Missing value interpolation method based on multi-method ensemble learning
An integrated learning and missing value technology, applied in integrated learning, special data processing applications, instruments, etc., can solve problems that affect the accuracy of later prediction models, data research cannot be carried out smoothly, and model-related information is missing
Pending Publication Date: 2021-05-18
胡安民 +2
View PDF0 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Since most of the current models based on statistical methods or machine learning algorithms require all data to be complete, the existence of missing data prevents research from proceeding smoothly
On the one hand, if the missing data variables are simply and roughly eliminated, the relevant information of the model will be missed. On the other hand, if the simple interpolation directly affects the accuracy of the later prediction model
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
[0019] The above and other technical features and advantages of the present invention will be described in more detail below in conjunction with the accompanying drawings.
[0020] Random Forest (RF) refers to a classification that uses multiple decision trees to train and predict samples, and each decision tree is unrelated. Random Forest randomly selects training data with replacement and then constructs a classifier, and finally combines learning to obtain a model to increase the overall effect. The random forest calculates the importance of each feature as a whole and sorts them in descending order, and then removes some features according to the importance of the features to obtain a new feature set, then sorts the importance and removes some features again, and iterates repeatedly; finally according to The different feature sets obtained and their corresponding out-of-bag error rates. The feature of the dependent variable is the feature set corresponding to the lowest o...
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 provides a missing value interpolation method based on multi-method ensemble learning. The missing value interpolation method comprises the steps of S1, extracting original data; s2, screening a part of missing variables; s3, carrying out simple interpolation on missing values; s4, screening feature variables of missing values; s5, carrying out 10-fold cross validation; s6, carrying out multiple interpolation of missing values; S7-S8, iteratively covering the original simple interpolation data; and S9, respectively predicting variables with large missing proportions. According to the method, multiple methods are used for predicting missing values, the potential uncertainty influence of interpolation data on the model is weakened as much as possible, meanwhile, real incomplete variable information is utilized to the maximum extent, and the accuracy and the efficiency of missing data prediction are improved.
Description
technical field [0001] The invention relates to the field of missing data filling, in particular to a missing value interpolation algorithm based on integrated learning of multiple methods. Background technique [0002] At present, big data related research has been widely used in many fields, and then there are often some missing data in the actual data extraction process. Since most of the current models based on statistical methods or machine learning algorithms require all data to be complete, the existence of missing data prevents research from proceeding smoothly. On the one hand, if the missing data variables are simply and roughly eliminated, the relevant information of the model will be missed. On the other hand, if the simple interpolation is used, it will directly affect the accuracy of the later prediction model. [0003] Purpose of the invention: In order to solve the above-mentioned technical problems, try to weaken the potential uncertainty impact of interpol...
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 Applications(China)
IPC IPC(8): G06F16/215G06K9/62G06N20/20
CPCG06F16/215G06N20/20G06F18/24323
Inventor 胡安民吴超然李镇
Owner 胡安民