Traditional Analysis of Variance (ANOVA) is a useful statistical test for group-wise comparison of means. However, two problems can undermine its effectiveness. First, implementing ANOVA can be complex because it assumes one has a thorough understanding of statistical concepts and assumptions. This can be a roadblock for less statistically adept researchers and practitioners for not benefitting from the full potential that ANOVA offers. Second, ANOVA results can often be difficult to interpret in practice, and significance tests such as p-values or F-statistics may not lead to actionable conclusions. To overcome these issues, we present a novel approach to perform ANOVA problem-solving with machine learning which is compared by using many other machine learning models. The best among intelligent classifiers was the Support Vector Machine (SVM) giving an accuracy of 95.2%. This study can significantly broaden the applicability of ANOVA, empowering researchers across domains to uncover valuable insights from their data.