Principal Component Analysis (PCA): Perform PCA on your input data to analyze its variance and reduce dimensionality, focusing on the most significant components.
Principal Component Analysis (PCA) is a statistical method used for reducing the dimensionality of large datasets while retaining most of the original data's variability. It works by identifying the principal components—orthogonal directions that capture the maximum variance in the data. By projecting the data onto these components, PCA reduces the number of variables, simplifying the dataset while preserving its essential patterns. The first principal component captures the most variance, followed by subsequent components. PCA is widely used for data compression, noise reduction, and feature extraction, making complex datasets easier to analyze and interpret.
Find the Pearson's correlation coefficient subject wise and activity wise and analyze your observation.
When you are working with subject wise, you have to find out how the subjects are related with respect to their basic details as given in the dataset. For example, Age, Height, weight, BMI Index etc.
When you are working with Activity wise, you have to find out how the activities are related with each other.
2. Rank Feature using Filter based technique (Correlation Coefficient)
3. Select the features by setting limits or threshold
4. Train a classification model SVM (Support Vector Machine) taking the selected feature as input of the model and find the performance (accuracy, f1 score, precision and recall) of the model.
5. Evaluate the performance with other classification models such as Decision Tree, Random Forest and justify it.
6. Extract the same features as given in the following available benchmark dataset UCI-HAR.
Link: https://archive.ics.uci.edu/dataset/240/human+activity+recognition+using+smartphones