6th International Workshop on Semantic Statistics

SemStats 2018


Software Systems



The goal of this workshop is to explore and strengthen the relationship between the Semantic Web and statistical communities, to provide better access to the data held by statistical offices. It will focus on ways in which statisticians can use Semantic Web technologies and standards in order to formalize, publish, document and link their data and metadata, and also on how statistical methods can be applied on linked data. It is the fifth workshop in a series that started at the International Semantic Web Conference in 2013 (SemStats 2013) and run since every year at ISWC (2014, 2015, 2016, 2017).
The statistical community shows more and more interest in the Semantic Web. In particular, initiatives have been launched to develop semantic vocabularies representing statistical classifications and discovery metadata. Tools are also being created by statistical organizations to support the publication of dimensional data conforming to the Data Cube W3C Recommendation. But statisticians see challenges in the Semantic Web: how can data and concepts be linked in a statistically rigorous fashion? How can we avoid fuzzy semantics leading to wrong analyses? How can we preserve data confidentiality?
The workshop will also cover the question of how to apply statistical methods or treatments to linked data, and how to develop new methods and tools for this purpose. Except for visualization techniques and tools, this question is relatively unexplored, but the subject will obviously grow in importance in the near future.
Motivation
The interest of the statistical community for linked data has recently increased in a spectacular way. Two illustrations of this phenomenon can be mentioned:
The UNECE High-level group for the modernization of official statistics, a group of ten directors of national or international statistical institutes around the world, launched in 2017 a project on implementing statistical standards. Two work packages of this project aim at implementing linked statistical metadata systems (classifications, glossaries, statistical models).
Eurostat, as part of its "Vision 2020" strategic program, has started a major project focusing on digital communication, user analytics and innovative products. Work package 3 of this project contains different tasks related directly to linked data.
The National Statistics Institutes of ​Japan, Ireland​, and ​Italy along with the ​Scottish Government and ​DCLG in the UK have opened up their statistical data using linked data technologies.
There is also a significant interest in exploiting linked statistical data inside public administrations in order to create innovative public services for citizens and businesses: see for example the new EU-funded H2020 OpenGovIntelligence project.
This growing interest is a tremendous opportunity for the SemStats community to leverage the work done in the previous years and to continue to elaborate the solutions that are needed for these initiatives.
Topics
The workshop will address topics related to statistics and linked data. This includes but is not limited to:
How to publish linked statistics?
- What are the relevant vocabularies for the publication of statistical data?
- What are the relevant vocabularies for the publication of statistical metadata (code lists and classifications, descriptive metadata, provenance and quality information, etc.)?
- What are the existing tools? Can the usual statistical software packages (e.g. R, SAS, Stata) do the job?
- How do we include linked data production and publication in the data lifecycle?
- How do we establish, document and share best practices?
How to use linked data for statistics?
- Where and how can we find statistics data: data catalogues, dataset descriptions, data discovery?
- How do we assess data quality (collection methodology, traceability, etc.)?
- How can we perform data reconciliation, ontology matching and instance matching with statistical data?
- How can we apply statistical processes on linked data: data analysis, descriptive statistics, estimation, correction?
- How to intuitively represent statistical linked data: visual analytics, results of data mining?
How to use statistical methods on IoT data streams?
- How can statistical processes be applied to Sensor streaming data at runtime and how can the results of these processes be stored and accessed?
- How can statistical and machine learning algorithms be used on time series data produced by IoT devices
Contributions
This workshop is aimed at an interdisciplinary audience of researchers and practitioners involved or interested in Statistics and the Semantic Web. All contributions must represent original and unpublished work that is not currently under review. Contributions will be evaluated according to their significance, originality, technical content, style, clarity, and relevance to the workshop.
At least one author of each accepted contribution is expected to attend the workshop. Workshop participation is available to ISWC 2018 attendants at an additional cost.
Full and Short articles (up to 12 and 6 ‘pages’)
The workshop will welcome long and short scientific articles related to the topics mentioned above. Long articles refer to mature research work, where ideas have been implemented and evaluated. Short articles refer to brave new ideas or position statements describing a vision for the Semantic Statistics community.
Challenge articles (up to 12 ‘pages’)
The workshop will also feature a data challenge based on a corpus of linked metadata that will be made available on the SemStats web site by the end of May. The challenge will consist in the realization of mashups or visualizations, but also on comparisons, alignment and enrichment of the data and concepts involved.
Application and Demo articles (up to 6 ‘pages’)
This year, the workshop calls for contributions more generally. This includes interactive demonstrations of applications, or useful and relevant software library and repository, described in short articles. All application and demo articles should include a link where readers can experiment with the live software. Additional pointers such as source code repository are also welcomed.
Writing your contribution
All contributions must be written in English and must be formatted according to the information for LNCS Authors (see http://www.springer.com/computer/lncs?SGWID=0-164-6-793341-0). Please note that HTML+RDFa contributions are also welcome as long as the layout complies with the LNCS style. Authors are welcome to use dokieli (source) or similar systems. Contributions are not anonymous. Please share your contributions through Easychair and before August 6, 2018, 23:59 Hawaii Time. All accepted articles will be archived in an electronic proceedings published by CEUR-WS.org.