Map Spam Detection via Category Density and Data Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Businesses engage in fraudulent marketing tactics known as 'map spam' by creating multiple listings in geographic areas where they do not actually operate, misleading customers and causing financial loss.
Innovation Solution
A computer-implemented method that assesses the likelihood of map spam by considering factors such as business category density, shared identifying data, and location characteristics, assigning a 'spam score' to determine if a business listing is likely to be map spam and adjusting this score based on thresholds and specific geographic conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple business listings are created in different geographic locations to attract more customers, then the quantity of search results increases, but the reliability of the search results decreases due to fraudulent listings
Solution Approach 1:
The system performs preliminary verification of business listings before they are included in search results. It calculates spam scores based on multiple factors including geographic distribution analysis, category density comparisons, and identifying data matching across listings. Listings that exceed spam thresholds are filtered out in advance, ensuring that only reliable listings appear in search results while maintaining high quantity of legitimate results
Solution Approach 2:
The system implements a feedback mechanism where search results are continuously monitored and evaluated. By analyzing patterns in listing distributions, category densities, and identifying data across multiple listings, the system dynamically adjusts spam scoring and filtering criteria to maintain reliability while preserving the quantity of legitimate search results
2Reliability
If spam detection filters are applied to search results, then the reliability of search results improves, but the productivity decreases due to loss of legitimate listings
Solution Approach 1:
The system applies differentiated spam detection criteria to different geographic regions and business categories. By analyzing local characteristics such as regional business density patterns, category-specific distribution norms, and location-based verification, the system tailors filtering thresholds to each context. This prevents over-filtering of legitimate listings in regions or categories where high density is normal, while maintaining strict filtering where spam patterns are prevalent
Solution Approach 2:
The system dynamically adjusts spam score thresholds and detection parameters based on observed patterns in the data. By monitoring metrics such as category density deviations, geographic distribution anomalies, and identifying data repetition rates, the system adapts its filtering criteria to maintain optimal balance between reliability and productivity, ensuring legitimate listings are not incorrectly filtered
3Measurement precision
If business category density thresholds are enforced to detect map spam, then the measurement precision of spam detection improves, but the device complexity increases due to multiple verification factors
Solution Approach 1:
The system segments the spam detection process into distinct modular components: geographic distribution analysis, category density comparison, identifying data matching, and spam score calculation. Each module independently evaluates specific aspects of listings and contributes to the overall spam score. This segmentation improves measurement precision by thoroughly examining multiple factors while managing complexity through modular, independent processing units
Data Source
AI summary
A determination of whether a mapped business listing that is produced as a search result corresponds to an actual location of operation is based on different factors. One factor identifies whether the business listing is associated with a business category that appears as search results for a particular geographic area in numbers that exceed average proportions for the same business category density in similarly situated geographic areas. Another factor determines whether different business listings in the same geographic area include the same identifying data. Specific characteristics of a neighborhood where the business listing is mapped provide an additional factor for identifying whether a search result for a business listing is map spam. The different factors may be considered together to determine the likelihood that a mapped search result is spam.