Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/zenitsu272/fault-detection-ml

Machine Learning based Fault Detection in machines using sensor data
https://github.com/zenitsu272/fault-detection-ml

artificial-intelligence decsion-tree machine-learning pandas pandas-dataframe pandas-python scikit-learn

Last synced: 9 days ago
JSON representation

Machine Learning based Fault Detection in machines using sensor data

Awesome Lists containing this project

README 

        
# Fault-detection-ML

So here, i tried to create a machine learning model that uses the random forest algorithm to predict the chances of a machine becoming faulty using sensory data.

Being a beginner to ML concepts, this basic project classifies whether a Machine would get faulty very soon or the machine is safe (no fault will be there in the near future.

The output mainly consisted of 2 possibilities [0,1]

1 being the machine would become faulty very soon or the chances of this machine being faulty is high

0 being the machine is safe in the near future or chances of this machine being faulty is low

Knowing that the possibilities of the output is only 2, I used the Desicion tree algorithm to evaluate this dataset.BUT, it turned out that my training set was overfitting and my testset accuracy was not great.

Considering all this, I used Random forest algorithm and evaluated this dataset.

This comes under supervised learning thus the dataset is splitted into 2 (I/O). 

Also there is seperate allocation of the data for training and testing in the ratio of 7:3.

Also this model gave me an accuracy of more than 90% in almost all the cycles.

Since this is just the beginning of me making projects, help me and correct me if there's something to be done.

Dataset link from kaggle:https://www.kaggle.com/datasets/umerrtx/machine-failure-prediction-using-sensor-data