Neural network-based dialogue semantic intention prediction method and learning training method

A neural network and neural network model technology, applied in the field of dialogue semantic intent prediction, learning and training, can solve problems such as the inability to use real-time experience online for autonomous learning, the lack of deep semantic information related to tasks, and the inability to obtain initialization vector representations, etc., to achieve high The effect of semantic intent prediction, saving time and economic cost, and improving the accuracy of description

Inactive Publication Date: 2018-08-03
北京十三科技有限公司
View PDF3 Cites 52 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] The LSTM neural network model needs to rely on the original word vector input. For the semantic intent recognition task in chat robots, it is difficult to obtain large-scale corpus for training word vectors (especially human-machine dialogue). initialization vector representation
[0012] The LSTM neural network model only takes each word vector of a sentence (instead of each sentence text in several consecutive sentences) as input, and inputs it into the corresponding Cell unit in time series, lacking deep semantic information related to the task, so , the model does not fully characterize the chat text
[0013] The LSTM ne

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
  • Neural network-based dialogue semantic intention prediction method and learning training method
  • Neural network-based dialogue semantic intention prediction method and learning training method
  • Neural network-based dialogue semantic intention prediction method and learning training method

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0066] Example one

[0067] The present invention provides a "Q-LSTM neural network-based chat robot dialogue semantic intention prediction method", and its model framework is as attached figure 2 As shown, the design of modules including input vector generation, Q-LSTM neural network, random intent selection, trial dialogue generation, human dialogue collection, intent matching verification and input state update, etc. The design steps of each module are as follows:

[0068] Step 1. Input vector generation module design.

[0069] Perform word segmentation and other preprocessing on a large data set (labeled corpus) collected in advance;

[0070] Generate vocabulary for non-repeated words in the statistical corpus;

[0071] At a certain time t, the sentence text of a certain length I in a certain time segment τ of the human interlocutor (said or just said) is expressed in one-hot vector space according to the vocabulary to form a sentence matrix S τ , As a certain group of inputs of t...

Example Embodiment

[0148] Example two

[0149] The learning and training process of the "Q-LSTM Neural Network-based Semantic Intention Prediction Method for Chat Robot Dialogue" provided by the present invention is as follows Figure 5 As shown, the specific steps are:

[0150] Step 1. Collect and sort out the conversation data of human chat sequence.

[0151] According to the service field that the chatbot is to be applied to, through some public social platforms, a large number of serial sample data of actual conversations between people and machines are collected, and after basic semantic processing, a conversation database is established.

[0152] Step two, sequence dialogue data preprocessing and semantic intention marking.

[0153] For the database generated in step (1), first use some keywords and template rules to preprocess the data to filter out some unintentional data; then perform semantic preprocessing and semantic intent tagging of sequence dialogue data samples. The specific steps are as f...

Example Embodiment

[0177] Example three

[0178] Taking the intelligent customer service (chat robot) of real estate intermediary as the application background, based on the chat record data of customer-person and human-machine conversations of a large domestic real estate agency in a certain city, according to the implementation of the present invention as described above A "Q-LSTM Neural Network-based Semantic Intention Prediction Method for Chat Robot Dialogue", the specific steps are as follows:

[0179] Step 1. Collect and sort out the conversation data of human chat sequence. Using the customer service platform of a large domestic real estate agency in a certain city, we collected two months of human-human and human-machine conversations and chat records about 30,000 sets of sequence sentences, each of which contains no more than 10 sentences. , Each sentence includes no more than 30 words, and after basic semantic processing, a dialogue database is established.

[0180] Step two, sequence dial...

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 a neural network-based chat robot dialogue semantic intention prediction method and a learning training method. The prediction method is used for a chat robot to predict a possibly continuous dialogue intention or point of interest (one of multiple dialogue intentions or points of interest) of a dialogist according to a sequence (said) statement eigenvector of the human dialogist; and by heuristically selecting an output dialogue of the chat robot and taking an answer of the dialogist as a verification feedback of a previous intention prediction result, online reinforcement learning and self-updating of an algorithm are realized, the accuracy of dialogue intention prediction is continuously improved, the round number of chat can be increased, and the interest of thedialogist can be enhanced.

Description

technical field [0001] The present disclosure relates to the technical field of semantic analysis, in particular to a neural network-based dialog semantic intent prediction method and a learning and training method. Background technique [0002] Through chatbots, people communicate with computers in natural language. In the process of chatting with the robot, a large amount of data containing rich user information is generated, which contains the user's potential needs and intentions. Intent recognition aims to judge whether the text or performance behavior posted by a user has a certain directional intention (such as consumption, leisure, knowledge, etc.), and accurately identifying user intentions can make chatbots more intelligent and enhance user experience; at the same time , and can also recommend consumer products in a targeted manner to better serve users. [0003] Conventional intent recognition uses template-matching-based methods and guided classification-based ...

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): G06F17/27G06F17/30G06N3/08
CPCG06N3/08G06F16/3329G06F40/30
Inventor 张宇
Owner 北京十三科技有限公司
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