This study trains three machine learning models with varying complexity-Random Forest, Support Vector Machine, and Neural Network-to predict cyclogenesis at a forecast lead time of 24 hr for given tropical disturbances identified by an optimized Kalman Filter algorithm. The overall performance is competent in terms of f1-scores (similar to 0.8) compared to previous research of the same kind. An assessment by SHapley Additive exPlanations (SHAP) values reveals that mid-level (500 hPa) vorticity is the most influential factor in deciding if a tropical disturbance is developing or non-developing for all three models. Wind shear and tilting are found to hold a certain level of importance as well. These results encourage further experiments that use physical models to explore the dynamical, mid-level pathway to tropical cyclogenesis. Another usage of SHAP values in this work is to explain how a machine learning model decides if an individual tropical disturbance case will develop, by listing the contribution of each feature to the output genesis probability, illustrated by a case study of Typhoon Halong. This increases the reliability of the machine learning models, and forecasters can take advantage of such information to issue tropical cyclone formation warnings more accurately. Several caveats of the current machine learning application in the studies of tropical cyclogenesis are discussed and can be considered for future research. These can benefit the interpretation and emphasis of certain output fields in the operational dynamical prediction system, which can contribute to more timely cyclogenesis forecasts. Machine learning methods are utilized to improve the prediction of typhoon formation. In addition to high accuracy, the models also provide additional information on decisive formation factors. The most evident general relationship found in typhoon formation is that the stronger the mid-tropospheric circulation, the higher the probability of typhoon formation. The models are capable of showing why individual disturbances may or may not grow into typhoons. The results are found to be physically consistent and helpful in enhancing the trustworthiness of the machine learning product. Hence, the methods and models have promising potentials for being applied to real-life typhoon formation forecasting as convenient and reliable tools. Machine learning methods are trained to forecast tropical cyclone genesis events with a decent accuracy of similar to 80% Mid-level vorticity and wind shear are important fields in the cyclogenesis predictions informed by the machine learning models SHAP values are used to interpret what information the models used to estimate the probability of cyclogenesis
