Workshop date: April 9th, 2020
Networks are ubiquitous in many real-life domains and applications: basically whenever/wherever there exists a relationship between two entities, we can talk about networks. The type of entities can be people (in which case we talk about social networks), proteins (biological networks), computers/routers/printers (computer networks), railway stations (transportation networks), etc. Entities can be characterized by their attributes (attributed networks), can be of different types (heterogeneous), can appear/disappear/change (them or their relations) over time (dynamic) etc.
Lately, quite extensive research focus has been put in different directions related to studying different types of networks and various research problems and challenges associated with them. This workshop focuses on three such directions.
First, it is important to understand the role of network modeling (also known as network design or network engineering). Although the majority of real life problems relate to network data, this does not necessary mean that the actual data is provided in the form of a network. This frequently requires transforming implicitly given network data into an explicit network and that does not necessarily mean that this transformation is always a unique one. In those cases, opting for a particular choice to model a network could significantly influence the outcome. Furthermore, modeling a network in a most straightforward way does not have to be the most optimal solution. Therefore, we seek for solutions which propose creative yet effective network models for solving particular applications.
Second, generating ad hoc/hand-crafted features based on the network structure or attributes is slowly but steadily fading away. Instead, there is a tremendously increasing tendency of performing network featurization in an automated way, primarily using representation learning methods. The latter attempt at learning representations of different parts of network: nodes, edges, sub-networks or even entire networks, typically with the goal of using learnt representation for downstream tasks such as node classification, link prediction or graph completion. However, despite a growing number of studies, research community is still facing a number of challenges e.g. how to learn the high quality representations in an efficient way on extremely large networks. Additionally, different network types bring along different challenges. Papers not only proposing novel algorithms and methods for addressing these challenges but also shedding light on how learnt representations could be used in concrete applications are of particular interest in this workshop.
Third, network analysis has demonstrated many benefits and provided many valuable insights in different application areas. However, growing amounts of data and appearance of different types of networks and underlying latent aspects, forces researchers to constantly innovate and adjust their perspectives on apparently known problems. Topics of interest in this stream range from discovering interesting patterns, detecting communities, via sampling and evaluation to network alignment, summarization and visualization.
List of topics (not exclusive):
Discovering interesting patterns/roles in the networks
Algorithms and methods for analyzing attributed/heterogeneous/signed/multiplex networks and knowledge graphs
Network clustering methods and applications
Network representation learning methods and applications
Methods for analyzing and learning on dynamic networks
Large-scale network analysis and learning
Semi-supervised learning, inductive learning on networks
Networks sampling, simulation and evaluation challenges and solutions
Case studies and empirical studies in different application areas, e.g. telco, finance, IoT, social media, social good, retail, biology, political sciences, etc.
November 30, 2019 December 7th, 2019 (AoE)
Notification: December 29, 2019
Camera-ready: January 8, 2020 (hard deadline because of Springer publishing deadlines)
Workshop date: Thursday, April 9th, 2020
Conference dates: April 7-10, 2020
Three types of papers can be submitted to NMLA workshop:
Full papers: Finished or consolidated R&D works. These papers are assigned a 10-page limit.
Short papers: Finished or consolidated R&D works and also ongoing work but with relevant preliminary results, open to discussion. These papers are assigned a 7-page limit.
Demo/visionary/insightful (potentially previously published) papers with a 3-page limit.
Please note that only full and short papers could be published at the Springer AISC proceedings. Demo/visionary/insightful (potentially previously published) papers will only be considered as poster presentations.
All submissions must be in PDF, written in English and must follow the formatting rules for Proceedings in Advances in Intelligent Systems and Computing Series (see Instructions for Authors at Springer Website).
All submissions will have three independent or in special cases two independent, non-conflicting and detailed reviews by the Program Committee members. The version of papers for evaluation by the Program Committee, saved in PDF format, must not include identification, e-mail and affiliation of the authors. This information must only be available in the camera-ready version of accepted papers, saved in Word or Latex format and also in PDF format. These files must be accompanied by the Consent to Publish form filled out, in a ZIP file, and uploaded at the conference management system.
In order to be published in the workshop proceedings, accepted full and short workshop papers must be registered and presented at the workshop.
The workshop proceedings will be published in Proceedings by Springer. Proceedings will be submitted for indexation by ISI Thomson, SCOPUS, DBLP, among others, and will be available in the SpringerLink Digital Library. Detailed and up-to-date information may be found at WorldCist’20 website: http://www.worldcist.org
To submit or upload a paper please go to: EasyChair Workshop Submissions
More detailed information on workshop program will follow.
For the information on NMLA’20 (and WorldCIST’20) venue see: http://www.worldcist.org/index.php/venue
1) Metadata Action Network Model for Cloud Based Development Environment, Mehmet N Aydin, Ziya N. Perdahci, Ilker Şafak and Jos Van Hillegersberg
2) Clustering Foursquare Mobility Networks to Explore Urban Spaces, Olivera Novović, Nastasija Grujić, Sanja Brdar, Miro Govedarica and Vladimir Crnojević
3) Open and closed triads in distributed online social networks, Cheick Tidiane Ba, Matteo Zignani, Sabrina Gaito and Gian Paolo Rossi
4) A Comparative Study of Representation Learning Techniques for Dynamic Networks, Carlos Ortega Vazquez, Sandra Mitrović, Jochen De Weerdt and Seppe Vanden Broucke
5) Finding Influential Nodes in Networks with Community Structure, Zakariya Ghalmane, Stephany Rajeh, Chantal Cherifi, Hocine Cherifi and Mohammed El Hassouni
6) The trilogy of algorithms for network embedding in the hyperbolic space, Carlo Vittorio Cannistraci
Sandra Mitrović, KU Leuven, Belgium
Jochen De Weerdt, KU Leuven, Belgium
Steven Skiena, Stony Brook University, NY, USA
Huan Liu, Arizona State University, AZ, USA
Aleksandar Bojchevski, Technical University of Munich, Germany
Aravind Sankar, University of Illinois Urbana-Champaign, IL, USA
Davide Mottin, Aarhus University, Denmark
Federico Battiston, Central European University, Hungary
Hocine Cherifi, University of Burgundy, France
Junting Ye, Facebook, WA, USA
Leto Peel, Université Catholique de Louvain, Belgium
Martin Atzmueller, Tilburg University, The Netherlands
Mohammad Al Hasan, Purdue University, IN, USA
Rémy Cazabet, Université de Lyon, France
Roy Ka-Wei Lee, University of Saskatchewan, Canada
Roya Imani Giglou, KU Leuven, Belgium
Sabrina Gaito, Università degli Studi di Milano, Italy
Sanja Brdar, University of Novi Sad, Serbia
Tyler Derr, Michigan State University, MI, USA
Wouter Verbeke, Vrije Universiteit Brussel, Belgium
Zhana Kuncheva, C4X Discovery, London, UK