Health Monitoring of Ship's Engine with Simulated Data by using Classifiers -Preliminary Result

Simulated data generation refers to the process of creating artificial data that mimics the characteristics of real-world data. This approach is commonly used in machine learning and statistical modeling to test and evaluate algorithms and models before applying them to real-world data. Second, simulated data can be used to test the robustness of algorithms and models against outliers, missing data, and other common data issues. Researchers can introduce these issues into the simulated data to evaluate how well the algorithms and models perform under these conditions. Simulated data generation is a powerful tool for testing and evaluating machine learning and statistical modeling algorithms. By creating artificial data that mimics the characteristics of real-world data, we can assess the performance of algorithms and models under a range of conditions, test their robustness against common data issues, and address privacy concerns associated with real-world data. This paper proposes a simple method for monitoring ship's engine health using various classification techniques like logistic regression, KNN (K- nearest neighbors), Naïve bayes, SVC (Support Vector Classifier) and Decision tree of machine learning. Before applying these algorithms synthetic data were generated and compared using various models under the library called “SDV (Synthetic data Vault).