Aim: The objective of the work is to detect and determine tweets that indicate suicide ideation using SMO over Naive Bayes. To achieve accuracy, a novel SMOClassifier function was used. Materials and Methods: Accuracy and Loss are performed with SUICIDAL_DATA dataset from the Github library. The total sample size is 242. The two groups considered were Sequential Minimal Optimization and Naive Bayes. Result: The accuracy of SMO is 93.5% and loss is 6.5%, which appears to be better than Naive Bayes whose accuracy is 81.7% and loss is 18.3%. Finally, SMO appears significantly better than the Naive Bayes algorithm. The two algorithms, Sequential Minimal Optimization (SMO) and Naive Bayes (NB) with independent sample T-Test value achieved is p=0.662 (p>0.05), statistically insignificant.. Conclusion: Detecting suicidal tweets significantly seems to be better in Sequential Minimal Optimization (SMO) than Naive Bayes (NB).