Ontology Graph Query Generation for Collaborative Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
During project planning and idea exploration, collaborators face challenges in efficiently sharing and investigating information across varying domain expertise levels, requiring significant time to search, filter, and structure relevant information, which can be iterative and time-consuming.
Innovation Solution
A computer-implemented method and system that automatically captures data during collaborative sessions, processes it to extract features like keywords and context, structures them into an ontology graph, generates search queries using a machine learning model, executes these queries, provides search results, receives feedback, tags results with weights, and re-trains the model for improved query generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual searching and information filtering is performed by collaborators, then relevant information can be found, but significant time investment is required
Solution Approach 1:
The system automatically captures data from collaborative tools, extracts features, generates search queries, and retrieves information without requiring manual intervention from collaborators. The machine learning model autonomously performs the information retrieval task that would otherwise require significant human time investment.
Solution Approach 2:
The patent introduces an intermediary system comprising a machine learning model that acts as a bridge between collaborators and information sources. This intermediary automatically translates collaborative data into search queries and retrieves relevant information, eliminating the need for collaborators to manually search and filter information themselves.
2Adaptability or versatility
If collaborators with varying domain expertise investigate information, then comprehensive coverage is achieved, but the process becomes complex and iterative
Solution Approach 1:
The machine learning model is designed to handle multiple domain areas and types of collaborative tools universally. It can process data from various sources including project planning software, task planning tools, and communication platforms, adapting to different domains without requiring domain-specific customization for each case.
Solution Approach 2:
The system adjusts the complexity of search queries and information retrieval strategies based on the characteristics of the input data and detected domain requirements. The machine learning model dynamically modifies search parameters and query structures to optimize information retrieval across varying domain expertise levels without manual intervention.
3Loss of information
If manual information filtering and structuring is performed, then relevant information is organized, but the process is time-consuming and iterative
Solution Approach 1:
The patent replaces the mechanical manual process of information filtering and structuring with an automated machine learning system. The model automatically extracts features from captured data, generates structured search queries, and organizes retrieved information, substituting human manual effort with automated computational processes.
Solution Approach 2:
The system performs preliminary feature extraction and query generation automatically as data is captured from collaborative tools. By preparing search queries in advance based on extracted features before full information retrieval is needed, the system reduces the overall time required for information organization and structuring.
Data Source
AI summary
Automatically capturing information and performing content searching may include extracting features from data content, for example, captured via an image capturing device or by another method. Features extracted from the data content are structured into an ontology graph representing keywords and contextual relationships. Search queries are generated based on the ontology graph, by inputting the ontology graph to a query generating machine learning model trained to predict one or more search queries. The search queries are executed and one or more search results are presented on a user interface, for example, a display device. Based on received feedback on the search results, the machine learning model is retrained.