Illegitimate Content Search via Query Dictionary Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing illegitimate content search methods are ineffective in identifying illegitimate content posted by unauthorized users, as they rely on titles that may be modified to appear relevant to legitimate content, leading to missed detections.

Innovation Solution

An illegitimate content search device and method that generates a query creation dictionary by extracting relevant phrases from illegitimate content titles, which are then used to create search queries, allowing for a more accurate identification of illegitimate content by comparing with legitimate content profiles.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If search queries are created using phrases from legitimate content titles, then the search can identify obviously illegitimate content, but unauthorized users can modify titles to evade detection

Engineering Contradiction:
Improvedetection accuracyVSAvoidevasion capability
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary actions by collecting and analyzing illegitimate content titles in advance to extract characteristic phrases and patterns. These pre-extracted features are stored and used to build detection models before actual detection occurs, enabling the system to anticipate and identify evasion techniques without relying solely on legitimate content titles

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms by continuously collecting detection results and illegitimate content samples, analyzing new evasion patterns, and updating the query creation dictionary and detection models accordingly. This closed-loop feedback enables the system to adapt to emerging evasion techniques and improve detection accuracy over time

Inventive Principle:
Principle #23Feedback

2Measurement precision

If the search system analyzes more content features to improve detection, then detection accuracy improves, but system complexity increases

Engineering Contradiction:
Improvecontent analysis accuracyVSAvoidsystem structure complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the content analysis process into distinct modules: title analysis, phrase extraction, pattern recognition, and comparison operations. Each module handles a specific aspect of detection, making the overall complex system manageable and maintainable while achieving comprehensive analysis through coordinated simple components

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces intermediary structures such as the query creation dictionary and feature extraction modules that act as mediators between raw content and detection logic. These intermediaries pre-process and structure data, simplifying the main detection algorithm while enabling sophisticated multi-feature analysis

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS12061584B2Illegal content search device, illegal content search method, and program
Publication Date: 2024.08.13 NIPPON TELEGRAPH & TELEPHONE CORP
  • US12061584B2 patent drawing
  • US12061584B2 patent drawing
  • US12061584B2 patent drawing

AI summary

An objective of the present invention is to make a search for illegal (illegitimate) content more efficient. The illegitimate content relates to content posted by an unauthorized user without a legitimate ownership of the content. An illegitimate content search device according to the present invention comprises: a query generation dictionary storage part for storing a query generation dictionary for generating a search query used in searching for illegitimate content; a query generation dictionary creation part for extracting, from a title of illegitimate content, associated wording being associated with wording included in a title of legitimate content, and registering the extracted associated wording in the query generation dictionary for generating the search query used in searching for the illegitimate content; and a search query generation part for generating the search query using the associated wording registered in the query generation dictionary.