System and method for knowledge pattern search from networked agents,申请号US201113283442-传众专利搜索

System and method for knowledge pattern search from networked agents

申请公布号：US8903756(B2)

申请号：US201113283442

申请日期：2011.10.27

申请公布日期：2014.12.02

发明人：Zhao Ying;Zhou Charles C.

分类号：G06F17/00;G06N5/04

主分类号：G06F17/00

代理机构：

Invent Capture, LLC.

代理人：Invent Capture, LLC. ;Cho Samuel S.

摘要：One or more systems and methods for knowledge pattern search from networked agents are disclosed in various embodiments of the invention. A system and a related method can utilizes a knowledge pattern discovery process, which involves analyzing historical data, contextualizing, conceptualizing, clustering, and modeling of data to pattern and discover information of interest. This process may involve constructing a pattern-identifying model using a computer system by applying a context-concept-cluster (CCC) data analysis method, and visualizing that information using a computer system interface. In one embodiment of the invention, once the pattern-identifying model is constructed, the real-time data can be gathered using multiple learning agent devices, and then analyzed by the pattern-identifying model to identify various patterns for gains analysis and derivation of an anomalousness score. This system can be useful for knowledge discovery applications in various industries, including business, competitive intelligence, and academic research.

主权项：1. A system for knowledge pattern search and analysis from multiple learning agents, the system comprising: a learning agent configured to process historical data with a contextualization module, a conceptualization module, and a cluster profile module, wherein at least a portion of the historical data is grouped and categorized by a context-concept-cluster (CCC) data analysis method; the contextualization module first analyzing the historical data as an unsupervised machine-learning process to identify information patterns in the historical data without a user entry of a search term, and then selectively taking in a user-defined context for narrowing of the information patterns found from the unsupervised machine-learning process in the historical data, wherein the user-defined context is a condition, a keyword, an element, or a desired target information defined by the user; the conceptualization module configured to group elements and contexts together as a concept based on a calculation of a proximity of relationship between elements and contexts (EC matrix); the cluster profile module containing one or more clusters, wherein each cluster stores a portion of grouped and categorized historical data based on the user-defined context in the contextualization module and the calculation of the proximity of relationship between elements and contexts in the conceptualization module; a pattern-identifying model as a lookup table comprising keys and values, wherein each key contains a keyword related to contexts, concepts, and clusters, and wherein each value contains one or more relationship calculations between elements and contexts (EC matrix), contexts and concepts (CC matrix), concepts and sequences (SC matrix), or averages of SC matrix per cluster; and a CPU and a memory unit operatively connected to the system, wherein the CPU and the memory unit are configured to operate and execute the learning agent, the contextualization module, the conceptualization module, and the cluster profile module to construct and operate the pattern-identifying model based on information stored in the one or more clusters.