Cognitive Computing System Using Multiple Model Structures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in efficiently generating accurate and complete client data sets due to inefficient query sequences that often result in incomplete or conflicting information, leading to frustration and reduced user engagement.
Innovation Solution
A model-based cognitive computing system generates tailored query sequences using multiple models (BPMN, CMMN, DMN) to optimize data collection by selecting the most efficient and relevant questions based on user context, historical data, and external data sources, ensuring accurate and complete data sets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single model structure is used to generate query sequences, then the system complexity is low, but the accuracy and completeness of client data sets deteriorate
Solution Approach 1:
The patent combines multiple model structures (BPMN model for process flow, CMMN model for case management, and DMN model for decision logic) into a unified cognitive computing system. This merging allows the system to leverage the strengths of each model type to generate more accurate and complete client data sets while maintaining manageable complexity through modular architecture.
Solution Approach 2:
The patent creates a multi-functional query generation system where a single cognitive computing platform can perform multiple functions: process modeling (BPMN), case management (CMMN), and decision-making (DMN). This universal approach allows the system to adapt to different data collection scenarios and generate comprehensive query sequences for various client situations.
2Loss of information
If comprehensive queries are generated to ensure complete data sets, then the data completeness improves, but the user frustration increases due to excessive questioning
Solution Approach 1:
The patent implements dynamic query generation that adapts to user context and historical data. The system adjusts the number, type, and sequencing of queries based on real-time user responses, previously collected data, and pattern recognition from historical interactions. This dynamic approach ensures comprehensive data collection while minimizing user frustration by avoiding redundant or unnecessary questions.
Solution Approach 2:
The system incorporates feedback loops where user responses to queries are analyzed in real-time to adjust subsequent query generation. The cognitive computing system learns from user interactions and modifies the query sequence to balance data completeness with user experience, reducing frustration while maintaining data quality through continuous adaptation.
3Measurement precision
If multiple external data sources are queried to verify client information, then the data accuracy improves, but the time required for data collection increases
Solution Approach 1:
The patent implements preliminary data verification by pre-configuring decision logic and validation rules that determine which external data sources need to be queried. The system performs preliminary assessments of client information using available data and historical patterns to identify only the critical verification steps required, reducing unnecessary external queries and minimizing data collection time while maintaining accuracy.
Solution Approach 2:
The system dynamically changes verification parameters based on the specific client context, data type, and risk assessment. Instead of uniformly querying all external sources for every client, the cognitive computing system adjusts verification depth and source selection based on parameters such as data criticality, client history, and confidence levels, optimizing the balance between accuracy and time efficiency.
Data Source
AI summary
A method of populating a data set includes generating a plurality of models that model the behavior of an agent process, where the plurality of models includes a first model, a second model, and a third model. The method also includes using the plurality of models to generate one or more requests for one or more external data sources, and using the plurality of models to select a plurality of queries from a data store of predefined queries. The plurality of queries are selected by the plurality of models to request information that is missing from the data set. The method also includes populating at least a portion of the data set using information received in response to the one or more requests for the one or more external data sources and the plurality of queries.