The current study deals with the evaluation of vibrations and noise generated by a diesel engine powered by Calophyllum inophyllum biodiesel blend (B20) and molybdenum trioxide (MoO3) nanoparticles, predicted using machine learning algorithms. The nanoparticles were spread out in B20 at a level of 75 ppm, along with a surfactant and a dispersant to make sure they mixed evenly with the fuel samples. The stability of the nanoparticles was then tested using the principle of photo spectroscopy and it was found that the stability of nanoparticles to which surfactants and dispersants had been added was increased. The experiment with the diesel engine was carried out to evaluate the vibrations and noise by varying different loads (3, 6, 9, and 12 kgf) and injection pressures (200, 220, and 240 bar). The vibrations and noise intensity were reduced with B20 compared to normal diesel and a further reduction was observed by adding nanoparticles together with the dispersant. With increasing load, the RMS velocity and the RMS noise also increased, but these decreased with increasing injection pressure. The most favourable result for RMS velocity and RMS noise was B20+MoO3 75ppm+ Dispersant 75ppm. At an injection pressure of 240 bar and a full load of 12 kgf, the RMS velocity was 0.568 m/s and 55.265 dB, which corresponds to an improvement of 28.76% and 9.78% respectively compared to conventional diesel for B20+MoO375ppm+ Dispersant 75ppm. Finally, the entire experimental data of 60 were predicted using various machine learning (ML) algorithms, such as Random Forest (RF), Decision Trees (DT), Artificial Neural Networks (ANNs), Support Vector Machines (SVM) and XG Boost and the predicted results correlated well with the experimental values. All algorithms have shown a good association, but XGBoost has shown a better correlation coefficient (R2). The R2 of each algorithm is in the range of 0.94-0.99, the MRE and RSME are also in the significant range for both RMS velocity and RMS noise.