Problem classification method and system for artificial intelligence customer service robot
A technology of problem classification and artificial intelligence, applied in the direction of instruments, special data processing applications, electrical digital data processing, etc., can solve the problem that the classification accuracy rate cannot better meet the industrial application, so as to improve the classification accuracy rate and prevent the gradient from disappearing Or gradient explosion, the effect of increasing the number of neural network layers
Inactive Publication Date: 2017-08-15
上海携程国际旅行社有限公司
View PDF3 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] The technical problem to be solved by the present invention is to overcome the defect that the classification accuracy of artificial intelligence customer service robots in the prior art cannot better meet the needs of industrial applications in problem classification, and to provide a problem classification method for artificial intelligence customer service robots and system
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0024] The present invention is further illustrated below by means of examples, but the present invention is not limited to the scope of the examples.
[0025] like figure 1 As shown, the problem classification method of the artificial intelligence customer service robot provided in this embodiment includes the following steps:
[0026] Step 101, receiving the entire question information.
[0027] In this step, the artificial intelligence customer service robot receives the question entered by the user, that is, the entire question information, and is ready to classify the question.
[0028] Step 102, perform dynamic convolution on the entire question information based on the CNN model, and extract several main feature information.
[0029] In this step, on the basis of the CNN model, the number of neural network layers is increased, and the original static convolution kernel is changed to a dynamic convolution kernel.
[0030] Step 103 , perform down-sampling processing, a...
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
Login to View More Abstract
The invention discloses a problem classification method and system for an artificial intelligence customer service robot. The problem classification method comprises the following steps that: S1: receiving a whole section of problem information, carrying out dynamic convolution on the whole section of problem information, and extracting multiple pieces of major characteristic information; S2: carrying out downsampling processing on the multiple pieces of major characteristic information, and extracting multiple pieces of target characteristic information from the multiple pieces of major characteristic information; and S3: according to the multiple pieces of target characteristic information, on the basis of an artificial intelligence classification algorithm, outputting problem type information. By use of the problem classification method and system for the artificial intelligence customer service robot, classification accuracy is greatly improved, the requirement of industrial application is favorably met, and the problem of gradient vanishing or gradient explosion can be favorably avoided.
Description
technical field [0001] The invention relates to the field of artificial intelligence, in particular to a problem classification method and system for an artificial intelligence customer service robot. Background technique [0002] There are many classification networks commonly used in the field of artificial intelligence, such as SVM (Support Vector Machine, support vector machine), Bayes (Bayesian classification algorithm), CNN (Convolutional Neural Network, convolutional neural network), RNN (Recurrent Neural Networks, cycle neural network), etc. However, in practical applications, especially in the field of text classification, the classification accuracy cannot better meet the needs of industrial applications. Taking the multi-classification problem as an example, usually in the case of large-scale corpus, even if a deep model such as CNN or RNN is used, the accuracy rate is low and it is difficult to further improve it. Contents of the invention [0003] The techni...
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
Login to View More IPC IPC(8): G06F17/30
CPCG06F16/3329G06F16/353
Inventor 李健于天池田志国
Owner 上海携程国际旅行社有限公司