Academic Papers Slides Generation Using Nlp & Ilp

Abstract

Presentation slides have been a popular and efficacious designates to present and transfer information, especially in academic conferences. So we investigate a very arduous task of automatically engendering presentation slides for academic papers with including both text elements and graphical elements. The engendered presentation slides can be utilized as drafts to avail the presenters prepare their formal slides in a more expeditious way. The proposed system is to address this task. It first employs the regression method to learn the paramountcy scores of the sentences in an academic paper, and then exploits the integer linear programming (ILP) method to engender well-structured slides by culling and aligning key phrases and sentences. Evaluation results on a test set of 200 pairs of papers and slides accumulated on the web demonstrate that our proposed system can engender slides with better quality. A utilizer study is withal illustrated to show that proposed system has a few evident advantages over baseline methods. The automatic presentation slides generation system’s Main quandary was images. Earlier system are failed to include images with the slides. The proposed system overcome this quandary.