amitthere/classification-algorithms

Nearest Neighbor, Decision Trees and Naïve Bayes classifiers implemented in python and evaluated based on 2 datasets

Classification Algorithms

3 classification algorithms implemented and tested on 4 datasets.

Performance of all classifiers is validated using 10-fold Cross Validation.

Accuracy, Precision, Recall, and F-1 measure of the classifiers is reported.

1. Nearest Neighbour Classifier

2. Decision Tree Classifier

* Random Forest Classifier based on the Decision Tree Classifier

* Boosting based on the Decision Tree Classifier

3. Naïve Bayes Classifier

Test Datasets

Dataset	Objects	Number of Classes
dataset1	569	2
dataset2	462	2
demo-dataset-1	150	2
demo-dataset-2	6	2

Dataset Format

Each row represents a gene:

1. Each line represents an object
2. Last column is the class label
3. Rest of the columns represent feature values, each of them can be a real-value (continuous type) or a string (nominal type)