Design of E-Nose
In this project, we studied a classification problem using a gas sensor dataset in UCI Machine Learning Repository.
Motivation:
A chemical detection platform composed of 8 chemo resistive gas sensors was exposed to turbulent gas mixtures generated naturally in a wind tunnel. The acquired time series of the sensors are provided. Through our project, we tried to predict the level of concentration of the gas Ethylene emitted from the turbines used for compiling this dataset.
Installation:
Please install Jupyter Notebook and python 3.7 in your laptop using the corresponding operating system from the anaconda website.
Features:
We used multiple machine learning algorithms including Logistic Regression, Random Forest, LDA, QDA and Decision Tree.
How to use?
The project is rather simple and straight forward. Once you install the jupyter notebook, download the relevant. ipynb file and place it in the same directory as the dataset.
Credits:
The major motivation for this project was the Machine learning course at San Jose state University and the datasets from the UCI repository.