Early Pattern Detection in Enterprise Data via Knowledge Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current big-data analytics tools fail to effectively capture useful signals from noisy data and recognize patterns in incomplete information, especially in the context of connected enterprises and customers, where dynamics and real-time interactions generate vast amounts of data.
Innovation Solution
An early pattern detection platform that builds and updates knowledge graphs using artificial intelligence to extract web-based data, detect domain-relevant patterns, and output actionable events to improve enterprise operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional big-data analytics tools are used to detect dependencies between enterprises, then data analysis capability is provided, but the tools are unable to capture useful signals from noisy data and recognize patterns in incomplete information
Solution Approach 1:
The system segments the data processing task into multiple specialized components: a knowledge graph construction module that builds structured representations of enterprise relationships, an event detection module that identifies specific patterns, and a machine learning module that ranks events. This segmentation allows each component to specialize in handling noisy or incomplete data according to its strengths, improving overall pattern detection accuracy while maintaining reliability.
Solution Approach 2:
The patent introduces a knowledge graph as an intermediary structure between raw noisy data and pattern detection algorithms. The knowledge graph serves as a mediator that structures and contextualizes incomplete information, enabling more reliable signal capture. By representing enterprise relationships, interactions, and attributes in a structured graph format, the system can infer missing information and reduce noise before pattern recognition occurs.
2Productivity
If traditional analytics models are used, then basic data processing is achieved, but the models are unable to keep pace with the dynamics that result from connectivity and real-time interactions
Solution Approach 1:
The system implements dynamic adaptability through multiple mechanisms: the knowledge graph is continuously updated as new enterprise interactions occur, machine learning models are retrained on evolving data patterns, and event detection thresholds are adjusted based on current market conditions. This allows the system to maintain high detection speed while adapting to changing dynamics in real-time connected enterprise environments.
Solution Approach 2:
The patent incorporates feedback loops where detected events and patterns are fed back into the knowledge graph and machine learning models for continuous improvement. The system monitors its own performance and uses this feedback to refine its detection algorithms, update knowledge representations, and adjust to new dynamics in enterprise interactions, thereby maintaining both speed and adaptability.
3Loss of information
If comprehensive data analysis is performed across multiple networks and platforms, then insight generation is enhanced, but the complexity of processing massive amounts of data increases
Solution Approach 1:
The system extracts and separates critical information from massive datasets by focusing on specific event types and patterns relevant to enterprise operations. Rather than processing all data uniformly, the knowledge graph construction and event detection modules extract only the most significant signals, reducing the effective data volume that requires complex processing while retaining essential information for insight generation.
Solution Approach 2:
The patent transforms high-dimensional noisy data into a structured knowledge graph representation that organizes information along meaningful dimensions (entities, relationships, attributes, events). This dimensional transformation simplifies the data structure, making it more manageable and reducing the complexity required for comprehensive analysis while preserving information integrity across multiple networks and platforms.
Data Source
AI summary
Implementations of the present disclosure include receiving a goal, providing a problem-specific knowledge graph that is responsive to at least a portion of the goal, determining a set of events from the problem-specific knowledge graph, processing data representative of events in the set of events through a first machine learning (ML) model to provide a set of event scores, each event score in the set of event scores being associated with a respective event in the set of events, determining a sub-set of events based on the set of event scores, for each event in the sub-set of events, determining at least one action by processing a sequence of actions through a second ML model, and outputting the sub-set of events and a set of actions for execution of at least one action in the set of actions.