Virtual Corpus Engine Weight Matrix for Cognitive Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed cognitive systems face inaccuracies due to the inclusion of untrusted sources of information across various corpora, which can lead to less accurate results when performing cognitive operations across different topics, concepts, or domains.
Innovation Solution
A virtual corpus engine is implemented to customize the selection of actual corpora based on user feedback, adjusting weight values in a matrix to prioritize trusted sources for specific topics, thereby eliminating untrusted sources and improving the accuracy of cognitive operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple actual corpora are included in the virtual corpus, then the coverage and versatility of information sources is improved, but the accuracy of cognitive operations deteriorates due to inclusion of untrusted sources
Solution Approach 1:
The patent applies local quality by assigning different weight values to different actual corpora based on their trustworthiness for specific topics. Each corpus receives a customized weight rather than uniform treatment, allowing the system to optimize for both coverage and accuracy by giving higher weights to trusted sources for particular topics while maintaining inclusion of multiple diverse corpora.
Solution Approach 2:
The patent changes the parameter of corpus weight values dynamically based on user feedback and topic relevance. The weight matrix stores and updates weight values for each corpus-topic combination, allowing the system to adapt the contribution of each corpus to the virtual corpus based on measured performance and user preferences, thereby resolving the contradiction between broad coverage and accurate results.
2Reliability
If a weight matrix is used to customize corpus selection, then the accuracy of cognitive operations is improved, but the device complexity increases
Solution Approach 1:
The patent segments the corpus selection problem into manageable components by using a weight matrix that separately stores weight values for each corpus-topic combination. This segmentation allows the system to handle complexity in an organized manner, where the virtual corpus engine can selectively apply weights based on the specific inquiry topic rather than requiring complex global optimization algorithms.
Solution Approach 2:
The weight matrix serves as an intermediary data structure that mediates between the multiple actual corpora and the cognitive operations. Rather than requiring direct complex interactions between corpora and queries, the weight matrix provides a simplified interface that translates topic-inquiry relationships into weighted corpus selections, reducing overall system complexity while maintaining accuracy.
3Measurement precision
If untrusted sources are eliminated from the virtual corpus, then the accuracy of results is improved, but the quantity of available information sources is reduced
Solution Approach 1:
The patent applies dynamics by making the composition of the virtual corpus dynamic rather than static. The weight values for each corpus are updated based on user feedback and performance metrics, allowing the system to adaptively include or exclude corpora based on their demonstrated trustworthiness. This dynamic approach maintains a large pool of potential sources while ensuring that only trusted sources actively contribute to results at any given time.
Solution Approach 2:
The patent uses partial action by selectively applying corpora to specific topics rather than using all corpora for all inquiries. The weight matrix enables the system to apply only the necessary subset of corpora for each topic, giving partial contribution to each corpus based on its relevance and trustworthiness for that particular topic, thereby maintaining accuracy without requiring all sources to be actively used.
Data Source
AI summary
Mechanisms are provided for implementing a virtual corpus engine that receives an inquiry to be processed and analyzes the inquiry to extract one or more features of the inquiry. The virtual corpus engine selects a weight matrix associated with a virtual corpus based on the extracted one or more features of the inquiry. The virtual corpus comprises a plurality of actual corpora of information. The weight matrix comprises a separate weight value for each actual corpus in the plurality of actual corpora. The virtual corpus engine processes the inquiry using a set of selected actual corpora selected from the plurality of actual corpora based on the weight values in the weight matrix and receives results of the processing of the inquiry using the set of selected actual corpora. The virtual corpus engine outputs the results of the processing of the inquiry.