Progressive Similarity Detection for Unwanted Conversational Media Sessions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conversational media sessions over packet switched networks, such as IP networks, face challenges in handling unwanted sessions that can be annoying and potentially compromise security, due to the low cost and ease of generating large volumes of unwanted calls without requiring dedicated hardware.
Innovation Solution
A system and method for detecting unwanted conversational media sessions by calculating progressive similarity scores between real-time media data and reference data items, using a progressive similarity evaluation process to determine if a session is unwanted, and implementing traffic control rules to manage such sessions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If progressive similarity evaluation is implemented to detect unwanted sessions, then security and annoyance reduction are improved, but device complexity and processing requirements increase
Solution Approach 1:
The system pre-calculates and stores reference data items from previous conversational media sessions before they are needed for comparison. This preliminary action allows the detection system to quickly compare incoming sessions against historical data without performing complex real-time analysis, thereby improving security detection capability while reducing the complexity of real-time processing.
Solution Approach 2:
The similarity evaluation process is divided into multiple progressive stages, where similarity scores are calculated incrementally as more media data becomes available. This segmentation allows the system to make early termination decisions when sufficient similarity is detected, reducing overall processing complexity while maintaining high security detection accuracy.
2Measurement precision
If real-time similarity scores are calculated during session progress, then unwanted session detection accuracy is improved, but processing time and computational resources increase
Solution Approach 1:
The system implements early termination of similarity evaluation when the progressive similarity score exceeds a predetermined threshold. This allows the system to rush through the evaluation process for clearly unwanted sessions, achieving high detection accuracy without unnecessarily extending processing time for sessions that are obviously similar to previous unwanted sessions.
Solution Approach 2:
The system calculates similarity scores based on progressively accumulating media data, using only the portion of data needed to reach a confident detection decision. This partial action approach avoids processing the entire session duration when sufficient evidence for unwanted session detection is obtained earlier, thereby reducing overall processing time while maintaining detection accuracy.
3Reliability
If multiple progressive similarity scores are calculated, then detection reliability is improved, but computational complexity and processing overhead increase
Solution Approach 1:
The system dynamically adjusts the number of similarity scores calculated and the evaluation depth based on the accumulating evidence. As progressive similarity scores are calculated, the system can adaptively terminate evaluation when reliability thresholds are met, rather than always computing a fixed number of scores. This dynamic approach improves detection reliability while reducing processing complexity for clear-cut cases.
Data Source
AI summary
Some embodiments of the invention relate to a method and a system for detecting unwanted conversational media session data. In accordance with one aspect of the invention, a method of detecting unwanted conversation media session data according to some embodiments of the invention may include calculating two or more progressive similarity scores each with respect to a different instant during a progress of a real-time conversational media session, wherein each of said scores is associated with a similarity between the conversational media session's media data that was available at the associated instant and a reference data item corresponding to media data of a previous conversational media session, and evaluating progressive similarity between the real-time conversational media session and the reference data item based upon the two or more progressive similarity scores.