Notification Message Authorship Classification via Linguistic Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer systems face challenges in efficiently distinguishing between genuine and misleading notification messages, particularly from scareware applications, which can deceive users by generating fictitious threats and demanding software purchases.
Innovation Solution
The method involves linguistic analysis techniques such as syntax parsing, stop word analysis, and character language modeling to classify notification messages by comparing their characteristics with those of known genuine and misleading messages, determining the authorship and authenticity of the messages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If linguistic analysis procedures are performed on notification messages to determine authorship, then the accuracy of detecting malicious messages is improved, but the complexity of the system increases
Solution Approach 1:
The system segments the analysis process into three distinct linguistic analysis procedures: syntax parsing, stop word usage analysis, and character language modeling. Each procedure independently analyzes different aspects of the notification message and generates separate results, which are then combined to determine authorship. This segmentation allows the complex analysis task to be divided into manageable components while maintaining high detection accuracy.
Solution Approach 2:
The message model serves as a universal database that stores characteristics from multiple sources (genuine notification messages and misleading notification messages). This single message model is used to compare against all unclassified notification messages, providing a multi-functional solution that handles both detection and classification tasks without requiring separate systems for each function.
2Reliability
If multiple linguistic analysis procedures are applied to each notification message, then the reliability of classification is improved, but the processing time increases
Solution Approach 1:
The system performs preliminary action by pre-processing genuine and misleading notification messages to extract and store their linguistic characteristics in the message model before actual detection begins. Syntax parsers, stop word lists, and language models are trained and stored in advance. When a new notification message arrives, the system only needs to compare it against the pre-computed message model, significantly reducing real-time processing time while maintaining high reliability through the use of multiple analysis procedures.
Data Source
AI summary
A computer-implemented method for determining authorship of an unclassified notification message is described. An unclassified notification message with one or more target sentences is received. A message model based on one or more classified notification messages stored in a data storage device is retrieved. One or more linguistic analysis procedures are performed on the one or more target sentences. Results of the one or more linguistic analysis procedures are compared with one or more characteristics of the message model. The unclassified notification message is classified based on the results of the comparison.