The 2nd International Workshop on Semantics-Powered Data Analytics (SEPDA 2017)

in conjunction with the IEEE International Conference on Bioinformatics and Biomedicine (BIBM 2017)
to be held at Kansas City, MO, USA
on November 13, 2017


Biomedical ontologies and controlled terminologies provide structured domain knowledge to a variety of health information systems. The rich thesaurus with concepts linked by semantic relationships has been widely used in natural language processing, data mining, machine learning, semantic annotation, and automated reasoning. The dramatically increasing amount of health-related data poses unprecedented opportunities for mining previously unknown knowledge with semantics-powered data analytics methods. However, due to the heterogeneity of different data sources, it is a challenging problem to exploit multiple sources to solve real-world problems such as designing cost-effective treatment plan for patients, designing generalizable clinical trials, drug repurposing, and clinical phenotyping. The goal of this workshop is to bring people in the field of knowledge representation, knowledge management, and health data analytics to discuss innovative semantic methods, applications, and data analytics to address problems in healthcare, biomedicine, public health, and clinical research with biomedical, clinical, behavioral, and social web data.

We are inviting original research submissions (FULL 8 pages), work-in-progress (SHORT 4 pages), and poster abstracts (2 pages, NEW TYPE).

All the accepted submissions will be presented in SEPDA 2017 and published in the IEEE BIBM 2017 Proceedings (in IEEE Xplore Digital Library); Selected FULL/SHORT papers will be invited to publish an extended version in the supplement of BMC Medical Informatics & Decision Making (IF: 1.643; Indexed in SCIE, MEDLINE, and PMC; and subject to an article processing charge of £1,122). Selected high-quality SHORT papers will also be invited to submit an extended version to the journal supplement for consideration.

Note: If a paper is selected for possible journal publication, the authors will be asked to shorten their workshop paper to be published in the BIBM 2017 Proceedings and then submit the journal version after the workshop. The authors can still choose to publish their full papers in the conference proceedings, in which case however, the authors will NOT be eligible to publish in the journal supplement due to the journal's self-plagiarism concern.

A new review process: To mitigate the biases and reduce the burden of the reviewer community, we will employ a hybrid approach in the review process of this workshop. We will invite external reviewers (in the Cyberchair system) as well as contributing authors to review the submissions (in the Pevals system). The final decision will be made based on the review reports from both systems. As such, the organizers, the external reviewers, as well as contributing authors will have a shared responsibility to maintain a high standard for the review process and ensure the quality of the workshop papers. To accommodate the busy schedule of the authors, you can choose the number of submission to review or not to participate in the peer-review in Pevals. In such a case, the decision will be made based on the external reviewers' review reports. (More details about the new review process)


We are inviting original research submissions as well as work-in-progress.

Topics of interest include but not limited to:

  • Ontologies and Controlled Terminologies
    • Ontology development and enrichment
    • Quality assurance of ontologies and controlled terminologies
    • Semantic harmonization and ontology alignment
    • Knowledge representation and reasoning
  • Semantics-based Data Analytics
    • Ontology-based text mining and natural language processing
    • Ontology-based analysis on biomedical, clinical, or social web data
    • Information Extraction on biomedical, clinical, or social web data
    • Data mining or machine learning on biomedical, clinical or social web data
    • Semantic annotation on biomedical, clinical or social web data
  • Data Integration
    • Linked open data
    • Novel approaches for data integration of heterogenous data sources
    • Large scale data integration
  • Application
    • Novel tools and ontologies for data interpretation and visualization
    • Pharmacovigilance and drug repurposing using ontologies
    • Clinical trial generalizability assessment using ontologies
    • Algorithmic phenotyping and cohort identification using ontologies
    • Improving the literacy of health information consumers

Important Dates

Sept 27, 2017: Due date for full / work-in-progress workshop papers submission

Oct 15, 2017: Notification of paper acceptance to authors

Oct 25, 2017: Camera-ready of accepted papers

Nov 13, 2017: SEPDA 2017 Workshop

Journal submission deadline: TBD


Please submit a full-length paper (up to 8 page IEEE 2-column format) through the online submission system (you can download the format instruction here ). Electronic submissions (in PDF or Postscript format) are required. Selected participants will be asked to submit their revised papers in a format to be specified at the time of acceptance.

Submission site: Submit a New Paper (


  • Organizing Committee
    • Zhe He, School of Information, Florida State University, USA
    • Cui Tao, School of Biomedical Informatics, University of Texas Health Center at Houston, USA
    • Rui Zhang, Institute for Health Informatics, University of Minnesota, USA
    • Jingshan Huang, School of Computing, University of South Alabama, USA
    • Jiang Bian, Biomedical Informatics, Health Outcomes & Policy, College of Medicine, University of Florida, USA
  • Program Committee
    • GQ Zhang, University of Kentucky, USA
    • James Geller, New Jersey Institute of Technology, USA
    • Guoqian Jiang, Mayo Clinic, USA
    • Zhihui Luo, University of Wisconsin, Milwaukee, USA
    • Tianyong Hao, The cisco School of Informatics, Guangdong University of Foreign Studies, China
    • Amanda Hicks, University of Florida, USA
    • Yongqun “Oliver” He, University of Michigan, USA
    • Fernando Martin-Sanchez, Weill Cornell College of Medicine, USA
    • James Cimino, University of Alabama at Birmingham, USA
    • Licong Cui, University of Kentucky, USA
    • Xia Jing, Ohio University, USA
    • Riccardo Miotto, Icahn School of Medicine at Mount Sinai, USA
    • Olivier Bodenreider, US National Library of Medicine, USA
    • Fleur Mougin, University of Bordeaux, France
    • Praveen Chandar, IBM Research, USA
    • Yonghui Wu, University of Florida, USA
    • Ronald Cornet, University of Amsterdam, Netherlands
    • Yi Guo, University of Florida
    • Juan Antonio Lossio-Ventura, University of Florida
    • Buzhou Tang, Intelligent Computing Research Center, Harbin Institute of Technology Shenzhen, China
    • Zhengxing Huang, Zhejiang University, China

Review Process

Introduction to the new peer-review system – Pevals (Participant-engaged evaluation system)

The review system we will use for this workshop is similar to the one NSF used in their 2013 pilot program to test an alternative review system ( Notable changes include using less reviewers for each paper and adding a discussion period before reviewers submitting the final scores.

The review process consists of the following steps:

Step 1: Each contributing author will be assigned to review and rank two or four other papers. Review assignments will be made so as to avoid organizational or individual conflicts-of-interest.

Step 2: All authors must complete their first round of review and ranking of the assigned papers within 10 days of the date of their assignment. 

Step 3: The authors will then have an opportunity to write rebuttals for the reviewers’ comments made to their papers and discuss with the anonymous reviewers about the comments and their replies. This step will take about three days.

Step 4: Each author will then re-evaluate the papers assigned to them based on the new information collected in the discussion period and update their scores and rankings. This step will take another three days.

Step 5: A composite ranking of all papers will be determined, and each author’s paper ranking will be adjusted based on a measure of the “quality” of the reviews provided by the author.  The adjustment is designed to provide an incentive to all authors to do an honest and thorough job of reviewing the papers to which they are assigned.

Step 6: Final acceptance/rejection decision making will be done by the workshop organizer as currently done.

Note: failure to complete this review and ranking within the allotted times will result in the disqualification of the author’s own paper.

Anonymity of reviewers will be preserved as authors will not know which of the other authors review their papers.

The advantages of adopting the new peer-review system are the following:

1. Your papers will not be reviewed by someone who does not have sufficient expertise in the topic area.
2. As you will commit significant time and efforts in reviewing the papers assigned to you, you would expect other authors who review your papers more committed than the reviewers of the current system.
3. You have the opportunity to clarify any misunderstanding the reviewers may have on your paper through interactive discussions. Such communications are expected to improve the overall review quality.
4. Potential review bias will be minimized since biases will have serious consequences, which is lacking in current review system.
5. The overall review process is usually much faster than current review system.



Accepted Papers



At least one author of an accepted paper needs to register as one full registration in order to have the paper published in the conference proceedings.


Please email Workshop co-chair, Zhe He (

Past Workshops