A Revised Publication Model for ECML PKDD

TL;DR

This paper proposes a new publication model for ECML PKDD that introduces a journal track with continuous submissions and revision cycles to address issues in the traditional conference publication process.

Contribution

It introduces a novel publication model with a journal track and ongoing submissions to improve review quality and community engagement.

Findings

Enhanced review quality and consistency

Increased community engagement and collaboration

More flexible publication process

Abstract

ECML PKDD is the main European conference on machine learning and data mining. Since its foundation it implemented the publication model common in computer science: there was one conference deadline; conference submissions were reviewed by a program committee; papers were accepted with a low acceptance rate. Proceedings were published in several Springer Lecture Notes in Artificial (LNAI) volumes, while selected papers were invited to special issues of the Machine Learning and Data Mining and Knowledge Discovery journals. In recent years, this model has however come under stress. Problems include: reviews are of highly variable quality; the purpose of bringing the community together is lost; reviewing workloads are high; the information content of conferences and journals decreases; there is confusion among scientists in interdisciplinary contexts. In this paper, we present a new publication model, which will be adopted for the ECML PKDD 2013 conference, and aims to solve some of the problems of the traditional model. The key feature of this model is the creation of a journal track, which is open to submissions all year long and allows for revision cycles.