Method and device for recognizing problem types and method and device for establishing recognition models

A technology of problem type and recognition model, which is applied in the field of computer network, can solve problems such as poor recognition accuracy, fuzziness, and inability to effectively distinguish, and achieve the effect of improving recognition accuracy and search accuracy

Active Publication Date: 2013-10-30
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF4 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Different question types may use the same interrogative word, so there is ambiguity about which question ty

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
  • Method and device for recognizing problem types and method and device for establishing recognition models
  • Method and device for recognizing problem types and method and device for establishing recognition models
  • Method and device for recognizing problem types and method and device for establishing recognition models

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] figure 1 Main method flow chart provided for the present invention, as figure 1 As shown, it mainly includes the following steps:

[0058] Step 101: Obtain a text segment containing interrogative words or demand words from a training corpus, wherein the training corpus includes a pre-classified question set.

[0059] Among them, if the question set in the training corpus is composed of question sentences containing interrogative words, the text fragments obtained in this step are text fragments containing interrogative words, and the model established in this case can be used to identify question types containing interrogative words , which will be specifically described in Embodiment 2. If the question set in the training corpus is composed of queries containing demand words, then the text fragments obtained in this step are text fragments containing demand words, and the model established in this case can be used for implicit question types containing interrogative ...

Embodiment 2

[0067] The problem type recognition model established in this embodiment is mainly aimed at the problem type recognition containing interrogative words, such as figure 2 As shown, the corresponding method for establishing a problem type identification model at this time includes the following steps:

[0068] Step 201: Create a training corpus using question sentences containing interrogative words, and these question sentences are pre-classified into types in the training corpus.

[0069] After a large number of question sentences that contain interrogative words can be marked as the type of question, as the training corpus, the types involved here can be large categories, such as: people, places, numbers, time, entities, descriptions (description includes methods, reasons, Definition, meaning, abbreviation, difference, expression, etc.), yes-wrong questions, choice questions, positive and negative questions, rhetorical questions, etc.; it can also be subcategories, for examp...

Embodiment 3

[0089] The question type identification model established in this embodiment mainly performs type identification for implicit questions that do not contain interrogative words, such as image 3 As shown, the corresponding method for establishing a problem type identification model at this time includes the following steps:

[0090] Step 301: Using the queries corresponding to the clicked questions in the search log to establish a training corpus, these queries are also pre-classified according to the types of the clicked questions corresponding to them in the training corpus.

[0091] During the user's search process, even if the query input sometimes does not contain interrogative words, it actually implies that the expression has the same intention as the question sentence containing interrogative words. For example, the user inputting "Yao Ming's height" actually expresses the same intention as "How much is Yao Ming's height". Based on this principle, the present invention...

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 provides a method and a device for recognizing problem types and a method and a device for establishing recognition models. During a process of establishing a question recognition model, whether a text fragment belongs to feature fragments of a class is judged on the basis of the absolute frequency of occurrence and the relative frequency of occurrence compared with other classes of the text fragment containing interrogatives or requirement words in the class, and accordingly the type to which the text fragment containing the interrogatives or the requirement words belongs can be positioned accurately, and the accuracy of recognition of question types can be improved. In addition, the question type recognition model can further be used for reclassifying training corpuses to serve as updated training corpuses, and the question type recognition model can be optimized gradually through this kind of iteration mode. If question type recognition and searching are performed on the basis of the mode, the searching accuracy can be further improved.

Description

【Technical field】 [0001] The invention relates to computer network technology, in particular to a method and device for identifying problem types, and a method and device for establishing a corresponding identification model. 【Background technique】 [0002] With the rapid development of computer technology, the network has gradually become the main means for people to obtain information. When people want to obtain information from the Internet, they often input search items (query) through search engines, knowledge platforms, etc., which requires search engines or knowledge platforms to identify user needs, and the identification of problem types lies in the identification of user needs. occupy an important position in. For example, when a user inputs a time-type query to a search engine, the search engine can rank the web pages containing the corresponding time attribute values ​​at the top of the search results. For another example, if the user enters "Yao Ming's height"...

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
IPC IPC(8): G06F17/30
Inventor 方高林
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD
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