A problem routing method based on word vector

A technology of word vectors and questions, applied in the field of question routing based on word vectors, can solve problems such as reducing accuracy, and achieve the effects of reducing noise, high data efficiency, and improving accuracy

Active Publication Date: 2018-01-23
DALIAN UNIV OF TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Such methods improve the recall rate of retrieval to a certain extent, but often reduce the accuracy due to the introduction of a large amount of noise information

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 problem routing method based on word vector
  • A problem routing method based on word vector
  • A problem routing method based on word vector

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] The present invention will be described below in conjunction with the accompanying drawings.

[0048] Such as figure 1 As shown, a word vector-based question routing method includes the following steps:

[0049] Step 1. Construction of user profiles: Build profiles for users based on their answer history. Users who have answered questions in the community Q&A are candidate answerers for a new question. The construction of each user profile includes the following sub-steps:

[0050] Step (a), download all the data from July 2008 when the website was established to March 2014 from the URL provided by the Stackoverflow website, the data is in XML format, read the XML file format to extract all questions, including questions Label tag domain, title title domain and content body domain;

[0051] Step (b), collecting those questions selected as the best answers among the questions answered by the user to form the user's file;

[0052] Step (c), ignoring the users who are s...

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 relates to a question routing method, in particular to a question routing method based on word vectors. The question routing method based on the word vectors includes the following steps that firstly, user files are established; secondly, data are preprocessed; thirdly, the word vectors are trained; fourthly, document vectors are expressed; fifthly, user authority is worked out; sixthly, user activity is worked out; seventhly, candidate answerers are ranked. The data are trained through word2vec, efficiency is high, the word vectors have the superposition property, and the defect that no co-occurrence word with the similarity of 0 exists among documents is overcome; meanwhile, document subject words are extracted, the document vectors are expressed by the word vectors, the authority, the activity and the similarity among the document vectors are worked out comprehensively, semantic information among the documents is considered, and noise is also reduced. Contrast experiments are carried out between the question routing method and classic TF_IDF and between the question routing method and a classic Language Model, and S@N of the question routing method is higher than those of the other two methods.

Description

technical field [0001] The present invention relates to a question routing method, more specifically, to a question routing method based on word vectors. Background technique [0002] Q&A communities in recent years, such as Yahoo! Answers, Baidu Zhizhi, and Stackoverflow have become more and more popular. The public shares knowledge in the community, and a large number of users ask questions every day. The content of the answers in the community provides users with optional answers. Generally, community Q&A websites will classify questions according to question categories. When a user asks a question, they will select an appropriate category, that is, a question label, and wait for other users to answer. The questioner must wait for other users to browse the community and read the question before they can provide an answer, and the best answer may be obtained after many users answer. The answer may already be moot to the questioner. On the other hand, if the user is an e...

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): G06F17/30
CPCG06F16/24578G06F16/9535G06Q50/01
Inventor 王健董华磊林鸿飞
Owner DALIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap