Halifax, Canada
11-13 October, 2021
DS'2021 provides an open forum for intensive discussions and exchange of new ideas among researchers working in the area of Discovery Science. The conference focus is on the use of artificial intelligence methods in science. Its scope includes the development and analysis of methods for discovering scientific knowledge, coming from machine learning, data mining, intelligent data analysis, and big data analytics, as well as their application in various domains.
We invite submissions of research papers addressing all aspects of discovery science. We encourage papers that focus on the analysis of different types of massive and complex data, including structured, spatio-temporal and network data, as well as heterogeneous, continuous or imprecise data. We also encourage papers in the fields of computational scientific discovery, mining scientific data, computational creativity and discovery informatics. We welcome papers addressing applications of artificial intelligence in different domains of science, including biomedicine and life sciences, materials science, astronomy, physics, chemistry, as well as social sciences.
After careful consideration of all pros and cons of the different alternatives, and given the current uncertainty concerning Covid-19 travel restrictions, we have decided that the conference will take place as a fully online event. The accepted papers will still be published by Springer and the special issue will proceed as announced. In these challenging times that the whole of humanity is going through, we hope that all of you are safe and remain healthy and positive.
2022/02/01 - Special Issue on the Springer Machine Learning journal announced. Check details below.
2021/10/08 - Web access the conference proceedings is now available in the program section of this web page!
2021/10/04 - The paper Neural Additive Vector Autoregression Models for Causal Discovery in Time Series by Bart Bussmann, Jannes Nys and Steven Latré was selected for the Best Student Paper Award sponsored by Springer! Congratulations!
2021/10/03 - Dr. Ross King will be one of the keynote speakers of DS'2021!
2021/09/15 - Dr. Rita Orji will be one of the keynote speakers of DS'2021!
2021/09/10 - Dr. Tanya Berger-Wolf will be one of the keynote speakers of DS'2021!
2021/08/07 - Conference registration is now open!
2021/08/03 - After careful consideration of all aspects related to this pandemic, we have decided that the conference will be 100% online.
2021/07/28 - Notifications sent to authors! 15 Long papers and 21 Short papers were accepted for DS'2021! Congratulations to all authors!
2021/07/20 - Due to the extension of the submission deadline, the notification and camera ready dates were also push forward a bit.
2021/06/15 - Due to numerous requests the abstract and full paper submission deadlines were extended one further week!
2021/05/12 - Abstract and full paper submission deadlines extended!
2021/05/01 - Submission site on Easy Chair open
2021/03/03 - First version of the conference Web site is live! Still work in progress, obviously!
Access the Conference Proceedings HERE
Incremental k-Nearest Neighbors Using Reservoir Sampling for Data Streams
Maroua Bahri and Albert Bifet
Neural Additive Vector Autoregression Models for Causal Discovery in Time Series
Bart Bussmann, Jannes Nys and Steven Latré
BEST STUDENT PAPER AWARD
Leveraging Grad-CAM to Improve the Accuracy of Network Intrusion Detection Systems
Francesco Paolo Caforio, Giuseppina Andresini, Gennaro Vessio, Annalisa Appice and Donato Malerba
Consensus Based Vertically Partitioned Multi-Layer Perceptrons for Edge Computing
Haimonti Dutta, Saurabh Amarnath Mahindre and Nitin Nataraj
FHA: Fast Heuristic Attack against Graph Convolutional Networks
Haoxi Zhan and Xiaobing Pei
Shapley-Value Data Valuation for Semi-Supervised Learning
Christie Courtnage and Evgueni Smirnov
Automated Grading of Exam Responses: An Extensive Classification Benchmark
Jimmy Ljungman, Vanessa Lislevand, John Pavlopoulos, Alexandra Farazouli, Zed Lee, Panagiotis Papapetrou and Uno Fors
Learning Time Series Counterfactuals via Latent Space Representations
Zhendong Wang, Isak Samsten, Rami Mochaourab and Panagiotis Papapetrou
Combining Predictions under Uncertainty: The Case of Random Decision Trees
Florian Peter Busch, Moritz Kulessa, Eneldo Loza Mencía and Hendrik Blockeel
Prioritization of COVID-19 literature via unsupervised keyphrase extraction and document representation learning
Blaž Škrlj, Marko Jukič, Nika Eržen, Senja Pollak and Nada Lavrač
Ranking Structured Objects with Graph Neural Networks
Clemens Damke and Eyke Hüllermeier
An Ensemble Hypergraph Learning framework for Recommendation
Alireza Gharahighehi, Celine Vens and Konstantinos Pliakos
KATRec: Knowledge Aware aTtentive Sequential Recommendations
Seyed Danial Mohseni Taheri, Mehrnaz Amjadi and Theja Tulabandhula
HTML-LSTM: Information Extraction from HTML Tables in Web Pages using Tree-Structured LSTM
Kazuki Kawamura and Akihiro Yamamoto
Controlling BigGAN Image Generation with a Segmentation Network
Aman Jaiswal, Harpreet Singh Sodhi, Mohamed Muzamil H, Rajveen Singh Chandhok, Sageev Oore and Chandramouli Shama Sastry
Multi-Scale Sentiment Analysis of Location-Enriched COVID-19 Arabic Social Data
Tarek Elsaka, Imad Afyouni, Ibrahim Hashem and Zaher Al Aghbari
A Network Intrusion Detection System for Concept Drifting Network Traffic Data
Giuseppina Andresini, Annalisa Appice, Corrado Loglisci, Vincenzo Belvedere, Domenico Redavid and Donato Malerba
An Analysis of Performance Metrics for Imbalanced Classification
Jean-Gabriel Gaudreault, Paula Branco and João Gama
Local Interpretable Classifier Explanations with Self-generated Semantic Features
Fabrizio Angiulli, Fabio Fassetti and Simona Nisticò
Automatic human-like detection of code smells
Chitsutha Soomlek, Jan N. van Rijn and Marcello Bonsangue
Spatially-Aware Autoencoders for Detecting Contextual Anomalies in Geo-Distributed Data
Roberto Corizzo, Michelangelo Ceci, Gianvito Pio, Paolo Mignone and Nathalie Japkowicz
Statistical Analysis of Pairwise Connectivity
Georg Krempl, Daniel Kottke and Tuan Pham
Local Exceptionality Detection in Time Series Using Subgroup Discovery
Dan Hudson, Travis Wiltshire and Martin Atzmueller
Elliptical Ordinal Embedding
Aissatou Diallo and Johannes Fürnkranz
Calibrated Resampling for Imbalance and Long-Tails in Deep learning
Colin Bellinger, Roberto Corizzo and Nathalie Japkowicz
Deriving a Single Interpretable Model by Merging Tree-based Classifiers
Valerio Bonsignori, Riccardo Guidotti and Anna Monreale
Ensemble of Counterfactual Explainers
Riccardo Guidotti and Salvatore Ruggieri
A Semi-Supervised Framework for Misinformation Detection
Yueyang Liu, Zois Boukouvalas and Nathalie Japkowicz
GANs for tabular healthcare data generation: a review on utility and privacy
João Almeida, Ricardo Correia and Pedro Pereira Rodrigues
Unsupervised Feature Ranking via Attribute Networks
Urh Primožič, Blaž Škrlj, Sašo Džeroski and Matej Petković
A Sentence-level Hierarchical BERT Model for Document Classification with Limited Labelled Data
Jinghui Lu, Maeve Henchion, Ivan Bacher and Brian Mac Namee
Sentiment Nowcasting during the COVID-19 Pandemic
Ioanna Miliou, John Pavlopoulos and Panagiotis Papapetrou
Privacy risk assessment of individual psychometric profiles
Giacomo Mariani, Anna Monreale and Francesca Naretto
Predicting reach to find persuadable customers: improving uplift models for churn prevention
Théo Verhelst, Jeevan Shrestha, Denis Mercier, Jean-Christophe Dewitte and Gianluca Bontempi
The Case for Latent Variable vs Deep Learning Methods in Misinformation Detection: An Application to COVID-19
Caitlin Moroney, Evan Crothers, Sudip Mittal, Anupam Joshi, Tulay Adali, Christine Mallinson, Nathalie Japkowicz and Zois Boukouvalas
Knowledge discovery of the delays experienced in reporting covid19 confirmed positive cases using time to event models
Aleksandar Novakovic, Adele Marshall and Carolyn McGregor
Access the Conference Proceedings HERE
9:00-16:25 ADT (UTC -3H)
Preferences & Recommender Systems - Session Chair: | |
---|---|
09:10 | An Ensemble Hypergraph Learning framework for Recommendation - Alireza Gharahighehi, Celine Vens and Konstantinos Pliakos |
09:35 | KATRec: Knowledge Aware aTtentive Sequential Recommendations - Seyed Danial Mohseni Taheri, Mehrnaz Amjadi and Theja Tulabandhula |
10:00 | Unsupervised Feature Ranking via Attribute Networks - Urh Primožič, Blaž Škrlj, Sašo Džeroski and Matej Petković |
10:20 | Elliptical Ordinal Embedding - Aissatou Diallo and Johannes Fürnkranz |
Short Break (10:40)
Neural Networks & Deep Learning - Session Chair: | |
---|---|
11:55 | Consensus Based Vertically Partitioned Multi-Layer Perceptrons for Edge Computing - Haimonti Dutta, Saurabh Amarnath Mahindre and Nitin Nataraj |
12:20 | A Sentence-level Hierarchical BERT Model for Document Classification with Limited Labelled Data - Jinghui Lu, Maeve Henchion, Ivan Bacher and Brian Mac Namee |
Long Break (12:40)
Neural Networks & Deep Learning - Session Chair: Paula Branco | |
---|---|
14:00 | Neural Additive Vector Autoregression Models for Causal Discovery in Time Series- Bart Bussmann, Jannes Nys and Steven Latré BEST STUDENT PAPER AWARD |
14:25 | Spatially-Aware Autoencoders for Detecting Contextual Anomalies in Geo-Distributed Data - Roberto Corizzo, Michelangelo Ceci, Gianvito Pio, Paolo Mignone and Nathalie Japkowicz |
14:45 | Local Exceptionality Detection in Time Series Using Subgroup Discovery- Dan Hudson, Travis Wiltshire and Martin Atzmueller |
Short Break (15:05)
Streams - Session Chair: Gjorgji Madjarov | |
---|---|
15:20 | Incremental k-Nearest Neighbors Using Reservoir Sampling for Data Streams- Maroua Bahri and Albert Bifet |
15:45 | A Network Intrusion Detection System for Concept Drifting Network Traffic Data- Giuseppina Andresini, Annalisa Appice, Corrado Loglisci, Vincenzo Belvedere, Domenico Redavid and Donato Malerba |
16:05 | Statistical Analysis of Pairwise Connectivity- Georg Krempl, Daniel Kottke and Tuan Pham |
9:00-16:25 ADT (UTC -3H)
Classification - Session Chair: Michelangelo Ceci | |
---|---|
09:10 | Combining Predictions under Uncertainty: The Case of Random Decision Trees - Florian Peter Busch, Moritz Kulessa, Eneldo Loza Mencía and Hendrik Blockeel |
09:35 | Shapley-Value Data Valuation for Semi-Supervised Learning - Christie Courtnage and Evgueni Smirnov |
10:00 | A Semi-Supervised Framework for Misinformation Detection- Yueyang Liu, Zois Boukouvalas and Nathalie Japkowicz |
10:20 | An Analysis of Performance Metrics for Imbalanced Classification- Jean-Gabriel Gaudreault, Paula Branco and João Gama |
Short Break (10:40)
Applications - Session Chair: Dino Ienco | |
---|---|
11:55 | Automated Grading of Exam Responses: An Extensive Classification Benchmark- Jimmy Ljungman, Vanessa Lislevand, John Pavlopoulos, Alexandra Farazouli, Zed Lee, Panagiotis Papapetrou and Uno Fors |
12:20 | Automatic human-like detection of code smells- Chitsutha Soomlek, Jan van Rijn and Marcello Bonsangue |
Long Break (12:40)
Applications - Session Chair: Vitor Cerqueira | |
---|---|
14:00 | HTML-LSTM: Information Extraction from HTML Tables in Web Pages using Tree-Structured LSTM- Kazuki Kawamura and Akihiro Yamamoto |
14:25 | Predicting reach to find persuadable customers: improving uplift models for churn prevention- Théo Verhelst, Jeevan Shrestha, Denis Mercier, Jean-Christophe Dewitte and Gianluca Bontempi |
14:45 | Knowledge discovery of the delays experienced in reporting covid19 confirmed positive cases using time to event models - Aleksandar Novakovic, Adele Marshall and Carolyn McGregor |
Short Break (15:05)
COVID-19 - Session Chair: | |
---|---|
15:20 | Prioritization of COVID-19 literature via unsupervised keyphrase extraction and document representation learning - Blaž Škrlj, Marko Jukič, Nika Eržen, Senja Pollak and Nada Lavrač |
15:45 | Sentiment Nowcasting during the COVID-19 Pandemic- Ioanna Miliou, John Pavlopoulos and Panagiotis Papapetrou |
16:05 | Multi-Scale Sentiment Analysis of Location-Enriched COVID-19 Arabic Social Data - Tarek Elsaka, Imad Afyouni, Ibrahim Hashem and Zaher Al Aghbari |
9:00-16:25 ADT (UTC -3H)
Graphs - Session Chair: Nathalie Japkowicz | |
---|---|
09:10 | Ranking Structured Objects with Graph Neural Networks - Clemens Damke and Eyke Hüllermeier |
09:35 | FHA: Fast Heuristic Attack against Graph Convolutional Networks - Haoxi Zhan and Xiaobing Pei |
10:00 | Deriving a Single Interpretable Model by Merging Tree-based Classifiers - Valerio Bonsignori, Riccardo Guidotti and Anna Monreale |
10:20 | Local Interpretable Classifier Explanations with Self-generated Semantic Features - Fabrizio Angiulli, Fabio Fassetti and Simona Nisticò |
Short Break (10:40)
Responsible AI - Session Chair: Anna Monreale | |
---|---|
11:55 | Learning Time Series Counterfactuals via Latent Space Representations - Zhendong Wang, Isak Samsten, Rami Mochaourab and Panagiotis Papapetrou |
12:20 | Ensemble of Counterfactual Explainers - Riccardo Guidotti and Salvatore Ruggieri |
Long Break (12:40)
Responsible AI - Session Chair: | |
---|---|
14:00 | Leveraging Grad-CAM to Improve the Accuracy of Network Intrusion Detection Systems- *Francesco Paolo Caforio, Giuseppina Andresini, Gennaro Vessio, Annalisa Appice and Donato Malerba * |
14:25 | Privacy risk assessment of individual psychometric profiles- Giacomo Mariani, Anna Monreale and Francesca Naretto |
14:45 | The Case for Latent Variable vs Deep Learning Methods in Misinformation Detection: An Application to COVID-19- Caitlin Moroney, Evan Crothers, Sudip Mittal, Anupam Joshi, Tulay Adali, Christine Mallinson, Nathalie Japkowicz and Zois Boukouvalas |
Short Break (15:05)
Spatiotemporal - Session Chair: | |
---|---|
15:20 | Controlling BigGAN Image Generation with a Segmentation Network- Aman Jaiswal, Harpreet Singh Sodhi, Mohamed Muzamil H, Rajveen Singh Chandhok, Sageev Oore and Chandramouli Shama Sastry |
15:45 | GANs for tabular healthcare data generation: a review on utility and privacy - João Almeida, Ricardo Correia and Pedro Pereira Rodrigues |
16:05 | Calibrated Resampling for Imbalance and Long-Tails in Deep learning- Colin Bellinger, Roberto Corizzo and Nathalie Japkowicz |
Each accepted paper should be accompanied by a processing fee of 150 Canadian dollars.
The deadline for paying the registration fee is August 25, 2021. After that date, papers without a fee, will not be included in the proceedings.
Registration Details
REGISTRATION FEES ARE NON-REFUNDABLE
Name | Institution |
---|---|
Alberto Cano | Virginia Commonwealth University |
Albrecht Zimmermann | Université Caen Normandie |
André L.D. Rossi | São Paulo State University (Unesp) |
Anna Monreale | Computer Science Dep., University of Pisa |
Bernhard Pfahringer | University of Waikato |
Bruno Cremilleux | Universite de Caen Normandie |
Catarina Oliveira | INESC TEC |
Chedy Raïssi | INRIA |
Colin Bellinger | NRC |
Daniel Castro Silva | FEUP-DEI / LIACC |
Dino Ienco | IRSTEA |
Dragan Gamberger | Rudjer Boskovic Institute |
Elio Masciari | Federico II University |
Francesca Alessandra Lisi | University of Bari Aldo Moro |
George Papakostas | Human-Machines Interaction (HMI) Laboratory, Department of Computer and Informatics Engineering, EMT Institute of Technology |
Gianvito Pio | University of Bari Aldo Moro |
Giuseppe Manco | ICAR-CNR |
Gjorgji Madjarov | Faculty of Computer Science and Engineering, Ss. Cyril and Methodius University |
Herna Viktor | University of Ottawa |
Ioannis Tsamardinos | Computer Science Department, University of Crete |
Ivica Dimitrovski | Faculty of Computer Science and Engineering |
Jaakko Hollmén | Aalto University |
Jan Ramon | INRIA |
Jerzy Stefanowski | Poznan University of Technology, Poland |
Johannes Fürnkranz | Johannes Kepler University Linz |
Kazumi Saito | Univesity of Shizuoka |
Kouichi Hirata | Kyushu Institute of Technology |
Luis Teixeira | FEUP-DEI / INESC TEC |
Makoto Haraguchi | Hokkaido University |
Marina Sokolova | Faculty of Medicine, University of Ottawa and Institute for Big Data Analytics |
Martin Atzmueller | Osnabrueck University |
Michelangelo Ceci | Universita degli Studi di Bari |
Mohamed Gaber | Birmingham City University |
Nada Lavrač | Jozef Stefan Institute |
Nathalie Japkowicz | American University |
Nicola Di Mauro | Università di Bari |
Pance Panov | Jozef Stefan Institute |
Pascal Poncelet | LIRMM Montpellier |
Pedro Larranaga | University of Madrid |
Pedro Pereira Rodrigues | University of Porto |
Rafael Gomes Mantovani | Federal Technology University of Paraná, campus Apucarana |
Rita P. Ribeiro | University of Porto |
Ruggero G. Pensa | University of Torino, Italy |
Sageev Oore | Dalhousie University and Vector Institute |
Saso Dzeroski | Jozef Stefan Institute |
Stefan Kramer | Johannes Gutenberg University Mainz |
Tomislav Lipic | Rudjer Boskovic Institute |
Tomislav Smuc | Rudjer Boskovic Institute |
Vincenzo Lagani | Ilia State University |
Wouter Duivesteijn | Eindhoven University of Technology |
The event will be fully online. Details on attending the conference will be sent later to registered participants.
Abstract submission (extended): May 16, 2021 June 18, 2021
Full paper submission (extended): May 23, 2021 June 20, 2021
Notification (extended): July 20, 2021 July 28, 2021
Camera ready version, author registration (extended): August 8, 2021 August 11, 2021
Conference: October 11-13, 2021
The international conference on Discovery Science provides an open forum for intensive discussions and exchange of new ideas among researchers working in the area of Discovery Science. The conference focus is on the use of artificial intelligence methods in science. Its scope includes the development and analysis of methods for discovering scientific knowledge, coming from machine learning, data mining, intelligent data analysis, and big data analytics, as well as their application in various domains.
We invite submissions of research papers addressing all aspects of discovery science. We encourage papers that focus on the analysis of different types of massive and complex data, including structured, spatio-temporal and network data, as well as heterogeneous, continuous or imprecise data. We also encourage papers in the fields of computational scientific discovery, mining scientific data, computational creativity and discovery informatics. We welcome papers addressing applications of artificial intelligence in different domains of science, including biomedicine and life sciences, materials science, astronomy, physics, chemistry, as well as social sciences.
Possible topics include, but are not limited to:
Papers must be written in English and formatted according to the Springer LNCS guidelines. Papers should be submitted in PDF form via the DS 2020 Online Submission System at EasyChair. Once a paper has been submitted to the conference, changes to the author list are not permitted.
Submitted papers should not exceed 15 pages (long papers) and 10 pages (short ones), in total (including references). All submissions will be subject to review by the DS 2020 Program Committee. The Program Committee reserves the right to offer acceptance as Short Papers (10 pages in the Proceedings) to some Long Paper submissions. All accepted papers will appear in the conference proceedings published by Springer LNCS series and will have allocated time for oral presentation in the conference.
The reviews are single-blind. Authors do not need to anonymize their submission. Submitted papers may not have appeared in or be under consideration for another workshop, conference or journal. They may not be under review or submitted to another forum during the DS 2020 review process.
To be announced
The authors of a number of selected papers presented at DS 2021 will be invited to submit extended versions of their papers for possible inclusion in a special issue of Machine Learning journal (published by Springer) on Discovery Science. Fast-track processing will be used to have them reviewed and published.
There will be a Best Student Paper Award in the value of 555 Eur sponsored by Springer.
Springer Machine Learning journal
The Machine Learning journal invites submissions of research papers addressing all aspects of discovery science – a research discipline concerned with the development and analysis of methods for discovering scientific knowledge, coming from machine learning, data mining, intelligent data analysis, and big data analytics, as well as their application in various domains. Submissions addressing all aspects of discovery science are welcome. Research papers that focus on the analysis of different types of massive and complex data, including structured, spatio-temporal and network data, as well as heterogeneous, continuous or imprecise data are encouraged. Research papers in the fields of computational scientific discovery, mining scientific data, computational creativity and discovery informatics are also welcome. Finally, submissions addressing applications of artificial intelligence in different domains of science, including biomedicine and life sciences, materials science, astronomy, physics, chemistry, as well as social sciences are encouraged.
Possible topics include, but are not limited to:
Papers which, at the time of submission, have appeared in the proceedings of Discovery Science 2021 or other relevant conferences will be considered provided that the submission constitutes a significant contribution beyond the conference paper containing at least 30% of new material (e.g., extensions of the method, additional technical results, etc.) as compared to the conference version of the paper. The guest editors (based on the reviews) will make the decision on whether the difference is significant enough to warrant publication. The journal version should include a short paragraph explaining how it extends the previously published conference paper.
To submit to this issue, authors have to make a journal submission to the Springer Machine Learning journal following the instructions on the following link https://www.springer.com/journal/10994/updates/20356458. It is highly recommended that submitted papers do not exceed 20 pages including references. Every paper may be accompanied with unlimited appendices.
The papers should be formatted in the Springer journal style (svjour3, smallcondensed). The journal requires authors to include an information sheet as a supplementary material that contains a short summary of their contribution and specifically address the following questions:
Queries relating to the special issue should be sent to the guest editors at csoares@fe.up.pt and ltorgo@dal.ca