The rapid growth of digital media is the main cause for the loss of integrity and confidentiality of news. Identifying whether the news is fake or true is the measure concern among the people these days. Traditional techniques lead to less accurate results require more time and more human intervention. To address these issues Machine Learning (ML) models are utilized which automate the process of identifying fake and true news. In the proposed work two datasets are used to train the models. One dataset is made up of true news and other is made up of fake news then these two datasets are merged to train the models. Three machine learning algorithms including Support Vector Machine (SVM), Logistic Regression (LR), and Random Forest (RF) are proposed for detection of true or fake news. The performance of each algorithm is assessed based on metrics: accuracy, precision, recall, support, and F1 score. Objectives: The aim of the research work is to evaluate the efficiency of these models in identifying true and false news. Methods: Proposed research addresses whether news is true or fake, using two datasets (true.csv and fake.csv), and these datasets are merged to analyse proposed models. For evaluation accuracy, precision, recall, support, and F1-score are calculated. Results: Support vector machine outperform among the proposed models for detection of true and fake news. SVM attained highest accuracy which is 93.05%, precision of 0.93(fake) and 0.94(true), recall of 0.94(fake) and 0.92(true), F1-score of 0.93(fake) and 0.93(true) and support of 638(fake) and 629(true). Conclusion: Support vector machine outperformed the proposed models: RF and LR. Simulation of the work showed the potential of machine learning models in the automatic detection of fake and true news. The proposed work automates the process of detecting true and fake news. It improves reliability of information sharing in society and integrity of news.