Chat UI Question Guidance for Accurate FAQ Target Text Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional FAQ search services often fail to provide desired search results due to overwhelming hits from keywords with high generality, and chatbots struggle to accurately reach target texts, making it difficult for users to obtain appropriate answers.
Innovation Solution
An information processing device that utilizes a chatbot to assist users through a chat-type UI, combining a guidance sentence dictionary and intent prediction search to present relevant question candidates and target texts, enhancing the conventional intent prediction search with machine learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional FAQ search services execute a search with a keyword that a typical user can come up with, then the search covers a wide range of possible questions, but an overwhelming number of search results are displayed making it difficult to reach the desired target text
Solution Approach 1:
The patent segments the search results by presenting multiple candidate questions (guidance sentence candidates) derived from the user's input keyword. Instead of displaying all matching FAQ items at once, the system divides them into ranked candidates and presents them sequentially, allowing users to navigate through segmented results to find the desired target text efficiently.
Solution Approach 2:
The system performs preliminary action by pre-generating multiple guidance sentence candidates from the user's keyword input before the user completes their search query. These candidates are prepared in advance based on the guidance sentence dictionary, so when the user selects a candidate, the target text is already identified and can be presented immediately, reducing the time to reach the desired information.
2Extent of automation
If a conventional chatbot is simply utilized, then the system can process user inputs, but it is difficult for the user to reach the desired target text or the like
Solution Approach 1:
The patent introduces an intermediary component (the guidance sentence candidate presentation mechanism) between the chatbot and the user. The chatbot generates candidates, but instead of directly answering or searching, the system presents multiple candidate questions for user selection. This intermediary step ensures that the automated processing aligns with user intent, improving the accuracy of reaching the desired target text while maintaining automation.
3Ease of operation
If multiple guidance sentence candidates are presented to the user, then the user can more easily find the desired question, but the system complexity increases
Solution Approach 1:
The patent applies universality by using a single guidance sentence dictionary that serves multiple functions: it stores pre-defined question patterns, generates candidate questions from keywords, and provides the basis for matching user inputs to target texts. This multi-functional dictionary reduces the need for separate complex modules for each function, thereby managing system complexity while enabling multiple guidance sentence candidates to be presented.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention makes it possible for a user to easily find a desired target text or the like through a chat and obtain an appropriate answer based on the target text or the like. A question sentence candidate extraction unit 103 extracts one or more question sentence candidates including, in at least a part thereof, a keyword accepted by a keyword or the like acceptance unit 102. A question sentence candidate presentation unit 104 presents the extracted one or more question sentence candidates to a user. A related text or the like presentation unit 106 extracts, from a question sentence dictionary DB 181, a target text or the like associated with one selected from among the one or more question sentence candidates, and presents the extracted target text or the like to the user. A chat type UI control unit 109 controls transfer of information between the keyword or the like acceptance unit 102 to the related text or the like presentation unit 106 and a chatbot, thereby causing the chatbot to generate an answer based on the related text or the like to be presented to a text inputted by the user.