Occupancy Detection Project
Using the provided data, supervised machine learning classification algorithms have been trained on training data set and tested against two test sets.
The original paper is “Accurate occupancy detection of an office room from light, temperature, humidity and CO2 measurements using statistical learning models” from Luis M. Candanedo and Véronique Feldheim.
The algorithms trained and tested on this project are,
Logistic Regression K-Nearest Neighbors Decision Tree Random Forest Support Vector Machine
For every algorithm, a table has been presented with several parameters and different feature combinations. At the end, there is a table that shows every algorithm with their best parameters and feature combinations alongside their scores for this project.
The initial dataset orginally were in text file but now converted to CSV file for better view and analysis. It would be attached in zip file. Train.csv - datatraining.txt Test1.csv - datatest.txt Test2.csv - datatest2.txt
Following libraries are uncomman and would be needed to run the iypnb file without any errors.
- pyplot
- sns
- export_graphviz
- plotly