Probabilistic Inference for Anomalous Behavior Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data processing systems fail to effectively analyze and identify anomalous criminal or terrorist activities due to the challenge of dealing with divergent data at different levels of granularity and the lack of ability to compare and assign probabilities to inferences across various data sources, leading to difficulties in detecting subtle crimes like identity theft and covert operations.
Innovation Solution
A computer-implemented method using a centralized database that conforms divergent data to common dimensions, applies rules to compare data, and executes queries to infer probabilities of anomalous behaviors, allowing for the identification of potential threats by analyzing metadata and associated keys related to cohorts, hierarchies, sources, and probabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a centralized database is used to store vast amounts of divergent data from multiple sources, then the quantity of available information increases, but the complexity of processing and analyzing this data increases
Solution Approach 1:
The patent segments the divergent data into standardized components with defined schemas, allowing complex data to be broken down into manageable, analyzable units that can be processed systematically
Solution Approach 2:
The patent introduces an intermediary processing layer that standardizes and normalizes divergent data from multiple sources before analysis, acting as a mediator between raw data and analytical processes
2Adaptability or versatility
If data from multiple sources with different granularities is integrated, then the versatility of analysis increases, but the difficulty of comparing and analyzing data increases
Solution Approach 1:
The patent transforms data from different granularities into a common parameter space through standardization, allowing comparison across diverse data types by changing their representation parameters to a unified schema
3Measurement precision
If probabilistic inference is applied to identify anomalous behaviors, then the accuracy of threat detection increases, but the computational resources required increase
Solution Approach 1:
The patent applies probabilistic inference selectively to specific data patterns and anomaly detection scenarios rather than uniformly across all data, performing partial analysis only where needed to identify potential threats
4Reliability
If rules are applied to compare data and infer probabilities, then the reliability of anomaly detection increases, but the time required for analysis increases
Solution Approach 1:
The patent performs preliminary data standardization and preparation before applying probabilistic rules, pre-processing the data into a format ready for rapid analysis, thereby reducing the time required during actual anomaly detection
Data Source
AI summary
Inferring a probability of a first inference absent from a database at which a query regarding the inference is received. The first inference relates to identifying anomalous behavior of cohorts members, which identification can be used to identify crimes that are particularly difficult to detect, such as identity theft. Each datum of the database is conformed to the dimensions of the database. Each datum of the plurality of data has associated metadata and an associated key. The associated metadata includes data regarding cohorts associated with the corresponding datum, data regarding hierarchies associated with the corresponding datum, data regarding a corresponding source of the datum, and data regarding probabilities associated with integrity, reliability, and importance of each associated datum. The query is used as a frame of reference for the search. The database returns a probability of the correctness of the first inference based on the query and on the data.