Submissions are solicited for the 2016 edition of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECMLPKDD). The conference provides an international forum for the discussion of the latest high-quality research results in all areas related to machine learning and knowledge discovery in databases and other innovative application domains. The 2016 conference will take place in Riva Del Garda, Italy, September 19-23. This edition will feature a full day of plenary presentations, for papers of general interest to the whole community, and two days of parallel sessions. All papers will be presented both orally and as posters. Papers on all topics related to machine learning, knowledge discovery, and data mining are invited.
Electronic submissions will be handled via CMT at the following address: https://cmt.research.microsoft.com/ECMLPKDD2016/. Please note that user accounts in each CMT conference is independent of other conferences so you will need to create a new account.
Abstracts need to be registered by Friday April 1, 2016, 23:59 Central European Time, and full submissions will be accepted until Monday April 4, 2016, 23:59 Central European Time.
Papers must be written in English and formatted according to the Springer LNAI guidelines. Author instructions, style files and copyright form can be downloaded at: http://www.springer.de/comp/lncs/authors.html.
The maximum length of papers is 16 pages in this format. Overlength papers will be rejected without review (papers with smaller page margins and font sizes than specified in the author instructions and set in the style files will also be treated as overlength).
Up to 10 MB of additional materials (e.g. proofs, audio, images, video, data or source code) can be attached to the submission. Note that the reviewers and the program committee reserve the right to judge the paper solely on the basis of the 16 pages of the paper; looking at any additional material is up to the discretion of the reviewers and is not required.
The review process is single-blind (authors identities known to reviewers). Submissions will be evaluated on the basis of technical quality, novelty, potential impact, and clarity. Authors will have the opportunity to point out factual errors, obvious mistakes, or misconceptions by reviewers during a rebuttal phase following the release of initial reviews.
Papers submitted should report original work. ECML PKDD 2016 will not accept any paper that, at the time of submission, is under review or has already been accepted for publication in a journal or another conference. Authors are also expected not to submit their papers elsewhere during the review period. The dual submissions policy applies during the whole ECMLPKDD 2016 reviewing period from April 1 to June 20, 2016.
Authors are encouraged to adhere to the best practices of Reproducible Research (RR), by making available data and software tools for reproducing the results reported in their papers. Authors of accepted papers may flag their submissions as RR and make software and data accessible to reviewers and to the program committee who will verify the accessibility of software and data. Links to data and code will be then inserted in the final version of RR papers. For the sake of persistence and proper authorship attribution, we require the use of standard repository hosting services such as Dataverse, mldata.org, OpenML, etc. for data sets, and mloss.org, Bitbucket, GitHub, etc. for source code. If data or code gets updated after the paper is published, it is important to enable researchers to access the versions that were used to produce the results reported in the paper.
The conference proceedings will be published by Springer in the Lecture Notes in Artificial Intelligence series (LNAI). There will be no difference in the proceedings between papers presented in parallel sessions and papers presented during the plenary day.
In addition to normal conference submissions, papers can be submitted to the ECMLPKDD 2016 journal track. Accepted papers will be presented at the conference and published either in Machine Learning or in Data Mining and Knowledge Discovery. For information about the journal track, please see the separate Journal Track call for papers.
For any additional questions you can contact the Program Chairs (Paolo Frasconi, Niels Landwehr, Giuseppe Manco, Jilles Vreeken) at email@example.com.
We invite submissions for the journal track of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECMLPKDD) 2016. The conference provides an international forum for the discussion of the latest high-quality research results in all areas related to machine learning, data mining, and knowledge discovery. The conference will take place in Riva Del Garda, Italy, September 19-23, 2016. This edition will feature a full day of plenary presentations — for papers of general interest to the whole community — and two days of parallel sessions. All accepted papers will be presented both orally and as posters.
Papers on all topics related to machine learning, knowledge discovery, and data mining are invited. However, given the special nature of the journal track, only papers that satisfy the quality criteria of journal papers and at the same time lend themselves to conference talks will be considered. This implies that journal versions of previously published conference papers, or survey papers will not be considered for the special issue. Papers that do not fall into the eligible category may be rejected without formal reviews but can of course be resubmitted as regular papers. Authors are encouraged to adhere to the best practices of Reproducible Research (RR), by making available data and software tools for reproducing the results reported in their papers. For the sake of persistence and proper authorship attribution, we require the use of standard repository hosting services such as Dataverse, mldata.org, OpenML, etc. for data sets, and mloss.org, Bitbucket, GitHub, etc. for source code. Authors who submit their work to the special ECMLPKDD issues of the journals commit themselves to present their results at the ECMLPKDD 2016 conference in case of acceptance.
To submit to this track, authors have to make a journal submission to either the Springer Data Mining and Knowledge Discovery journal or the Springer Machine Learning journal, and indicate that the submission is for the "ECMLPKDD 2016" special issue. It is recommended that submitted papers do not exceed 20 pages including references and appendices, formatted in the Springer journal slyle (svjour3,smallcondensed). This is a soft limit, but if a submission exceeds the limit, please provide a brief justification regarding the length in the cover letter.
Authors are required to include a cover letter containing a short summary of their contribution (2 pages max), where they address the following questions:
The journal track allows continuous submissions from the end of September 2015 to March 2016. We expect two cutoffs per month on which we distribute papers to reviewers with the first one being the 27th of September 2015 and the last one being the 30th of March 2016. For papers accepted without revisions we strive for a turn around time of two months. This means that we should be able to include all of those submissions in the special issue. However, in the last years there has often been the need for revisions and we therefore recommend to submit papers as early as possible.
The 2016 cut-off dates for the bi-weekly batches are: Jan 3, Jan 17, Jan 31, Feb 14, Feb 28, Mar 13, Mar 30.
We solicit submissions for demos for ECML PKDD 2016 in Riva del Garda, Italy. Submissions must describe working systems and be based on state-of-the-art machine learning and data mining technology. These systems may be innovative prototype implementations or mature systems that use machine learning techniques and knowledge discovery processes in a real setting. Systems that use basic statistics are not acceptable. A commercial software is not acceptable.
The accepted papers for demos will be included in the conference proceedings, to be published by Springer Verlag in the "Lecture Notes in Artificial Intelligence" (LNAI) Series. The demos will be presented in a special demonstration session. At least one of the demo submitters must register for the conference, and perform the demo on site.
All aspects of the submission and notification process will be handled online via the conference CMT submission site.
Please choose the right track during the submission. Instructions concerning the submission (except for the "reproducible research" part which is not relevant for a demo paper), camera-ready formatting and copyright transfer for conference papers also hold for demo papers, unless otherwise specified.
A demonstration submission must be up to 4 pages long. It must provide adequate information on the system's components and the way the system is operated, including e.g. screenshots.
Submitters should keep in mind that the description of a demo has inherently different content than a research paper submitted to the main conference. A successful demonstration paper provides satisfactory answers the following questions:
The formatting guidelines of Springer Verlag for the LNAI series apply, and the author instructions and style files under http://www.springer.de/comp/lncs/authors.html must be used. For inquiries concerning submissions please contact the Demo Track Chairs.
The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECMLPKDD) provides an international forum for the discussion of the latest high-quality research results in all areas related to machine learning and knowledge discovery in databases and related application domains.
The goal of the Nectar Track, started in 2012, is to offer conference attendees a compact overview of recent scientific advances at the frontier of machine learning and data mining with other disciplines, as published in related conferences and journals. For researchers from the other disciplines, the Nectar Track offers a place to present their work to the ECMLPKDD community and to raise the community's awareness of data analysis results and open problems in their field. We invite senior and junior researchers to submit summaries of their own work published in neighbouring fields, such as (but not limited to) artificial intelligence, data analytics, bioinformatics, games, computational linguistics, natural language processing, computer vision, geoinformatics, health informatics, database theory, human computer interaction, information and knowledge management, robotics, pattern recognition, statistics, social network analysis, theoretical computer science, uncertainty in AI, network science, complex systems science, and computationally oriented sociology, economy and biology, as well as critical data science/studies.
Particularly welcome is work that summarises a line of work that comprises older and more recent papers. The described work should be relevant to a broad audience within ECMLPKDD, and (a) illustrate the pervasiveness of data-driven exploration and modelling in science, technology, and the public, as well as innovative applications, and/or (b) focus on theoretical results.
Note that papers focusing only on software implementations rather than on the interdisciplinary use of ML/DM should rather be submitted to the demo track. Work at the core of ML/DM should target the main tracks of ECMLPKDD rather than the Nectar Track.
Papers must be 4 pages and should be formatted according to the Author instructions, style files and copyright form that can be found at http://www.springer.de/comp/lncs/authors.html.
Submissions must clearly indicate which corresponding original publication(s) are presented, and must clearly motivate the relevance of the work in the context of machine learning and data mining. Papers should be submitted through the conference CMT submission system (select from the menu the Nectar track). Accepted Nectar contributions will be presented as oral presentations and included in the conference proceedings.
In case you have any question, please do not hesitate to contact the Nectar Track Chairs (Bettina Berendt, Pauli Miettinen) at firstname.lastname@example.org. We are looking forward to your proposals.
The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD, provides an international forum for the discussion of the latest high-quality research results and applications in all areas related to machine learning, data mining and knowledge discovery in databases, as well as other innovative application domains. The 2016 edition of ECML PKDD will take place in Riva del Garda, Italy, September 19-23.
The INDUSTRIAL, GOVERNMENTAL & NGO Track of ECML PKDD 2016 follows the success of the previous year with a separate Program Committee and a separate Proceedings volume. The track aims to bring together participants from academia, industry, governments and NGOs (non-governmental organizations) in a venue that highlights practical and real-world studies of machine learning, knowledge discovery and data mining. This track wants to encourage mutually-beneficial links between those engaged in scientific research and practitioners working to improve big data mining and large scale machine learning analytics. Novel ideas, controversial issues, open problems and comparisons of competing approaches are strongly encouraged. Experiences from practitioners provide crucial input into future research directions and allow others to learn from successes and failures.
Submissions are invited on innovative real-world data systems and applications, state-of-the-art practices, surveys from real-world projects and industrial experiences, and identification of unsolved research challenges of machine learning, knowledge discovery and data mining. The INDUSTRIAL, GOVERNMENTAL & NGO track is distinct from the Research Track in that submissions solve real-world problems and focus on applications and challenges. Submissions must clearly identify one of the following three areas they fall into: "Engineering Systems", "Data Science", or "Challenges".
The criteria for submissions is the following:
The INDUSTRIAL, GOVERNMENTAL & NGO Track proceedings of ECML/PKDD 2016 will be published by Springer as a specific volume of the Proceedings of ECML/PKDD 2016, in the Lecture Notes in Artificial Intelligence series (LNAI). At least one of the authors of each accepted paper must register for the conference to present the paper on site.
The papers must be written in English and formatted according to the Springer LNAI guidelines. Author's instructions and style files can be downloaded at: http://www.springer.de/comp/lncs/authors.html
The maximum length of papers is 16 pages in this format. Longer papers will be rejected without review (papers with smaller page margins and font sizes than specified in the author instructions and set in the style files will also be treated as over length). Papers submitted should report original work; ECML PKDD 2016 will not accept any paper which, at the time of submission, is under review or has already been accepted for publication in a journal or another conference. Authors are also expected not to submit their papers elsewhere during the review period.
To submit a paper:
Submissions will be evaluated on the basis of relevance, novelty, originality, technical soundness, clarity, empirical and/or practical validation, external significance and validity, quality and consistency of presentation, and appropriate comparison to related work. Special emphasis will be placed on the relevance of the proposed contribution to practitioners. Authors are strongly encouraged to make data and code publicly available whenever possible.
Authors of papers submitted to the INDUSTRIAL, GOVERNMENTAL & NGO Track of ECML PKDD 2016 must identify the application domain that is the subject of their paper. Example application domains are finance, government, e-commerce, retail, mobile, medicine, healthcare, security, public policy, science, engineering, law, manufacturing, and telecommunications.
The INDUSTRIAL, GOVERNMENTAL & NGO Track of ECML PKDD 2016 has a separate Program Committee from the Research Track of ECML PKDD 2016. Papers submitted to the INDUSTRIAL, GOVERNMENTAL & NGO Track of ECML PKDD 2016 will be reviewed by at least three referees. The review process is single-blind (reviewer identities unknown to authors) and there will be no opportunity for author rebuttal. This decision was made to minimize reviewer workload and to concentrate it in time, which may ultimately result in better quality reviews and decisions. If necessary, a discussion will take place among the reviewers of a paper until a decision is reached.
You can contact the Industrial Track Chairs (Björn Bringmann, Gemma Garriga, Volker Tresp) at email@example.com.
The ECML-PKDD 2016 Organizing Committee invites proposals for half-day tutorials to be held on the first and last days of the conference (September 19 and 23, 2016), which will take place in Riva del Garda, Italy.
Tutorials are intended to provide a comprehensive introduction to established or emerging research topics of interest for the machine learning and the data mining community. These topics include related research fields or applications but also well-developed tools and suites that support ML/DM research. The ideal tutorial should attract a wide audience. It should be broad enough to provide a basic introduction to the chosen area, but it should also cover the most important topics in depth. Each tutorial should be well-focused so that its content can be covered in a half-day slot. Proposals that exclusively focus on the presenter’s own work or commercial demonstrations are strongly discouraged.
Tutorial slides will be made available online at the ECML-PKDD website, although authors can provide them on additional websites as well. Depending on the budget, we expect to offer a waived registration fee to one speaker of each accepted tutorial (i.e., no guarantee).
Tutorial proposals should contain at least the following:
Especially for a relatively novel but rapidly maturing topic, a half day tutorial (4h incl. one 30 minute break) followed by a half day workshop (4h incl. one 30 minute break) could be a good format. We kindly ask you to write a single proposal for the tutorial and workshop that covers the different guidelines and requirements for tutorials and workshops. In addition, the links should be clearly described and the proposal should be submitted to the workshop and tutorial chairs (select both 'Workshop' and 'Tutorial' as topics). For workshop guidelines, please see the separate Call For Workshop Proposals.
The proposal will be reviewed by the workshop and tutorial co-chairs, who may use the help of external reviewers, expert on the submission topics.
The features that will be evaluated are:
Please submit your tutorial proposals in PDF format using EasyChair.
The following deadlines are important for the tutorial organizers:
In case you have further questions, please do not hesitate to contact the Workshop and Tutorial Chairs (Matthijs van Leeuwen, Fabrizio Costa, Albrecht Zimmermann) at firstname.lastname@example.org. We are looking forward to your proposals.
The ECML-PKDD 2016 Organizing Committee invites proposals for workshops to be held on the first and last days of the conference (September 19 and 23, 2016), which will take place in Riva del Garda, Italy. We invite proposals for both full- and half-day workshops in current and emerging topics in machine learning and data mining.
Workshops provide an opportunity to discuss novel topics in a small and interactive atmosphere. They can concentrate in-depth on research topics, but can also be devoted to application issues, or to questions concerning the economic and social aspects of machine learning and data mining. Multidisciplinary workshops that bring together researchers and practitioners from different communities are particularly welcome.
If the budget permits, one organizer or invited speaker of an accepted workshop will be offered the possibility of waived registration fee for attendance on the workshop day (only).
We welcome both full- and half-day workshop proposals. Full-day workshops will have a program of typically 8 hours including two 30-minute coffee breaks and a 90-minute lunch break. Half-day workshops will have a 4 hours program with a 30-minute coffee break.
We would like to encourage proposers to aim for a program that is both varied and interesting. Especially where the format of the workshop is concerned, we would like you to think about ways of going beyond the usual list of presentations of accepted papers. Keep in mind that the main conference is necessarily more time-constrained and workshops therefore allow for group explorations of interesting topics, for example by means of discussions, demo sessions, invited talks, and panels.
Another way of extending the usual format is to include a specific challenge problem that can be addressed by the workshop participants, with a dedicated challenge session in the workshop program. Note, however, that the challenge should be only one of the components of the workshop, targeting a problem which is specific to the workshop topic(s).
For some workshops, it may be useful to first present an introduction to the state-of-the-art in the field given by experienced invited presenters, and afterwards discuss more technical or novel work in a standard (or non-standard!) workshop setting.
Especially for a relatively novel but rapidly maturing topic, giving the afore-mentioned introduction of the state-of-the-art may go beyond the scope of an invited presentation. In this case, a half day tutorial (4h incl. one 30 minute break) followed by a half day workshop (4h incl. one 30 minute break) could be a good format. We kindly ask you to write a single proposal for the tutorial and workshop that covers the different guidelines and requirements for tutorials and workshops. In addition, the links should be clearly described and the proposal should be submitted to the workshop and tutorial chairs (select both 'Workshop' and 'Tutorial' as topics). For tutorial guidelines, please see the separate Call For Tutorial Proposals.
Workshop proposals should contain the necessary information for the workshop chairs and reviewers to judge the importance, quality, and community interest in the proposed topic (a minimum of 15-20 expected participants is required). Each workshop should have one or more designated organisers and a program. When proposing a workshop, please provide (at least) the following information:
Please submit your workshop proposals in PDF format using EasyChair.
Proposals will be reviewed in close collaboration with the conference chairs and the program committee.
The following deadlines are important for the workshop organizers:
For paper submission, reviewing and final revisions, please consider the following deadlines:
These deadlines are somewhat flexible, but consider as constraints that the paper submission deadline should be after conference author notification (June 20) and acceptance notification should be before the conference early registration deadline (July 30).
In case you have further questions, please do not hesitate to contact the Workshop and Tutorial Chairs (Matthijs van Leeuwen, Fabrizio Costa, Albrecht Zimmermann) at email@example.com. We are looking forward to your proposals.
The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECMLPKDD) includes a PhD Forum on machine learning and knowledge discovery.
The purpose of this forum is to provide an environment specifically for junior PhD students to exchange ideas and experiences with peers in an interactive atmosphere and to get constructive feedback from senior researchers in data mining, machine learning, and related areas.
The focus of the discussion at the PhD Forum would be the work in progress of junior PhD students, with 1-3 years of research experience, towards their dissertation.
During the forum, senior researchers with experience in supervising and examining PhD students will be able to participate and provide feedback and advice to the participants.
It will be an excellent opportunity for developing person-to-person networks to the benefit of the PhD students in their future careers.
The PhD Forum spans various topics of data mining, machine learning, and work in related fields such as databases, artificial intelligence, statistics, information retrieval, multimedia and the Web. Topics in specific domains such as bioinformatics and the more general science informatics are also encouraged. Participants with interdisciplinary work across the areas are particularly welcome.
The PhD forum is open for two types of submissions:
The papers must be written in English and formatted according to the Springer LNAI guidelines.
Author instructions and style files can be downloaded at: http://www.springer.de/comp/lncs/authors.html.
To submit, please go to the CMT submission system at https://cmt.research.microsoft.com/ECMLPKDD2016/, and select from the menu the PhD track.
For any additional questions you can contact the PhD Forum Chairs (Leman Akoglu, Tijl De Bie) at firstname.lastname@example.org.
This year Discovery Challenge has a quite intriguing set of competitions for those researchers that want to prove their ability in solving real-life problems. In particular we have four different context for participans:
The ECML/PKDD Discovery Challenge 2016 on Bank Card Usage Analysis asks you to predict the user behavior of the OTP Bank Hungary, a key bank in CEE Region. We give you one year list of card payment events with geolocation information.
The Bank wants to know wich branch will be visited by each customer to be able to optimize proactive contact list and plan distribution.
The customer will be proactively called in campaigns from the branch that will be visited with the highest probability. The bank expects higher conversion rates in branch campaigns if the call is made in the branch mostly prefered by the customer.
Obesity, depression, stroke, falls, cardiovascular and musculoskeletal disease are some of the biggest health issues and fastest-rising categories of health-care costs. The financial expenditure associated with these is widely regarded as unsustainable and the impact on quality of life is felt by millions of people in the UK each day. Smart technologies can unobtrusively quantify activities of daily living, and these can provide long-term behavioural patterns that are objective, insightful measures for clinical professionals and caregivers.
To this end the EPSRC-funded “Sensor Platform for HEalthcare in Residential Environment (SPHERE)” Interdisciplinary Research Collaboration (IRC) has designed a multi-modal sensor system driven by data analytics requirements. The system is under test in a single house, and will be deployed in a general population of 100 homes in Bristol (UK). The data sets collected will be made available to researchers in a variety of communities.
Data is collected from the following three sensing modalities:
Prizes will be awarded to the first three winners:
Due to the extended use of Web forums, such as Yahoo! Answers or Stackoverflow, there has been a renewed interest in Community Question Answering (cQA). cQA combines traditional question answering with a modern Web scenario, where users pose questions hoping to get the right answers from other users. The most critical problem arises when a new question is asked in the forum. If the user's question is similar (even semantically equivalent) to a previously posted question, she/he should not wait for answers or for another user to address her/him to the relevant thread already archived in the forum. An automatic system can search for previously-posted relevant questions and instantaneously provide the found information.
In this challenge, given a new question and a set of questions previously posted to a forum, together with their corresponding answer threads, a machine learning model must rank the forum questions according to their relevance against the new user question.
Even if this task involves both Natural Language Processing (NLP) and Information Retrieval, the challenge focuses on the machine learning aspects of reranking the relevant questions. Therefore, we provide both the initial rank and the feature representation of training and test examples to the participants. We extract features from the text of the user and forum questions using advanced NLP techniques, e.g., syntactic parsing. Most interestingly, we also provide the Gram matrices of tree kernels applied to advanced structural tree representation. A few other features express the relevance of the thread comments, associated with the forum questions, against the user question.
Participants are expected to exploit these data for building novel and effective machine learning models for reranking the initial question list in a better rank according to Mean Average Precision (MAP).
Prizes will be awarded to the two best performing teams:
In recent years, there have been many proposals pushing for the use of Machine Learning (ML) in automatic network management. This challenge is one of the first explorations of ML for automatic network analysis. Our goal is to promote the use of ML for network-related tasks in general and, at the same time, to assess the participants' ability to quickly build a learning-based system showing a reliable performance. Additionally, one difficulty of using ML for network-related applications is the lack of datasets for training and evaluating different algorithms. The challenge provides one of the few datasets for this field, which may become a reference point for future and more advanced research.
As this is one of the first initiative in network classification, we started with a relatively simple multi-class single label classification task, where the labels are standard applications and signals are static network parameters. A more detailed description can be found on the challenge website.
The best-scoring submission will receive a prize of 1000 euros.
Note: All deadlines are at 23:59 CET (Central European Time) unless otherwise stated.