In the era of digital communication, the spread of fake news has grown to be a serious problem since it causes divisiveness, disinformation, and social unrest. This study proposes an ensemble model, Integrated Classifier for Fake News Detection (ICFND) which integrates multiple machine learning algorithms- SVM, Decision Tree, Naïve Bayes, K-NN and Random Forest to identify false information. The methodology involves the acquisition of a heterogeneous dataset comprising authentic and fraudulent news articles. This data is then used to train machine learning models to make logical conclusions and arrive at reasonable assumptions. This study’s primary objective is to identify the key traits that distinguish authentic news articles from counterfeit ones and to efficiently and accurately categorize news articles. The research also examines how well the developed models and an ensemble model may be scaled and used to a variety of news sources and topics. Results obtained from this research demonstrate the analysis of various Classification Models in fake news detection. This study shows that the ensembled model (ICFND) has higher accuracy of 92% than other individual algorithms that have been used in this study. Some other parameters like F1-Score, Precision, Recall, and RoC scores have also been calculated in which the performance of ICFND is better than SVM, Naïve Bayes, Decision Tree, K-NN, and Random Forest algorithms. Nevertheless, the ICFND model shows the best performance in FND tasks; however, it shows high computational cost and affects the quality of data that can be resolved in the future work.