Content-based Approach for Title Generation of Academic Papers

Mohammed, Mohammed Ali; Jabur, Ahmed Jasim; Naser, Noor Maher

doi:10.24271/psr.2026.509186.1973

Content-based Approach for Title Generation of Academic Papers

Document Type : Original Article

Authors

Mohammed Ali Mohammed ¹

Ahmed Jasim Jabur ²

Noor Maher Naser

¹ College of Business Informatics, University of Information Technology and Communications (UOITC), Baghdad, Iraq.

² niversity of Information Technology and Communications (UOITC), Baghdad, Iraq.

10.24271/psr.2026.509186.1973

Abstract

In a document, text, or article, the title should encapsulate the whole content with a few possible words. Thus, forming a powerful title that successfully summarizes the core of an academic paper is often a hard challenge. This paper introduced an automated title generation for academic papers to support the scholarly community that includes two types of process: sentence process and word process. The experimental results show that the title generation is close to the original scientific title. The title generated by the sumy summarization algorithm and pre-model k2t is a better from other ways due to the highest similarity values (0.732). This approach reached close to real-time, reliable, presenting a methodology to improve the effectiveness of the title creation process.

Keywords

Title generation

content-based approach

text frequency

text summarization

automatic title generation tools