Unsupervised Learning Applied Exercises
In this session, we will practice how data dimensionality reduction and clustering methods work.
We will be using the *Human Activity Recognition dataset* provided by Samsung . This data comes from the accelerometer and gyroscope of the Samsung Galaxy S3 phone, you can find out more about these features via the link above). These types of activities include: walking, standing, lying down, sitting, or climbing stairs.
During the remainder of the challenge, we first feigned ignorance of the type of activity and attempted to cluster samples purely based on features. We then solve the problem of determining the type of physical activity as a classification problem.
Let's first import the modules and functions that may be used in this challenge.
from sklearn.svm import LinearSVC
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import PCA
<