Below you will find pages that utilize the taxonomy term “classification”
Post
PCA and classifications
Principal component analysis: Some few tool to analyze the specific impact of each feature, and their impact on the label.
The tool is applied to a dataset of flowers, where we try to classify which features of the plant correspond to the specific flower.
We have four features sepal width, sepal length, petal width, and petal length. All are in cm.
The labels or the flowers we are trying to classify are Iris Setosa (label = 0), Iris Versicolour (label = 1), Iris Virginica (label = 2)
Post
Support vector machine - Linear
Link to github repo
The Support Vector Machine Support vector machine is a classification algorithm. The idea behind the algorithm is that it tries to seperate 2 different labeled datasets. Compared to kNN and RandomForest, this is a supervised algorithm. Hence meaning that we use our labeled data to divide/classify our datapoints.
We divide the dataset utilizing a hyperplane. A hyperplane is a function of degree (Features-1), 2 features will entail a function of degree 1 (a line).