Basic Profiling Extraction Based on XGBoost

Wenhan Yang, Boyu Sun, Banruo Liu

Abstract: With the deepening of human academic research in various fields and the diversification of research branches, it has become an important work to obtain the information of scholars in the same field and conduct reference research on their research results. Thus, it is of vital importance to obtain relevant scholar information through information extraction and prediction by the result of search engines. Through XGBoost, KNN, information extraction and other methods, we realized the function of predicting scholars’ home page, email address, language, gender, title and other information through the search engine search results of scholars’ names and institutions, and achieved high accuracy in some aspects.