Abstract

ConfSys, a conference management system, has been used for over 15 years to manage a number of international academic conferences, such as C3S2E, IDEAS, and ICCM. It supports multi-event, multi-track conferences with a large number of user participation and submissions. It provides essential services such as setting up a conference, user sign-ups, call for papers, paper submissions, paper auction, paper allocation and review, blind debate, paper decision, author registration, final version submissions, automatic session management, presentation uploads, program generation, managing event sessions and electronic proceedings creation to efficiently manage and support the running of academic conferences and journals.

This thesis presents the fourth iteration in the ConfSys system to further accelerate enhancements and to incorporate new features keeping in sync with recent technological advancements. It presents new approach for Automatic Semantic Header Generation (ASHG 2) in Information Retrieval from academic documents such as research papers. ConfSys4 includes modules for document processing, information retrieval, and document classification. Document processing involves conversion of PDF documents to XML-formatted documents. Information retrieval involves extraction of paper-related details such as title, abstract, keywords, author names, emails, organizations, locations, affiliations, and author references present within the document. The extraction of author-related details ensures verification of the author metadata and references section for the submitted document. Document classification involves extraction of important subject headings (keywords) based on the contents of the submitted document. Thus, improving the paper submission, single/double/triple-blind review, and paper allocation process by reducing manual data entry by users. Additionally, ConfSys4 includes improvements to existing features, such as adding automatic reminder emails to program committee members for updating their topics of interest for improved paper allocation and using entity matching technique for author pairs identification with conflicting interests. Furthermore, the PayPal payment interface is improved to include a standard checkout feature for payments and PDF document generation for the final program, invoice, and payment receipt for user registration to events. These improvements in ConfSys4 ensure consistent metadata generation for papers, improved transparency, ease of usability, and operability for organizing committee members, authors, and system users of the ConfSys system.