Academic Potential Index Algorithm And System Implementation Using Scholarly Big Data

With the development of scientific research and the increase of scholars,research for scholars has become more and more extensive.Researchers always hope to use some methods to select outstanding scholars in the academic circle.Based on this idea,many evaluation indicators of scholars influence were proposed,and some prediction methods of scholars influence have emerged,hoping to early identify the outstanding scholars of the future.Most of the studies are more focused on evaluating the academic capabilities that scholars have demonstrated,but seldom do research on the capabilities that scholars have not demonstrated.This paper innovatively proposes an academic potential index algorithm that attempt to measure academic potential of scholars.The potential of scholars is a relatively abstract concept.When we say that a scholar has great potential,it is actually a prediction of the scholar's academic influence in the future.Therefore,the algorithm first looks for the characteristics of scholars who determine the future influence of scholars.Regard the H-index and citation as the measure of academic influence,utilize machine learning's multiple linear regression model to predict scholars' academic influence in the future.However,the greater scholar's academic influence in the future does not mean that the scholar's current potential is greater,because the scholar's potential needs to be combined with scholars' current and other scholars' academic status.For the question of how to measure the potential of scholars,the idea of acceleration in physics gives a hint.The reason why an object has acceleration is because of the effect of force,we can translate this theory to scholars.The acceleration of scholars' academic influence must also be influenced by “force”,this “force” can be regarded as the potential.Therefore,after calculating the academic influence of the scholars in the future,the algorithm calculates the scholar's current academic influence acceleration and obtains the final academic potential index.Finally,based on Microsoft's MAG dataset,this paper implements the proposed academic potential index algorithm.Calculates the potential index for the academics in the field of computer,and develops the academic potential index system.Show the results of the calculations in a systematic way and provide reference for researchers.