DMData Mining Techniques
All experiments

Experiment 08

K-Nearest Neighbors (K-NN) Classifier Using WEKA

Aim

To perform k-nearest neighbors (k-nn) classifier in weka using WEKA tool.

Procedure

  1. Open WEKA → Launch Explorer.
  2. Load dataset: Click Open file... → Select iris.arff.
  3. Go to the Classify tab.
  4. Choose classifier:
    • Click Choose → lazy → IBk.
  5. Set k-NN parameters:
    • Click on the classifier → Set KNN = 3 (or any value).
    • Distance function = Euclidean (default).
  6. Select Test Options:
    • Cross-validation (10 folds) is recommended.
    • Or Percentage split (e.g., 70% train, 30% test).
    • Or Supplied test set if you have a separate dataset.
  7. Run classifier: Click Start.
  8. View results:
    • Confusion matrix
    • Classification accuracy
    • Precision, Recall, F1-score

Output

K-Nearest Neighbors (K-NN) Classifier Using WEKA output screenshot 1K-Nearest Neighbors (K-NN) Classifier Using WEKA output screenshot 2K-Nearest Neighbors (K-NN) Classifier Using WEKA output screenshot 3

Result

The experiment was successfully executed, and the expected output was obtained using WEKA.