BlueMind missing data imputation Project

版本 1.5 (298.7 KB) 作者: Abolfazl Nejatian
missing value prediction by use of deep learning algorithm and shallow learning algorithm
5.0
(2)
376.0 次下载
更新时间 2022/2/21

查看许可证

a deep learning version for the mising data prediction and imputation.
as preprocess part of work, it might be useful for Every scientist,statistician and engineers who work with database and datasets.
If you need further information, feel free to email me.
Best regards
Abolfazl Nejatian
mail: Abolfazl.nejatian@gmail.com
|| ----------------------------- work structure -----------------------------||
The purpose of this app is to find the Missing values by training the network with complete data (rows without any missing cells).
So, in the first phase, the data are divided into two sections: complete data and missing data.
The missing data are then sorted according to the minimum to the maximum number of missing values in their rows.
After that, according to the missing values pattern in the examined line, a network is trained on the complete data.
For instance, suppose that the pattern of missing values in that line be like below,
a, b, r, t, NaN, w, f, NaN, p, NaN
algorithm try to train a network with the complete data so that the network takes the data
Of the first, second, third, fourth, sixth, seventh, and ninth observers as the input and, the fifth, eighth, and tenth as the target.
After the training process is completed, use the trained network on the mentioned row (a, b, r, t, NaN, w, f, NaN, p, NaN)
and (a, b, r, t, w, f, p) are given to the trained network as input and then expect the network to predict the NaNs values of that row.
After predicting the NaN values, that row becomes clean and added to the complete data.
The algorithm then goes to the next rows and this continues until all the rows are cleared.
|| ----------------------------- How to Run -----------------------------||
  • Load Data
data should be a microsoft excel file.
  • choosing the imputation method. (MLP or LSTM)
  • and then click on start.
  • finaly the App gives a clean dataset.

引用格式

Abolfazl Nejatian (2024). BlueMind missing data imputation Project (https://www.mathworks.com/matlabcentral/fileexchange/69291-bluemind-missing-data-imputation-project), MATLAB Central File Exchange. 检索时间: .

MATLAB 版本兼容性
创建方式 R2018a
兼容任何版本
平台兼容性
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

版本 已发布 发行说明
1.5

some minor bugs were fixed.
some optimization on code performance.
1.1

fix some minor bogs
1.0