Likefarm Detection via Domain Confidence Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing reliance on 'like' manipulations by web services to artificially inflate popularity rankings leads to the proliferation of 'likefarm' sites, which compromise the actual popularity of web content and clutter user feeds with spam, making it difficult to distinguish genuine content from manipulative tactics.
Innovation Solution
A system that identifies and blacklists web domains associated with 'likefarm' entities by determining domain-level 'like' counts, using a decision tree function based on attributes like classification values, domain registration time, and blacklist status to calculate a 'likefarm' domain confidence score, thereby preventing these sites from being highly ranked in search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If web services use 'like' functionality to rank web pages, then popularity-based ranking improves user relevance, but likefarm manipulation compromises ranking accuracy
Solution Approach 1:
The patent introduces an intermediary analysis system that sits between the like-counting mechanism and the ranking algorithm. This intermediary component analyzes web documents to detect likefarm patterns and generates confidence scores that modify or filter the raw like counts before they influence ranking, thereby preserving the usefulness of popularity metrics while blocking manipulation
Solution Approach 2:
The system performs preliminary analysis of web documents to identify likefarm patterns before the like counts are used for ranking. By detecting suspicious patterns in advance and assigning low confidence scores to potentially manipulated content, the system prevents compromised popularity metrics from affecting ranking accuracy
2Measurement precision
If the system analyzes domain attributes to detect likefarms, then detection accuracy improves, but processing complexity increases
Solution Approach 1:
The patent segments the detection process into distinct components: document acquisition, domain attribute determination, like count aggregation, confidence score calculation, and blacklist management. Each component handles a specific aspect of the analysis, making the overall complex system more manageable and maintainable while preserving detection accuracy
3Reliability
If the system processes multiple domain attributes, then detection reliability improves, but computational resources increase
Solution Approach 1:
The system determines a subset of domain attributes that are most indicative of likefarm patterns rather than analyzing every possible attribute. By focusing on the most discriminative features (such as domain age, like count velocity, and document patterns), the system achieves high detection reliability while consuming fewer computational resources
Data Source
AI summary
A plurality of web documents that include indicators corresponding to one or more selectable like objects may be obtained. A corresponding web domain associated with each of the plurality of obtained web documents may be determined. A domain total like object count of the indicators corresponding to the one or more selectable like objects may be determined for each one of the obtained plurality of web documents, for each determined corresponding web domain. A candidate group of the corresponding web domains may be determined based on a comparison of a predetermined first threshold value with one or more of the domain total like object counts. A likefarm domain confidence score may be determined for each one of a second group of the corresponding web domains based on a decision tree function that is based on a plurality of domain attributes.