Supercharge Your Innovation With Domain-Expert AI Agents!

A Conversational Information Acquisition Method Based on Hybrid Drive

An acquisition method and hybrid-driven technology, applied in digital data information retrieval, instruments, calculations, etc., can solve the problem of no guarantee, does not take into account the high degree of overlap between the redundancy problem and the concise problem, and cannot better pay attention to the redundancy Problems such as the difference between redundant problems and concise problems, to achieve the effect of improving model performance

Active Publication Date: 2022-03-22
SHANDONG UNIV
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, they cannot guarantee that the generated questions will contain the information the user needs, that is, they cannot guarantee the validity of the questions
2. The second type of method is the retrieval method
However, they did not consider the high overlap between the redundant problem and the concise problem, but treated each word equally, which makes the generator always tend to generate some easy words, such as in the original redundant problem Words appearing, could not better focus on the difference between redundant and brevity problems

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Conversational Information Acquisition Method Based on Hybrid Drive
  • A Conversational Information Acquisition Method Based on Hybrid Drive
  • A Conversational Information Acquisition Method Based on Hybrid Drive

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] In order to facilitate understanding of the present invention, the present invention will be described in more detail below with reference to the accompanying drawings and specific embodiments. Preferred embodiments of the present invention are shown in the accompanying drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described in this specification. Rather, these embodiments are provided so that a thorough and complete understanding of the present disclosure will be provided.

[0036] The present invention provides a method for acquiring conversational information based on hybrid drive. The process is as follows: figure 1 shown, including the following steps:

[0037] Step 1: Build the rewrite model

[0038] Paraphrase model consists of two modules: Paraphrase module and Phrase module. The rewriting module is used to determine which rewriting method is used for the current word, mark it with a re...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the field of information acquisition, and in particular relates to a dialog information acquisition method based on a hybrid drive. A conversational information acquisition method based on a hybrid drive, comprising the following steps: constructing a rewriting model; using a dynamic programming sampling-training mechanism to train the parameters of the model; using the trained model, inputting redundant questions and rewriting it, Output concise questions. Based on the fact that the modification process has no labels and cannot supervise learning, the present invention creatively proposes a sampling-training method based on dynamic programming. Compared with the traditional reinforcement learning sampling method, such sampling-training can consider the internal relationship between labels. Encourage the model to sample better labels and improve model performance.

Description

technical field [0001] The invention belongs to the field of information acquisition, and particularly relates to a hybrid-driven-based conversational information acquisition method. Background technique [0002] Conversational information acquisition, that is, the traditional retrieval-based information acquisition process is more convenient to use through the man-machine dialogue interface. At this time, the process of human-machine dialogue can not only be used to meet the information needs of users to a great extent, but also improve user satisfaction. At present, the related work of conversational information acquisition has already produced preliminary products, and it has brought great convenience to human life. For example: conversational products in the form of virtual assistants (Apple Siri and Microsoft Cortana), conversational products in the form of small talk (Microsoft Xiaoice), and conversational products in the form of smart speakers (Amazon Echo, Baidu Xia...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F3/16G06F16/332G06F40/216
CPCG06F3/16G06F16/3329G06F40/216
Inventor 陈竹敏刘中坤任鹏杰任昭春
Owner SHANDONG UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com