Unlock instant, AI-driven research and patent intelligence for your innovation.

User complaint prediction method and device

A prediction method and user technology, applied in the direction of prediction, neural learning methods, data processing applications, etc.

Inactive Publication Date: 2021-02-05
GCI SCI & TECH +1
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For the remaining unlabeled data, the user may feel that the network quality is poor, but choose not to complain due to time or other factors; it may be that the user feels that the network quality meets expectations, and there is no need to complain

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
  • User complaint prediction method and device
  • User complaint prediction method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] see figure 1 , the embodiment of the present invention provides a user complaint prediction method, including step S1-step S7:

[0036] S1. Relabel the sample complained by the user and a large number of unlabeled samples, mark the sample complained by the user as the first positive sample, and mark the large number of unlabeled samples as the first negative sample; wherein, the The sample of user complaints is the data of user complaints due to network quality factors.

[0037] In the embodiment of the present invention, the large amount of unlabeled data may be the data that the user feels that the network quality is very poor, but chooses not to complain due to time or other factors. Complaint data.

[0038] S2. Divide the sample data into training data and test data according to a preset ratio; wherein, the sample data includes the first positive sample and the first negative sample.

[0039] S3. Using an ensemble learning algorithm to train the training data to ...

Embodiment 2

[0055] see figure 2 , an embodiment of the present invention provides a user complaint prediction device, including:

[0056] Marking module 1, for re-marking the samples complained by the user and a large number of unlabeled samples, marking the sample complained by the user as the first positive sample, and marking the large number of unlabeled samples as the first negative sample; Wherein, the samples of user complaints are the data of user complaints due to network quality factors.

[0057] In the embodiment of the present invention, the large amount of unlabeled data may be the data that the user feels that the network quality is very poor, but chooses not to complain due to time or other factors. Complaint data.

[0058] A segmentation module 2, configured to segment the sample data into training data and test data according to a preset ratio; wherein, the sample data includes the first positive sample and the first negative sample;

[0059] The strong classifier tra...

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 user complaint prediction method and device, and the method comprises the steps: marking a user complaint sample and a label-free sample again, marking the user complaint sample as a positive sample, and marking the label-free sample as a negative sample; segmenting the sample data into training data and test data; training the training data by adopting an ensemble learning algorithm to obtain a strong classifier for predicting the probability that the sample is a positive sample; inputting the test data into a strong classifier to obtain the probability that each sample in the test set belongs to a positive sample; taking samples of which the positive sample probabilities are greater than a first threshold value in the test set as second positive samples, and taking samples of which the positive sample probabilities are less than a second threshold value as second negative samples; inputting the second positive sample and the second negative sample into a neural network model for training to obtain a user complaint prediction model; and inputting the real-time microphone data of the user into the user complaint prediction model to predict the probability of user complaint. According to the invention, the accuracy of user complaint prediction is ensured.

Description

technical field [0001] The invention relates to the technical field of mobile service support, in particular to a user complaint prediction method and device. Background technique [0002] The rapid development of mobile communication technology has made the use of mobile smart terminals popular rapidly. During the user's use of the communication service, the user may be dissatisfied with the service perception. Existing complaint data only have positive samples and a large number of unlabeled samples. The so-called positive sample is the behavior that users feel that the network quality is too poor, so they choose to complain. For the remaining unlabeled data, users may feel that the network quality is poor, but choose not to complain due to time or other factors; users may feel that the network quality meets expectations, and there is no need to complain. Like this kind of data that only has positive samples and unlabeled samples, how to train to ensure the accuracy of ...

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): G06Q10/04G06Q30/00G06Q50/30G06N3/08G06N20/20
CPCG06Q10/04G06Q30/016G06N3/08G06N20/20G06Q50/40
Inventor 杜翠凤陈少权蒋仕宝
Owner GCI SCI & TECH