Text Message Identification Using Octet Distribution Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying abnormal text messages, such as spam or scam messages, are ineffective when the number of reports is low or when malicious users change their phone numbers, leading to a blind area in identification and reduced accuracy.
Innovation Solution
The method involves acquiring and comparing first and second predetermined octet distribution characteristics of normal and abnormal text messages to determine if a target message is abnormal, using hexadecimal octet sequences and first-order correlation probabilities to identify whether a text message is normal or abnormal, thereby eliminating the blind area in identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional identification methods based on report numbers are used, then the system is simple to operate, but the identification accuracy deteriorates when malicious users change phone numbers or when report numbers are low
Solution Approach 1:
The patent changes the identification parameter from report numbers to octet distribution characteristics of text messages. By analyzing the hexadecimal octet sequences and their distribution patterns, the system can identify abnormal messages based on their intrinsic statistical properties rather than external report counts, thereby maintaining high accuracy even when malicious users change phone numbers or when report numbers are low.
Solution Approach 2:
The patent replaces the mechanical counting system (report numbers) with a statistical analysis system (octet distribution characteristics). Instead of manually or mechanically counting reports, the system uses automated statistical analysis of message content patterns, substituting a simple counting mechanism with a more sophisticated but automated analytical approach.
2Reliability
If report-based identification is used, then the implementation is straightforward, but it creates blind areas in identification when malicious users change phone numbers
Solution Approach 1:
The patent creates a statistical model (copy) of normal message octet distribution patterns. By comparing actual messages against this pre-established model, the system can reliably identify deviations without needing to track each individual report or user history, thereby eliminating blind spots when users change phone numbers while maintaining implementation feasibility.
3Measurement precision
If analysis of octet distribution characteristics is performed, then identification accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent applies partial action by focusing analysis only on the octet distribution characteristics of message content rather than analyzing all possible message attributes. This selective approach achieves high identification accuracy by concentrating computational resources on the most discriminative features (octet patterns) while avoiding unnecessary analysis of other message properties.
Data Source
AI summary
One embodiment provides a method, including: receiving, at an information handling device, a first predetermined data characteristic and a second predetermined data characteristic; receiving, at the information handling device, text data comprising a third predetermined data characteristic; comparing, using a processor, the third predetermined data characteristic with the first predetermined data characteristic and the second predetermined data characteristic; and determining, based on the comparing, whether the third predetermined data characteristic is associated with the first predetermined data characteristic or the second predetermined data characteristic. Other aspects are described and claimed.