User growth portrait construction method based on feature screening and semi-supervised learning
A semi-supervised learning and feature screening technology, applied in special data processing applications, instruments, calculations, etc., can solve problems such as high cost and insufficient labeled data, and achieve the effect of improving accuracy and stability
Active Publication Date: 2021-03-26
DALIAN UNIV OF TECH
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The common problem in user growth value prediction using supervised learning is insufficient labeled data. Labeled data often requires a lot of expert knowledge and takes a lot of time to complete. Therefore, high-quality labeled data is often very precious. A large amount of labeled data usually means a lot of cost
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
[0033] The present invention will be further described below in conjunction with accompanying drawing.
[0034] Such as figure 1 As shown, a user growth portrait construction method based on feature screening and semi-supervised learning includes the following steps:
[0035] Step 1. Preprocess the original user data, select the CSDN user portrait data set for the 2017 National Social Media Processing Conference SMPCUP evaluation task, and preprocess it, specifically including the following sub-steps:
[0036] (a) The user data in 9 files, including user browsing records, user posting records, user comment records, user like records, user click records, user like records, user follow records, user private message records and marked User growth value, integrated through user ID;
[0037] (b) According to the time records of each behavior of the user, count the number of times of each behavior of the user, and then fill the empty value of the user data table after integration ...
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 present invention relates to a user growth profile construction method, a user growth profile construction method based on feature screening and semi-supervised learning, comprising the following steps: (1) preprocessing the original user data, (2) according to the user's Behavior data, extract user behavior features and time features, (3) perform feature screening on behavior features and time features, (4) use semi-supervised learning to expand training set, (5) train first-level model, (6) model fusion , (7) Prediction of user growth value. In addition to paying attention to the user's behavior characteristics, the present invention also pays attention to the user's time characteristics, and uses the method of feature screening to select the features with obvious discrimination, combines the semi-supervised method to expand the training set, and finally uses the method of model fusion to improve the final model. accuracy and stability.
Description
technical field [0001] The present invention relates to a method for constructing user growth portraits, more specifically, to a method for constructing user growth portraits based on feature screening and semi-supervised learning. Background technique [0002] With the development of the Internet industry becoming more and more mature, major Internet companies are now stepping up their efforts to compete for user resources, and an important part of it is user precision marketing services. On the other hand, user portraits are used to label each user on the basis of big data, so it can be seen that user portraits are a great help to achieve precision marketing. And the user's growth portrait is a very important part of the user portrait. The growth value of users reflects the potential vitality of users on social media platforms, such as blogs and Weibo. If the user growth value in a certain topic area can be predicted, it will be helpful for the design of early products. ...
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 Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/2458G06K9/62
CPCG06F18/24323
Inventor 王健钱凌飞董哲瑾林鸿飞
Owner DALIAN UNIV OF TECH