Illegitimate Content Search Using Profile-Based Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Efficiently searching for illegitimate content on large content delivery sites is hindered by the time-consuming process of downloading and comparing numerous pieces of content, making it difficult to identify unauthorized uploads.
Innovation Solution
An illegitimate content search device and method that utilizes a query creation dictionary to detect and filter out illegitimate content by identifying profiles matching previously detected illegitimate content, prioritizing comparisons based on profile similarity and probability estimates, and using a previously-detected illegitimate content removing unit to streamline the search process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a search query is created based on the title of legitimate content and all possible content is downloaded and compared, then the search comprehensiveness is improved, but the search time becomes excessively long
Solution Approach 1:
The patent applies preliminary action by creating a query creation dictionary in advance that stores predetermined phrases and their associated weights. This dictionary is prepared before the actual illegitimate content search, allowing the system to quickly filter and prioritize search results without performing exhaustive comparisons of all possible content, thus reducing search time while maintaining comprehensiveness.
Solution Approach 2:
The patent segments the search process into multiple stages: first using the query creation dictionary to filter and score potential matches, then performing detailed comparisons only on high-scoring candidates. This segmentation divides the overwhelming task of comparing all possible content into manageable phases, improving both efficiency and comprehensiveness.
2Measurement precision
If the number of content comparisons is increased to ensure accurate detection, then the detection accuracy is improved, but the processing time increases significantly
Solution Approach 1:
The patent changes the parameter of content selection by introducing a scoring mechanism based on phrase matching weights. Instead of uniformly processing all content, the system assigns scores and selectively processes only those contents that exceed a threshold, thereby maintaining high detection accuracy while dramatically reducing the number of comparisons needed.
Solution Approach 2:
The patent applies partial action by performing complete detailed comparisons only on a small subset of high-probability candidates identified by the preliminary scoring stage. For lower-probability candidates, the system uses lighter-weight filtering, thus achieving sufficient detection accuracy without the excessive processing time that would result from uniform exhaustive comparison.
3Measurement precision
If all possible content is downloaded for comparison, then the completeness of illegitimate content detection is improved, but the resource consumption increases
Solution Approach 1:
The query creation dictionary serves as a preliminary filtering mechanism that identifies and prioritizes relevant content before the main detection process. This preliminary action reduces the volume of data that needs to be downloaded and processed, while the multi-stage approach ensures that detection completeness is maintained for the prioritized subset.
Solution Approach 2:
The patent extracts and utilizes specific characteristic phrases from legitimate content titles to create the query creation dictionary. By taking out these key identifying phrases and using them for targeted filtering, the system reduces the overall data processing volume while maintaining the ability to detect illegitimate content that contains these characteristic phrases.
Data Source
AI summary
The present disclosure relates to a search for illegal (illegitimate) content In particular, the present disclosure provides the search in an efficient manner by excluding pieces of presumed illegitimate content from the candidates of illegitimate content for comparing with legitimate content. The illegitimate content comprises content posted by an unauthorized user without a legitimate ownership of the content. An illegitimate content sensing device according to the present disclosure comprises a presumed illegitimate content deletion part for, on the basis of a profile of candidate illegitimate content, identifying pieces of the candidate illegitimate content having a profile which is within a degree of identicalness to a profile of previously identified illegitimate content as presumed illegitimate content.