Latent Intent Clustering for Report Feature Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face confusion and inefficiency when creating new reports due to the overwhelming number of potential features, leading to difficulties in determining the optimal combination of fields, filters, and derived fields.
Innovation Solution
A computer-implemented method of latent intent clustering, which encodes features from user reports into a binary matrix, calculates cosine similarities, and clusters reports based on these similarities to identify common features and intent behind each cluster, thereby suggesting appropriate features for new reports.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If users manually select features for reports, then report customization is possible, but the process becomes time-consuming and inefficient due to overwhelming options
Solution Approach 1:
The system automatically performs latent intent clustering on historical reports to identify patterns and suggest optimal feature combinations for new reports, eliminating the need for users to manually evaluate numerous feature options. The system serves itself by learning from past reports and providing intelligent recommendations.
Solution Approach 2:
The patent introduces an intermediary system that acts as a mediator between users and the overwhelming number of feature options. This system processes historical report data, performs clustering analysis, and presents simplified recommendations to users, reducing the complexity perceived by users while maintaining customization capabilities.
2Manufacturing precision
If users review all potential features, then optimal report configuration is achieved, but the process becomes confusing and time-consuming
Solution Approach 1:
The system performs preliminary action by pre-processing historical report data to identify latent intent patterns before the user needs to create a new report. Through pre-clustering analysis, the system prepares and organizes feature combinations in advance, so when a user needs a new report, ready-made recommendations are immediately available without requiring the user to review all possibilities from scratch.
Solution Approach 2:
The system uses feedback from historical report creation patterns to continuously improve its clustering models. By analyzing what features were commonly combined in successful reports and what user preferences emerge from usage patterns, the system refines its recommendations, providing increasingly accurate and time-saving suggestions for report configuration.
Data Source
AI summary
A method of latent intent clustering is provided. The method comprises encoding identified features in a number of electronic user reports in a database. A binary matrix is created, wherein each row of the binary matric represents a different report and each column represents a different available feature. A 1 is placed in each cell of the matrix that matches a feature present in a user report. Cosine similarities are calculated for the user reports, and a similarity matrix is created, wherein each row and column of the binary matrix represents a different report, and wherein the cosine similarities of the reports are placed in corresponding cells of the matrix. The reports are clustered according to the cosine similarities. Features common to reports in each cluster are identified, and an intent of each report cluster is labeled according to the common features.