Proteins engage in physical interactions to create extensive, highly interconnected networks, representing a significant study domain. Protein–Protein Interaction Network (PPIN) are beneficial as they enhance biological and biomedical applications while offering fundamental scientific abstraction like pharmaceutical development and therapeutic repurposing. Due to the biological and experimental challenges in identifying proteins associated with diseases in PPIN, it requires substantial time and financial resources for validation. Consequently, the computational techniques can serve as a viable alternative. Recent studies indicate a correlation between PPIN and the diagnosis of Endometrial Cancer (EC) and Ovarian Cancer (OC). Comparative analysis of gene data also yields common factors within these two types of cancer which might lead to pregnancy problems as well. So, an effort has been undertaken to identify the essential target proteins in PPIN associated with these two cancer types. The proposed work employs a total of 2821 gene data pertaining to EC and OC genes retrieved from DisGeNET. A method employing Machine Learning (ML) for the identification of EC and OC proteins in PPIN is proposed, which effectively collects data, extracts crucial features, conducts data cleansing, integrates data, and eventually predicts using ML models. Features are extracted with Cytoscape and the Pfeature web service. Then a multifaceted classification model is trained and assessed utilizing several classifiers like XGBoost, K-Neighbours, AdaBoost, Support Vector Classification, Decision Tree model, Logistic Regression, Naive Bayes, and Random Forest model. The Random Forest model exhibits superior performance relative to other state-of-the-art models.