To advance scientifically, talented individuals need to explore it. Traditional evaluation methods based on bibliometric indicators such as Sinta Score and Scopus often fail to identify the complexity of research change. These evaluation methods are also included based on bibliometric indicators such as Google Scholar and H-Index. This study proposes a talented researcher identification model using rotating machine learning (QML) to expand the limitations of predicting researchers’ rankings based on scientometric indicators. This model incorporates explicitly the reconstructed kernel Hilbert space (RKHS). Four main stages were followed: (1) data collection and preprocessing; (2) training of the QSVM model; (3) identification and visualization of researcher score; and (4) performance evaluation based on the difference between actual and performance scores. A dataset of researchers with different profiles was determined for the evaluation results of QSVM. The results showed that QSVM effectively predicted researchers with minimal deviation. The difference between the total and aggregate scores was between −0.25 and 0.05. The robustness of the model was revealed when comparing it strongly with the total and aggregate scores. Additionally, the analysis of the difference in rankings showed a low error rate, indicating the ability of QSVM to evaluate academic performance. This study highlights the explanations of QML-based classified models that offer a scalable and data-driven solution over traditional research methods.