You are on page 1of 1

Lab-7

Python for AI
23AIE232M

Experiment 1: Perform colour classification by implementing SVM, k-NN, DT,


and NB classifiers using 80-20% and 10-fold cross-validation. List all the
performance matrices for the above-given classifier.
Note: Select the model performing the highest classification accuracy for
further experiments.
Experiment 2: Observe the classification accuracy by expanding the above
experiment using the dimensional reduction technique (PCA). Select the most
appropriate PC component for optimal accuracy.
Experiment 3: Further try to improve the performance of the colour
classification using parameter tuning by grid search technique of the highest-
performing classifier from the above experiment.
Experiment 4: Perform K-means and DBSCAN clustering in the mall-
customers dataset link of the dataset is provided below. Decide the suitable
number of clusters using the elbow plot from no of clustering 1 to 10.
Link: https://www.kaggle.com/datasets/shwetabh123/mall-customers
Reference:
https://scikit-learn.org/stable/modules/generated/
sklearn.cluster.DBSCAN.html
https://scikit-learn.org/stable/modules/generated/
sklearn.cluster.KMeans.html
https://www.kaggle.com/code/fazilbtopal/popular-unsupervised-clustering-
algorithms

You might also like