Message-Centric Social Network Analysis for Structured Data Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data analysis tools are inadequate for correlating and analyzing relationships between disparate documents from both structured and unstructured data stores, particularly in identifying entities across various names and accounts, and visualizing complex relationships effectively.
Innovation Solution
The approach characterizes all electronic data files as 'messages' and correlates them using social network analysis, creating a social network to explore and visualize relationships between entities and message content, employing multi-aspect viewing tools to analyze structured and unstructured data in a unified manner.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional text-mining tools are used to analyze unstructured documents, then semantic network relationships between concepts can be visualized, but the ability to correlate and analyze relationships between disparate documents from structured and unstructured data stores is insufficient
Solution Approach 1:
The patent merges structured and unstructured data stores into a unified message-centric framework. All electronic data files (emails, documents, financial transactions) are characterized as messages and correlated through a common message store, enabling comprehensive relationship analysis across previously disparate data types.
Solution Approach 2:
The system creates a universal message store that handles multiple types of electronic data files uniformly. By characterizing all documents as messages with standardized properties (sender, recipient, content, timestamp), the system achieves multi-functionality in analyzing relationships across different data sources.
2Measurement precision
If extensive databases and indices are created to track terms and concepts, then document relationships can be analyzed, but the complexity of the system increases and meaningful analysis of unstructured content becomes difficult
Solution Approach 1:
The patent extracts the essential relationship-correlation functionality from complex database systems and implements it directly within the message store structure. By storing messages with inherent relationship properties rather than requiring separate indexing systems, the solution reduces system complexity while maintaining analytical precision.
Solution Approach 2:
The message store acts as an intermediary layer between raw electronic data files and analysis tools. By standardizing all documents as messages with uniform properties, it simplifies the interface between diverse data sources and analytical functions, reducing overall system complexity.
3Productivity
If Boolean logic queries are used against structured information, then simple term retrieval is achieved, but effective analysis of unstructured information and complex relationships is limited
Solution Approach 1:
The patent adds a new dimension to information retrieval by incorporating social network analysis and multi-aspect viewing capabilities. Instead of relying solely on Boolean logic, the system provides visual interfaces that display relationship networks, enabling users to analyze complex relationships through graphical representations rather than complex queries.
Solution Approach 2:
The system replaces mechanical Boolean logic querying with automated social network analysis algorithms. By using computational methods to automatically detect and visualize relationships between messages and entities, the system improves both productivity and ease of operation compared to manual Boolean querying.
4Loss of information
If data is presented in tabular formats with lists of lists, then structured information is organized, but intuitive insights into complex relationships between dozens or thousands of documents are difficult to obtain
Solution Approach 1:
The patent transforms flat tabular data into multi-dimensional visual representations using social network graphs and multi-aspect viewing interfaces. These visualizations display relationships between entities and messages in spatial arrangements that preserve contextual information and enable intuitive understanding of complex relationships.
Solution Approach 2:
The system uses color coding and visual differentiation in its interface to represent different types of relationships, entities, and message properties. By applying color changes and visual cues, the system enhances the ease of operation for analyzing complex relationships while preserving relationship context.
Data Source
AI summary
Electronic data files are broadly characterized as “messages” and a social network is constructed by analyzing one or more messages exchanged between various entities. Additionally, messages from structured and/or unstructured data stores are correlated using one or more common/related data elements from two or more messages. Further, the social network and the concepts contained in the exchanged messages (i.e. semantic network) can be visualized using a series of multi-aspect viewing tools. Finally, in conjunction with the social network and the semantic network, a message network based on the chronological relationship of the messages (event network) can be constructed to analyze and visualize how the messages relate to each other in a time-based reference model. Once visualized, the relationship of the concepts contained in the messages as well as the relationship between the entities and the timing involved in the exchange of messages can be analyzed for desired information.