured content and compares them with completed cases to autonomously identify relevant precedents.
Solution Approach 2:
The system changes the search parameters dynamically by extracting tags from the actual case content and flow history rather than using fixed search criteria. This allows the system to adapt its search parameters to the specific context of each case, improving relevance when caseworkers are unsure of search terms.
2Measurement precision
If the system requires users to provide search parameters to find relevant cases, then the search can be targeted, but the system fails when users are unsure of the search parameters
Solution Approach 1:
The system eliminates the need for user-provided search parameters by automatically analyzing case content and flow history to determine relevant cases. This maintains high search relevance while making the system accessible to users who may not know appropriate search terms.
Solution Approach 2:
The system performs preliminary analysis of the current case content and flow history before generating case recommendations. By extracting tags and identifying patterns in advance, the system can provide relevant recommendations without requiring users to first formulate search parameters.
3Quantity of substance
If the system provides a static list of relevant cases, then the list can be comprehensive, but it cannot adapt to dynamic changes in case context and flow history
Solution Approach 1:
The system transforms the static case list into a dynamic recommendation list that automatically updates based on changing case context and flow history. As the case progresses and new information is added, the system re-analyzes the content and generates updated recommendations, maintaining adaptability while providing comprehensive relevant cases.
Solution Approach 2:
The system incorporates feedback loops where the actual case progression and flow history continuously refine the relevant case recommendations. This feedback mechanism ensures the list adapts to dynamic changes in case context while maintaining comprehensiveness through ongoing analysis of completed cases.
Data Source
AI summary
An approach for dynamically providing a prioritized list of relevant cases includes a computer generating a set of tags for a case and retrieving a set of tags for completed cases. The approach includes the computer sorting the set of tags for the case and the set of tags for the completed cases into a set of tag types. The approach includes the computer determining a set of matching tags for each of the set of tag types between the case and completed cases. Furthermore, the approach includes the computer determining an ordered list of the completed cases. The ordered list is based, at least in part, on a number of matching tags in the set of matching tags corresponding to the completed case in the completed cases, and a number of matching tags in the set of matching tags in a tag type in the set of tag types.