Using textual content for predicting scientific success.

• Grant number: 24/05715-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2024
• End date: June 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Diego Raphael Amancio
• Grantee: Pedro Manicardi Soares
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

Text representation is one of the most important methods in natural language processing and is increasingly being applied in various areas. One application, for example, is based on the use of embeddings to classify the emotional intent of a text. In addition, it's possible to use textual representation to predict the success of a work, such as a scientific article, a novel, or a book.In this sense, there are many studies in the literature dealing with the prediction of scientific impact, but most have been limited to techniques using the title and abstract of articles. Therefore, there is a need for new studies that use combinations of advanced techniques to analyze the entire textual content, as well as authors data, providing a more complete view. In this scenario, this Scientific Initiation research proposes the investigation of using embeddings to predict the success of a scientific article. The objective is to utilize textual content and divide the articles into sections, in order to discover if a section has more impact on the work. Finally, information about the authors of the works will be added to investigate if a certain author is more likely to have greater recognition.