Automated Call Routing via Frequency Time Series Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated telephone calls are difficult to detect and block within communications networks, often requiring users to employ specific equipment or systems to avoid disturbances, and existing methods may not effectively differentiate between legitimate and nuisance calls.
Innovation Solution
A method that generates a call frequency time series to classify calls based on features such as transitions from quiet to busy periods and vice versa, using a sliding window detector and Mahalanobis distance to identify automated calls, and routes further calls accordingly, either to a voicemail server or the original destination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If users employ telephone answering machines or systems to block calls from unidentified numbers or blacklisted numbers, then automated calls can be blocked, but users must take significant action or acquire specific equipment
Solution Approach 1:
The network performs automated call detection and classification autonomously by analyzing call frequency time series patterns, eliminating the need for users to manually configure blocking systems or answer machines. The network self-services the harmful factor by automatically identifying and routing automated calls without user intervention.
Solution Approach 2:
The system performs preliminary analysis of call frequency patterns to classify calls as automated or manual before they reach the user. By analyzing transitions between quiet and busy periods in advance, the network can pre-rout automated calls to appropriate destinations, preventing them from disturbing users.
2Object-affected harmful factors
If existing blocking systems are used, then some automated calls can be blocked, but the systems are complex and require user configuration
Solution Approach 1:
The system transforms the detection approach by changing from analyzing individual call characteristics to analyzing temporal patterns in call frequency. By converting call data into time series representations and analyzing transitions between quiet and busy periods, the system achieves automated detection through parameter transformation rather than complex rule-based blocking.
Solution Approach 2:
The invention replaces mechanical/user-based blocking systems with an automated analytical system. Instead of relying on users to configure filters or answer machines, the network uses computational analysis of call frequency patterns to automatically identify and route automated calls, substituting manual operations with automated data processing.
3Measurement precision
If call frequency time series analysis is performed to detect automated calls, then automated calls can be effectively classified, but processing time and computational resources are required
Solution Approach 1:
The system segments the call frequency data into discrete time intervals and identifies transitions between quiet and busy periods. By dividing the continuous call stream into manageable temporal segments and analyzing characteristic variations within these segments, the system achieves accurate classification while optimizing processing efficiency through structured data organization.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
There is provided a mechanism that allows automated calls made from a set of one or more numbers to be detected and routed in an appropriate manner. The calls are classified based, at least in part, on one or more features of a call frequency time series that is generated from data representing a plurality of calls previously made from a set of one or more numbers by determining a respective number of calls made by the set of numbers during each of a plurality of time intervals. The classification indicates whether the calls include automated calls. Further calls from the set of numbers are routed in accordance with the classification.