Radial Time Schema for Anomaly Detection in User Transactions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer security systems fail to effectively detect unauthorized account access and misuse by analyzing transaction patterns based on time of day, leading to potential damage from unauthorized transactions.
Innovation Solution
A system that generates a smoothed probability distribution map of user transaction times to identify unusual transaction times, preventing potentially unauthorized transactions by transforming time of day into a two-dimensional representation, which helps in determining the likelihood of a transaction's authenticity based on past patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional transaction monitoring methods are used, then system simplicity is maintained, but detection accuracy of unauthorized transactions deteriorates
Solution Approach 1:
The patent transforms one-dimensional time data into two-dimensional coordinates using sinusoidal transformations (x=sin(2πt/24), y=cos(2πt/24)), creating a circular time representation that preserves temporal relationships while enabling sophisticated pattern analysis. This dimensional transformation allows the system to detect unauthorized transactions with high accuracy without requiring complex rule-based systems.
Solution Approach 2:
The patent replaces traditional mechanical rule-based monitoring systems with a probabilistic machine learning approach using smoothed probability distribution maps. Instead of predefined rules, the system uses kernel density estimation to learn transaction patterns automatically, achieving high detection accuracy while maintaining computational efficiency.
2Reliability
If detailed transaction pattern analysis is performed, then security detection capability is improved, but processing time increases
Solution Approach 1:
The patent pre-computes smoothed probability distribution maps from historical transaction data and stores them for rapid querying. When a new transaction arrives, the system only needs to evaluate its probability against the pre-computed map, dramatically reducing processing time while maintaining high security detection capability.
Solution Approach 2:
The patent transforms time parameters from linear format to circular coordinates and applies kernel density estimation to create continuous probability surfaces. These parameter transformations enable efficient probability calculations that balance detailed pattern analysis with fast processing speeds.
3Measurement precision
If circular time coordinate system is used, then temporal pattern recognition is improved, but computational complexity increases
Solution Approach 1:
The patent maps linear time data to two-dimensional circular coordinates using simple sinusoidal transformations, which preserve temporal relationships and enable sophisticated pattern recognition. The circular representation naturally handles wraparound at midnight and captures periodic patterns in user behavior.
Solution Approach 2:
The patent replaces complex temporal pattern matching algorithms with probabilistic density estimation on the circular coordinate system. This substitution simplifies the computational approach while improving temporal pattern recognition accuracy through the geometric properties of the circular representation.
Data Source
AI summary
Computer system security and efficiency of processing operations may be improved using techniques that are described relating to analyzing user actions based on time of day of occurrence, and using time of a day as a factor in determining whether a particular action should be allowed or disallowed. Past action times can be transformed to a two-dimensional representation using a radial time schema that avoids discontinuity. A probability distribution can indicate a likelihood of whether a new action fits a previous pattern. If a new user action is relatively unlikely due to time of day, the new user action might be denied/prevented from completing, thus enhancing computer system security and avoiding unnecessary computational processing costs.