CS3244 project - working on SMS spam classifier using K-Nearest Neighbors (k-NN) algorithm.
This repo contains the code used for implmenting k-NN algorithm to classify spam and ham SMS messages from the dataset spam.csv.
File contents list: -
| File | Details |
|---|---|
| knn_v2.py | Contains the source code for k-NN implementation. This version randomly shuffles and obtains 20% of training data to be the validation set. |
| knn_v3.py | Contains the source code for k-NN implementation. This version implements 10-fold cross validation. |
| Training data | Contains the training data used to classify SMS messages. |
| Experiment results | Contains the experimental accuracy results for classifying messages using k-NN. (Note: x-axis denotes k-value, y-axis denotes validation error) |