Trigger Query Apparatus Using Graph Analysis for Keyword Acquisition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Keyword-targeted advertising faces challenges in displaying advertisements due to the limited number of preset advertising keywords, making it difficult to obtain a comprehensive set of keywords that cover a certain category, which is labor-intensive for humans to achieve.

Innovation Solution

A trigger query obtaining apparatus that automatically generates a set of keywords for a category by creating a minimal directed graph from search logs, using score calculation based on link analysis to identify relevant queries, and outputting trigger queries that match a predetermined condition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If keyword-targeted advertising uses a small number of preset advertising keywords, then the device complexity is low, but the productivity of advertisement display is reduced

Engineering Contradiction:
Improveadvertisement display frequencyVSAvoidkeyword management complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system automatically generates advertising keywords by analyzing search query logs without requiring manual human intervention. The keyword generation unit extracts keywords from search logs and automatically creates advertising keyword sets, allowing the system to self-service the keyword management task that would otherwise require human effort

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical human effort of manually creating and managing keyword sets with an automated computational system. The keyword generation unit uses search log analysis and algorithmic processing to substitute human manual work in keyword management, transforming a labor-intensive process into an automated digital process

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If human effort is used to obtain a comprehensive set of keywords for a category, then the manufacturing precision of keyword selection is high, but the productivity is reduced

Engineering Contradiction:
Improvekeyword acquisition speedVSAvoidkeyword coverage accuracy
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The system replaces manual human analysis of search logs with automated computational processing. The keyword generation unit uses algorithms to analyze search query patterns, extract relevant keywords, and generate comprehensive keyword sets automatically, achieving both high speed and maintained accuracy through digital substitution of human effort

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system uses search log data as feedback to continuously improve keyword generation. By analyzing actual search queries submitted by users, the system learns which keywords are relevant and effective, then uses this feedback to refine and expand advertising keyword sets, maintaining accuracy while operating at automated speeds

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS9940408B2Trigger query obtaining apparatus, trigger query obtaining method, and non-transitory computer readable recording medium
Publication Date: 2018.04.10 YAHOO JAPAN CORP
  • US9940408B2 patent drawing
  • US9940408B2 patent drawing
  • US9940408B2 patent drawing

AI summary

A trigger query obtaining apparatus includes a graph creating unit, a score calculation unit, a trigger query obtaining unit, and an output unit. The graph creating unit creates a minimal directed graph having nodes each corresponding to a search query included in one of two search logs, and a link directed from one node to another node. The score calculation unit calculates a score for each node of a directed graph composed of minimal directed graphs, using a seed query. The trigger query obtaining unit obtains a trigger query that is a search query corresponding to a node having a score that satisfies a condition. The output unit outputs the trigger query.