Aim: The proposed study aims to detect Spam SMS using a Novel Attribute Selection Measure in Decision Tree Classifier Algorithm in comparison with K-Nearest Neighbor. Materials and Methods: The dataset considered in the current research is available on Kaggle, a machine learning repository. The dataset “SMS spam collection dataset” contains 5572 instances and two attributes v1 and v2. The v2 is the input messages which are either spam or nonspam. The predicted label v1 has two classes: 0 = nonspam and 1 spam. In the data, 4900 are nonspam samples and 672 are spam samples. The sample size was calculated using G Power(95%). The accuracy and sensitivity of the classification of SMS spam detection were evaluated and recorded. Results: The accuracy was maximum in the classification of SMS spam detection using the Decision Tree Classifier Algorithm (95%) which uses a Novel Attribute Selection Measure with a minimum mean error when compared with K-Nearest Neighbor (93%). There is a significant difference of 0.12 between the classifiers. Conclusion: The study proves that the Decision Tree Classifier Algorithm which uses a Novel Attribute Selection Measure exhibits better accuracy than the K-Nearest Neighbor in the Classification of SMS spam detection.