Abstract: Helicobacter pylori (H. pylori) can be transmitted through oral-oral or fecal-oral routes and is characterized by strong transmissibility, high infection rate, high disease burden, and poor compliance. Precise clinical decision-making is crucial for the treatment and prognosis of the patient. Machine learning, with its powerful capabilities of data mining and processing, plays a significant role in the diagnosis and prediction of H. pylori infection. This article reviews the latest advancements of machine learning in diagnosis of this field, prediction of infection, prediction of gastrointestinal hemorrhage, prediction of gastric cancer prediction, prediction of treatment programs and prediction of drug resistance, summarizing and comparing the performance, utilization potentiality, existing risks and model generalization ability. It aims to provide references for healthcare workers to implement the optimal clinical decision-making and further complete the management mode of the patients with H. pylori infection.