Spam Detection Module Using Call Pattern Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional spam call blocking systems face limitations in identifying new or pattern-changing spam calls, relying on user reports and content analysis, which can lead to delayed recognition and bypassing of spam calls.
Innovation Solution
A spam blocking method using call log analysis that trains a spam determination module with Gaussian Naive Bayes AI logic to analyze call patterns of both the originator and recipient, determining spam calls based on propensity and re-training based on user interactions, providing originator information and calculating a spam index for blocking decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If user report-based spam blocking is used, then system simplicity is maintained, but spam identification accuracy deteriorates for new or pattern-changing spam calls
Solution Approach 1:
The system performs self-learning by automatically analyzing call logs and call patterns to train the spam determination module without requiring manual user input for each spam case. The system serves itself by collecting call data, extracting features, and updating its classification model autonomously based on observed patterns and user feedback.
Solution Approach 2:
The system incorporates feedback mechanisms where user responses to calls (answering, rejecting, or ignoring) are recorded and used to re-train the spam determination module. This feedback loop allows the system to learn from actual user behavior and improve its accuracy over time, adapting to evolving spam patterns.
2Reliability
If content analysis of call logs is performed, then spam detection capability is improved, but processing time increases
Solution Approach 1:
The system performs preliminary action by pre-extracting and storing call pattern features during normal operation. Features such as call frequency, timing patterns, and recipient distribution are collected and stored in advance, allowing the spam determination module to quickly query and analyze this pre-processed data during actual spam detection without performing complex analysis in real-time.
Solution Approach 2:
The system changes parameters by transforming raw call log data into extracted call pattern features (such as average call duration, call frequency per hour, time of day patterns). This parameter transformation converts complex raw data into simplified feature representations that can be quickly processed by the spam determination module using classification algorithms.
3Reliability
If call pattern analysis is performed for every call, then spam identification accuracy is improved, but system response speed deteriorates
Solution Approach 1:
The system segments the call analysis process into distinct components: call log collection, feature extraction, pattern matching, and spam determination. Each segment operates independently and processes specific portions of the data, allowing the system to maintain high accuracy through comprehensive analysis while improving response speed by avoiding the need to perform all analysis steps sequentially for every incoming call.
Data Source
AI summary
The inventive concept relates to a spam blocking method, and the spam blocking method includes training a spam determination module using call pattern data of an originator and a recipient secured based on call logs and re-training the spam determination module according to a maneuver of a user for an incoming call received in a recipient terminal, thus enabling spam determination customized for each user.