Meeting Analytics Platform With Knowledge Graph Augmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies lack robust software-based approaches for capturing and analyzing deeper meeting data, extracting insights, and augmenting meeting content with additional information over time, including speaker identification and sentiment analysis, across various online interactions and documents.
Innovation Solution
A data modeling and analytics platform utilizing machine learning and artificial intelligence to perform speaker identification, generate transcriptions, and augment them with knowledge graphs, providing observability and explainability, and enabling dynamic information discovery and analysis of online interactions and documents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If existing technology is used for online interactions, then basic audio and video capabilities are provided, but deeper meeting data cannot be captured and analyzed
Solution Approach 1:
The patent implements nested processing layers where basic transcription is enhanced with speaker identification, which is further enhanced with knowledge graph integration. Each layer nests within and builds upon the previous layer, allowing deep analysis while maintaining a modular structure that manages complexity.
Solution Approach 2:
The patent introduces knowledge graphs as an intermediary layer between raw meeting data and analysis outcomes. The knowledge graph serves as a mediator that structures unstructured meeting content, enabling deeper insights without directly increasing the complexity of the core processing system.
2Measurement precision
If no speaker identification is performed, then transcription is simpler, but participant performance tracking over time is impossible
Solution Approach 1:
The patent performs speaker identification and clustering as a preliminary step before detailed transcription and analysis. By pre-segmenting the meeting data by speaker, the system enables accurate participant performance tracking while organizing the processing workflow to reduce overall complexity.
3Loss of information
If meeting content is not augmented with additional information, then processing is faster, but contextual understanding and explainability are limited
Solution Approach 1:
The patent segments the augmentation process into distinct modules: transcription generation, speaker identification, knowledge graph integration, and sentiment analysis. This segmentation allows the core transcription process to proceed quickly while optional augmentation modules can be applied selectively to preserve contextual information.
4Loss of information
If no knowledge graphs are applied, then entity recognition is simpler, but cross-references and inferences between data points cannot be made
Solution Approach 1:
The patent introduces knowledge graphs as an intermediary layer between raw meeting data and analysis outcomes. The knowledge graph serves as a mediator that structures unstructured meeting content, enabling deeper insights without directly increasing the complexity of the core processing system.
Data Source
AI summary
A data modeling and analytics platform augments and annotates content captured from a user's online interactions and other documents. The data modeling and analytics platform is performed within a machine learning and artificial intelligence-based processing environment that enables observability, explainability, and data analytics for dynamic information discovery over time within a user library that includes files representing the online interactions and documents containing information of user interest.