Subscriber Data Protection via Multi-Stage Call Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to effectively detect and protect subscriber data from unwanted calls, leading to high false positive and false negative errors, as attackers use sophisticated methods to hide their activities and deceive users.
Innovation Solution
A system and method that intercepts network traffic associated with calls, analyzes call parameters and stream data using techniques such as speech recognition, image recognition, and machine learning to identify unwanted calls, and inserts tags to declare the call as unwanted, thereby protecting the subscriber's data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional fraud detection methods are used, then the system is simple to operate, but the detection precision is low leading to high false positive and false negative errors
Solution Approach 1:
The fraud detection system is segmented into multiple independent analysis modules: call parameter analysis module, speech recognition module, image recognition module, and machine learning analysis module. Each module processes specific aspects of the call independently and contributes to the overall detection decision, allowing the system to achieve high detection precision through specialized analysis while maintaining manageable complexity through modular design.
Solution Approach 2:
The system transitions from traditional single-dimension fraud detection (analyzing only call parameters) to multi-dimensional analysis by incorporating speech recognition, image recognition, and machine learning techniques. This dimensional expansion enables comprehensive analysis of call content, speaker characteristics, and behavioral patterns, significantly improving detection precision while the systematic integration of these dimensions prevents overwhelming complexity.
2Measurement precision
If sophisticated analysis techniques are applied to detect unwanted calls, then the detection precision improves, but the processing time increases
Solution Approach 1:
The system performs preliminary analysis of call parameters and basic speech patterns during the call setup phase before the actual conversation begins. This preliminary action identifies obvious fraud indicators early, allowing the system to flag suspicious calls for more intensive analysis or reject them immediately, thereby improving detection precision while minimizing the time consumed by sophisticated analysis techniques on clearly fraudulent calls.
Solution Approach 2:
The system applies full sophisticated analysis techniques only to calls that meet specific suspicion criteria, rather than analyzing every call with maximum depth. For low-risk calls, the system performs partial analysis using streamlined methods, while reserving comprehensive speech, image, and machine learning analysis for calls that exhibit suspicious patterns, thus achieving high detection precision for fraudulent calls while reducing average processing time across all calls.
Data Source
AI summary
A method for protecting subscriber data includes intercepting network traffic associated with a call. The network traffic includes call parameters and call stream data. A first set of the call parameters is analyzed. A first probability value of the call being declared as unwanted is determined. The call stream data is analyzed to define a second set of call parameters. The first set of call parameters is reanalyzed based on the second set. A second probability value of the call being declared as unwanted is determined. A determination is made if the second probability value exceeds a second threshold value. The call is declared as unwanted, in response to determining that the second probability value exceeds the second threshold. The first and second sets of call parameters are transmitted to an application configured to protect data of a protected subscriber.