Identification of Scientists of Future Success using Bibliometric Indicators
Although quantifying the current relevance of academics is of substantial interest to scholarly decision makers, there is a wide range of situations in which future relevance is more important than the current one (e.g., in guiding funding allocations, recruitment decisions, and rewards). In this research project, we aim to develop models/methods based on bibliometric indicators for early identification of academics of future scientific success. Exploring the current relevance and temporal (sequential) aspect of the evolution of the scientific impact (and research output) of academics, we create a model in order to predict the future scientific impact (and future research output) of academics using deep learning techniques. Subsequently, we use it as part of a successful academic prediction framework formulated as a linear programming problem. Citation and publication data for two bibliometric data sets widely recognized in the literature (Arnetminer e Semantic Scholar) were used. Finally, the experimental results indicate that the framework can early identify a large portion of academics of future scientific success.