Fraudulent Software Promotion Detection via Comment Entropy Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies lack an effective method to detect fraudulent software promotion, specifically 'review fraud' and 'download number fraud', which misleads users into downloading unwanted software applications, leading to adverse social effects.
Innovation Solution
A method and system that utilize statistical analysis based on comment information and download data to calculate average similarity, information entropy, and comment-to-download ratios, defining determination threshold ranges using probability statistical distribution parameters to identify fraudulent software promotions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If download quantity and reputation indicators are used to guide user selection, then user convenience and software selection efficiency are improved, but fraudulent promotion activities increase leading to reduced reliability
Solution Approach 1:
The patent performs preliminary detection of fraudulent promotion before users make selection decisions. By pre-calculating fraud probabilities based on comment information analysis and comparing them against threshold ranges, the system proactively identifies suspicious software, preventing users from being misled by fraudulent promotions in the first place
Solution Approach 2:
The patent introduces an intermediary detection mechanism that sits between the software promotion system and user selection. This intermediary layer analyzes comment information, calculates similarity metrics, determines fraud probabilities, and provides fraud warnings to users, thereby mediating the trust relationship without requiring users to directly evaluate promotion authenticity
2Productivity
If automated detection methods are implemented to identify fraudulent promotion, then detection efficiency and productivity are improved, but system complexity increases
Solution Approach 1:
The patent implements a self-service detection system where the software evaluation platform automatically collects comment information, calculates similarity metrics, determines fraud probabilities, and generates detection results without requiring external intervention. The system serves itself by using its own resources (comment data, processing capabilities) to detect fraud, eliminating the need for complex external detection infrastructure
Solution Approach 2:
The patent transforms the detection problem into a parameter-based probability calculation. By converting qualitative fraud assessment into quantitative metrics (similarity values, information entropy, fraud probability scores, threshold range comparisons), the system simplifies complex fraud detection into manageable parameter transformations that can be automatically processed
3Measurement precision
If detailed comment information analysis is performed to detect review fraud, then measurement precision is improved, but loss of time increases due to processing large volumes of comment data
Solution Approach 1:
The patent extracts only the essential features from comment information that are relevant to fraud detection. Instead of analyzing every aspect of comments, the system focuses on extracting key characteristics (comment text, timing patterns, similarity metrics) that directly indicate fraud, discarding irrelevant information to reduce processing time while maintaining detection accuracy
Solution Approach 2:
The patent uses similarity comparison by creating simplified representations (copies) of comment information. Instead of analyzing original complex comment data from scratch, the system creates similarity metrics that copy essential characteristics and compares these against known fraudulent patterns, significantly reducing processing time while preserving detection precision
Data Source
AI summary
A method, system and apparatus for detecting fraudulent promotion of a software application, the method including acquiring comment information associated with a software application, the comment information including one or more comments; calculating an average similarity or average information entropy of the comment information, wherein the average similarity is calculated based on a similarity between the one or more comments and wherein the average information entropy is calculated based on an information gain between the one or more comments; defining a determination threshold range for a category associated with the software application, wherein the determination threshold range is defined by a plurality of probability statistical distribution parameters based on average similarities or average information entropies of other applications associated with the category; and identifying the software application as fraudulent if the average similarity or average information entropy of the comment information is within the determination threshold range.