Conversational Agent Scam Caller Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies are ineffective in detecting scam callers before a subscriber answers a call, as they rely on caller blacklisting or whitelisting and can be evaded by scammers using advanced calling technologies, leading to a high volume of unwanted calls.
Innovation Solution
A system that employs a conversational agent with an IVR model trained to simulate subscriber speech and a machine learning model to analyze phone calls, allowing the agent to engage callers in a simulated conversation and detect scam callers without requiring the subscriber to answer, thereby automatically disconnecting or alerting the subscriber to potential scams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If caller blacklisting or whitelisting is used to detect scam callers, then detection capability is provided, but it can be evaded by scammers using advanced calling technologies
Solution Approach 1:
The system performs preliminary actions by answering the call and engaging the caller in conversation before the subscriber receives the call. The conversational agent analyzes caller responses in real-time to detect scam indicators, allowing the system to identify and block scam callers before they can reach the subscriber, thereby improving detection reliability while maintaining adaptability against evolving scam techniques.
2Measurement precision
If the subscriber answers the phone call to determine whether it is a nuisance phone call, then detection accuracy is improved, but the burden and exposure to unwanted calls increases
Solution Approach 1:
The system introduces a conversational agent as an intermediary between the subscriber and the caller. The agent answers calls, engages in conversation, and analyzes caller responses to detect scam indicators. This intermediary performs the detection function that would otherwise require the subscriber to answer and evaluate the call themselves, thereby maintaining detection accuracy while eliminating the burden and exposure to unwanted calls.
3Reliability
If a conversational agent engages callers in simulated conversation to detect scams, then protection effectiveness is improved, but system complexity increases
Solution Approach 1:
The conversational agent is designed to autonomously engage callers in conversation and self-evaluate their responses against known scam indicators. The system automatically detects scam patterns, makes determination decisions, and blocks suspicious calls without requiring manual intervention or complex external analysis systems, thereby improving protection effectiveness while managing system complexity through self-service automation.
Data Source
AI summary
Systems and methods for detecting indications of a scam caller are disclosed. Call data, such as call audio, is received and used to create a training dataset. Using the training dataset, a machine learning model is trained to detect indications of a scam caller in a phone call. An Interactive Voice Response (IVR) model is trained or configured, using voice samples of speech of a subscriber of a telecommunications service provider, to simulate speech and conversation of the subscriber. A conversational agent is generated using the IVR model and the trained machine learning model. The conversational agent receives a phone call, engages a caller in simulated conversation, and detects indications of whether the caller is a likely scam caller. If the caller is determined to be a likely scam caller, an alert can be generated and/or the call can be disconnected.