Access Authentication via Historical Access Result Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fraud detection systems in resource security are not accurate, leading to unauthorized access as they fail to differentiate between legitimate and fraudulent access requests, resulting in delayed detection of compromised user information.
Innovation Solution
Implementing a system that analyzes the aggregation of historical access result values over time intervals to authenticate access requests, using vectors of historical access results to classify requests and determine access rules, thereby enhancing detection accuracy and precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional detection methods analyze individual data elements to determine access request validity, then the detection process is simple and fast, but the accuracy is insufficient leading to fraudulent access approval
Solution Approach 1:
The patent segments the detection process into multiple independent detectors, each analyzing specific data elements (username, email, IP address, device fingerprint, etc.) separately. Each detector generates an individual access result value that is then aggregated. This segmentation allows comprehensive analysis while maintaining modularity and manageability of the complex detection system.
Solution Approach 2:
The patent merges the results from multiple individual detectors by aggregating their access result values. The aggregation process combines information from username detector, email detector, IP address detector, device fingerprint detector, and other detectors to form a comprehensive assessment. This merging enables the system to leverage multiple data sources to improve overall detection accuracy.
2Reliability
If fraud detection systems use individual data element analysis, then the response time is fast, but the system cannot differentiate between legitimate and fraudulent requests resulting in delayed detection
Solution Approach 1:
The patent performs preliminary actions by continuously collecting and storing access result values from historical requests in vectors for each detector. This pre-computation and storage of historical data enables the system to quickly retrieve and aggregate relevant information when a new access request arrives, eliminating the need for time-consuming real-time analysis of all historical data while maintaining high detection reliability.
Solution Approach 2:
The patent implements feedback mechanisms where access result values from previous requests are fed back into the detection system through historical vectors. The aggregation process incorporates this feedback by comparing current request patterns against historical behavior patterns, allowing the system to learn from past fraudulent and legitimate requests and improve future detection accuracy without significant time delay.
3Measurement precision
If the system stores and processes aggregation of historical access result values, then the detection accuracy improves, but the memory access and processing complexity increases
Solution Approach 1:
The patent segments historical data storage into separate vectors for each detector type (username vector, email vector, IP address vector, etc.). This segmentation organizes the historical access result values in a structured manner that simplifies memory access patterns. When processing a new request, the system only needs to access the specific vectors corresponding to the detectors triggered by that request, rather than scanning through all historical data.
Solution Approach 2:
The patent applies local quality by storing access result values in detector-specific vectors rather than a single monolithic database. Each vector contains localized historical information relevant to its specific detector type. This localized organization allows the system to efficiently retrieve only the relevant historical data needed for each aspect of the analysis, reducing overall memory access complexity while maintaining high classification accuracy.
Data Source
AI summary
Embodiments of the present disclosure provide systems, methods, and apparatuses for addressing the above problems through the use of access rules that involve analyzing historical access request result data for various data elements individually and in combination over a predefined time interval. An automated determination can be made for whether a transaction can be authenticated based upon the historical access request result data (e.g., ultimately, deciding whether the data element or set of data elements are associated with a valid access request).