Intent Mining via Seed Phrase Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current intent mining processes are complex due to multiple modes of expressing intent and lack a well-defined vocabulary, making it difficult to analyze web user reviews effectively.
Innovation Solution
A method involving a preliminary search using seed phrases to generate intent results, followed by producing action search strings around action verbs and applying them to non-constrained sources to identify and analyze user intent, leveraging constrained sources to build patterns that can be generalized for broader data sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional intent mining methods are used to analyze web user reviews, then the analysis can be performed on general text data, but the process becomes complex due to multiple modes of expressing intent and lack of well-defined vocabulary
Solution Approach 1:
The patent applies preliminary action by first performing a constrained source search using seed phrases to generate a controlled set of preliminary search results before expanding to broader analysis. This preliminary structuring of intent expressions using predefined seed phrases and constrained sources simplifies the subsequent mining process by establishing a foundation of known intent patterns before tackling the complexity of general web user reviews
Solution Approach 2:
The patent segments the intent mining process into distinct phases: constrained source search with seed phrases, identification of intent expressions, generation of action search strings, and application to non-constrained sources. This segmentation breaks down the complex task of analyzing multiple intent expression modes into manageable steps, reducing overall process complexity while maintaining adaptability
2Measurement precision
If constrained sources are used to build intent patterns, then the precision of intent identification is improved, but the quantity of available data for analysis is reduced
Solution Approach 1:
The patent implements nesting by first conducting analysis on constrained sources to build precise intent patterns, then embedding these patterns within a broader analysis framework that applies to non-constrained sources. The constrained source results serve as a nested foundation that is subsequently expanded to larger data sets, allowing precision to be preserved while scaling to greater data quantities
Solution Approach 2:
The patent transitions from two-dimensional analysis (constrained sources only) to three-dimensional analysis by adding the layer of non-constrained sources. The constrained source patterns serve as a base layer that is then applied across additional dimensions of data, enabling both precision maintenance and data quantity expansion through dimensional expansion of the analysis framework
Data Source
AI summary
A method for intent mining is provided. The method includes performing a preliminary search of a constrained source using one or more seed phrases to generate multiple preliminary search results representing different ways of expressing a desired intent. The method also includes identifying each of the plurality of preliminary search results that have expressed the desired intent to generate a plurality of intent results. The method also includes producing multiple action search strings around one or more action verbs in each of the multiple intent results. The method further includes applying each of the multiple action search strings on one or more non-constrained sources to generate multiple action search results.