Intention recognition method and device

A recognition method and intent technology, applied in text database query, unstructured text data retrieval, instruments, etc., can solve the problems of lack of information, low accuracy of intent recognition results, etc., achieve fast recognition speed, improve accuracy and efficiency high effect

Active Publication Date: 2020-08-21
ALIBABA GRP HLDG LTD
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the selected feature words are separated from the original sentence, resulting in a large amount of information loss. For example, the same word with different part-of-speech marks expresses different meanings, and the same word may also be in different positions in the sentence. have different meanings, so that the meaning of the original text cannot be correctly understood, resulting in low accuracy of intent recognition results

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
  • Intention recognition method and device
  • Intention recognition method and device
  • Intention recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0024] This embodiment provides an intent recognition method, which can be used in the data preparation stage, is a part of feature engineering, and is mainly used for the purpose of simplifying features and extracting key effective information. Specifically, as figure 1 As shown, the method includes the following steps:

[0025] Step 100: Perform symbol processing and word segmentation processing on the text data fed back by the user to obtain segmented sentence fragments.

[0026] Wherein, the text data may be text data fed back by the user in a specific application scenario; the symbol processing is used to remove symbols irrelevant to semantics in the text data, such as emoticons, exclamation marks, etc. It is beneficial to reduce recognition noise; the word segmentation process is used to segment the entire sentence into sentence segments, and specifically, word segmentation can be performed according to an existing word segmentation algorithm.

[0027] For example, a p...

Embodiment 2

[0065] This embodiment provides an intention recognition device, such as Figure 4 As shown, it includes: a word processing module 10 , a part-of-speech tag labeling module 20 , a part-of-speech tag replacement module 30 , a redundant deletion module 40 , a feature generation module 50 and an intent recognition module 60 . The device has relatively small requirements for hardware resources, and can run on a single machine locally or on a cloud server. Specifically, its working principle is as follows:

[0066] When receiving the text data of user feedback, carry out symbolic processing and word segmentation processing to text data by word processing module 10 and obtain the sentence segment after segmentation; Then, the part-of-speech tag replacement module 30 replaces the part-of-speech tag of the sentence segment included in the preset specific language dictionary in the sentence segment with a specific part-of-speech tag. Specifically, when performing replacement, the par...

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 discloses an intention recognition method and device, and the method is characterized in that the method comprises the steps: carrying out the symbol processing and word segmentation oftext data to obtain segmented statement segments; marking part-of-speech tags for the statement fragments; replacing the part-of-speech tags of the statement segments contained in a preset specific language dictionary in the statement segments with specific part-of-speech tags; deleting redundant fragments in the statement fragments according to the specific part-of-speech marks; labeling positioninformation on the remaining statement fragments after the redundant fragments are deleted to generate text features; and performing intention recognition on the text features. By labeling the part-of-speech marks and the position information, the problem of information loss caused by word segmentation processing is solved, so that the meaning of an original text can be correctly understood, andthe accuracy of an identification result is improved. Manual participation is not needed, so that the recognition speed is high, and the efficiency is high.

Description

technical field [0001] The invention relates to an intention recognition method and device, belonging to the technical field of artificial intelligence. Background technique [0002] Intent recognition technology refers to the technology that analyzes the text data fed back by the user to identify the user's intent, and the intent refers to what the user wants to express or what he wants to do. For example, in the application scenario of a customer service robot, the user enters a piece of text and wants to ask the price of a product. The background program uses the intent recognition technology to identify the user's intention, and finds the corresponding product price and returns it to the user as an answer. In this way, the user and the customer service Bots enable continuous conversations. [0003] The existing intent recognition technology mainly adopts the following two methods: [0004] Approach 1: Rule-Based Recognition [0005] This method needs to manually summa...

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 Applications(China)
IPC IPC(8): G06F16/33G06F40/253G06F40/289
CPCG06F16/3344
Inventor 武东旭
Owner ALIBABA GRP HLDG 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