Machine learning (ML) is increasingly used to improve immunotherapy by analyzing multi-omics data. This review discusses the application of ML in immunotherapy, including predicting response, identifying biomarkers, and analyzing the tumor microenvironment (TME). ML models can identify significant biomarkers, enhance understanding of immunotherapy mechanisms, and optimize decision-making. Current challenges include the need for more accurate prediction methods and the integration of multi-omics data. ML models have shown potential in various oncology applications, including early diagnosis, cancer classification, and predicting response and prognosis. However, existing methods based on simplistic thresholds lack accuracy. ML technologies offer the potential to refine the precision of immunotherapy response prediction by analyzing extensive datasets and discerning intricate patterns and interactions among molecular biomarkers. ML models can adapt to the dynamic nature of immune responses, offering a more comprehensive approach than traditional methods. Recent studies have focused on developing complex ML models incorporating multi-omics datasets for immunotherapy prediction. These models can provide a more comprehensive scope of tumor profiles, leading to improved performance in immunotherapy response prediction. ML models have also been used to predict the response to chimeric antigen receptor (CAR) T therapy. Additionally, ML models have been employed to identify biomarkers in the TME, including microsatellite instability (MSI), tumor mutational burden (TMB), cancer stem-like cells (CSCs), cancer-associated fibroblasts (CAFs), and tumor-infiltrating lymphocytes (TILs). These models can identify biomarkers associated with immunotherapy response and help in the development of personalized immunotherapy strategies. However, challenges remain in the clinical translation of ML models, including the need for rigorous clinical validation and the integration of multi-modal data. Despite these challenges, ML has emerged as a promising approach for evaluating TME, identifying TME-related biomarkers, and unraveling the intricate relationship between TME and immunotherapy. The biomarkers derived from ML approaches hold great potential for predicting clinical outcomes of immunotherapy and enhancing personalized immunotherapy strategies, thereby facilitating the advancement and wider application of immunotherapy in cancer treatment.Machine learning (ML) is increasingly used to improve immunotherapy by analyzing multi-omics data. This review discusses the application of ML in immunotherapy, including predicting response, identifying biomarkers, and analyzing the tumor microenvironment (TME). ML models can identify significant biomarkers, enhance understanding of immunotherapy mechanisms, and optimize decision-making. Current challenges include the need for more accurate prediction methods and the integration of multi-omics data. ML models have shown potential in various oncology applications, including early diagnosis, cancer classification, and predicting response and prognosis. However, existing methods based on simplistic thresholds lack accuracy. ML technologies offer the potential to refine the precision of immunotherapy response prediction by analyzing extensive datasets and discerning intricate patterns and interactions among molecular biomarkers. ML models can adapt to the dynamic nature of immune responses, offering a more comprehensive approach than traditional methods. Recent studies have focused on developing complex ML models incorporating multi-omics datasets for immunotherapy prediction. These models can provide a more comprehensive scope of tumor profiles, leading to improved performance in immunotherapy response prediction. ML models have also been used to predict the response to chimeric antigen receptor (CAR) T therapy. Additionally, ML models have been employed to identify biomarkers in the TME, including microsatellite instability (MSI), tumor mutational burden (TMB), cancer stem-like cells (CSCs), cancer-associated fibroblasts (CAFs), and tumor-infiltrating lymphocytes (TILs). These models can identify biomarkers associated with immunotherapy response and help in the development of personalized immunotherapy strategies. However, challenges remain in the clinical translation of ML models, including the need for rigorous clinical validation and the integration of multi-modal data. Despite these challenges, ML has emerged as a promising approach for evaluating TME, identifying TME-related biomarkers, and unraveling the intricate relationship between TME and immunotherapy. The biomarkers derived from ML approaches hold great potential for predicting clinical outcomes of immunotherapy and enhancing personalized immunotherapy strategies, thereby facilitating the advancement and wider application of immunotherapy in cancer treatment.