# Data-Mining-in-Industrial-Processes-Evaluation-of-different-machine-learning-models-for-Prediction

**Repository Path**: greatwallisme/Data-Mining-in-Industrial-Processes-Evaluation-of-different-machine-learning-models-for-Prediction

## Basic Information

- **Project Name**: Data-Mining-in-Industrial-Processes-Evaluation-of-different-machine-learning-models-for-Prediction
- **Description**: Data Mining in Industrial Processes: Evaluation of different machine learning models for product quality prediction. Evaluated model types are Random Forest, Naive Gaussian Bayes, Logistic Regression, K Nearest Neighbour and Support Vector Machine. Comparision of non time based state based approach with time series based approach. Final result in precision 99.83 %.
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-09-18
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Data Mining in Industrial Processes: Evaluation of different machine learning models for product quality prediction

#######################################################################################
LinkedIn: https://www.linkedin.com/in/andreas-braun-6796ba12a/.
#######################################################################################

## Project Strucutre
Data Science as a Software, extension of (https://github.com/drivendata/cookiecutter-data-science)

## Some Frameworks:
Preprocessing, Pandas: https://github.com/pandas-dev/pandas  
Machine Learning, Scikit-learn: https://github.com/scikit-learn/scikit-learn  
Auto Machine Learning, TPOT: https://github.com/EpistasisLab/tpot  
Time Series Feature Extraction, TSFRESH: https://github.com/blue-yonder/tsfresh  

## Final Result
![](gfx/result.PNG?raw=true)

## Data Processing Pipeline
![](gfx/data_processing.PNG?raw=true)

## Data Type Portability
![](gfx/data_type_portability.PNG?raw=true)