Machine Learning Analysis for Temperature Classification using Bioelectric Potential of Plant

Plants can respond to changes in their surroundings, such as temperature, humidity, and human behavior. This ability is due to photosynthesis and respiration in plants, resulting in low electrical signals. We focus on room temperature change by plant bioelectric potential detection. The temperature topic is interesting for the actual research because it is one of important factor in health aspect. This study aims to analysis of temperature of Bioelectric potential of plant using machine learning. This implementation identifies the plants' bioelectric potential in distinguishing three-room temperature categories: A, B, and C temperature. We conducted the experimental in room control. For the transformation process is using wavelet transform. Furthermore, the analysis process, we used the Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighborhood (K-NN), Naïve Bayes, and linear regression. Then, for the evaluation is using confusion Matrix. The result showed that the best level of transformation using Wavelet is 7
^th
level for reducing the noise. Furthermore, the best accuracy of analysis process is using Decision tree which obtained 75%.