National Research Data Archive
MIDAS: development decisions
and usage peculiarities
Saulius Maskeliūnas
Vilnius Univer...
MII 2
Content
1. Introductory facts about
National Research Data Archive (MIDAS) project
2. Implementation aims and pr...
MII 3
1. Introductory facts about
MIDAS project (1)
• Project Title: National Open Access Research
Data Archive (LT: N...
MII 4
1. Introductory facts about
MIDAS project (2)
• Funded by: EU Structural Funds and
national budget
• Project bu...
MII 5
2. Implementation aims
and principles of MIDAS
MIDAS implementation purpose
• to establish the infrastructure th...
MII 6
Prospective MIDAS users
• Researchers, lecturers, professors, students;
• Science and studies institutions
[and/...
MII 7
Development principles
• privacy and security
(i.e., information confidentiality,
integrity and non-repudiation)...
MII 8
MIDAS compatibility
• MIDAS archive will be based on usage of open
code software, XML format and other open
meta...
MII 9
Integration with other
data archives and registers
• Lithuanian Academic E-Library eLABa www.elaba.lt
• Lithuani...
MII 10
3. Planned MIDAS
outcomes and peculiarities
MIDAS outcomes (1)
• The infrastructure
that enables collection, o...
MII 11
MIDAS outcomes (2)
• National united research data archive
with analytical software tools;
• Infrastructure for...
MII 12
MIDAS implementation advantages
• Guaranteed safety and
effective sharing of research data
• Increased quality ...
4. Data mining tool DAMIS
(slides by Olga Kurasova <......................................> )
Functionalities of DAMIS
• DAMIS is a tool for analysis of the MIDAS data;
• The following data mining methods are
impl...
Functionalities of DAMIS
• DAMIS is a web-based system http://dev.damis.lt
(user name/password: demo/demo , 1234/1234 );...
A sample of multidimensional data
(breast cancer data)
C
5 1 1 1 2 1 3 1 1 b
5 4 4 5 7 10 3 2 1 b
3 1 1 1 2 2 3 1 1 b...
DAMIS GUI
Data upload
Data preprocessing
Experiments
Statistical primitives
Dimensionality reduction
Data classification and clustering
Matrix view of Iris after
dimensionality reduction by PCA
Iris graphical representation
MII 26
5. Conclusions (1)
• MIDAS will provide virtual services for
researchers and other participants in research and ...
MII 27
5. Conclusions (2)
• National Research Data Archive MIDAS
will increase research cooperation possibilities,
bec...
MII 28
6. Demonstration of MIDAS
http://midas.insoft.lt:8888/web/
User name / password:
101/101
MII 29
7. Demonstration of DAMIS
http://dev.damis.lt
User name / password:
demo/demo
Thanks for Your Attention !
Questions ?...
of 30

National Research Data Archive MIDAS

Presentation at "Theory Days – 2014" in Ratnieki, Latvia, 2-5.10.2014 http://home.lu.lv/~df/tdays-ratnieki/
Published on: Mar 3, 2016
Published in: Technology      
Source: www.slideshare.net


Transcripts - National Research Data Archive MIDAS

  • 1. National Research Data Archive MIDAS: development decisions and usage peculiarities Saulius Maskeliūnas Vilnius University Institute of Mathematics and Informatics Akademijos str. 4, Vilnius LT-08663, Lithuania .
  • 2. MII 2 Content 1. Introductory facts about National Research Data Archive (MIDAS) project 2. Implementation aims and principles of MIDAS 3. Planned MIDAS outcomes and peculiarities 4. MIDAS data mining tool (DAMIS) 5. Conclusions 6. Demonstration of MIDAS 7. Demonstration of DAMIS
  • 3. MII 3 1. Introductory facts about MIDAS project (1) • Project Title: National Open Access Research Data Archive (LT: Nacionalinis atviros prieigos Mokslo Informacijos Duomenų Archyvas, MIDAS) • Lead institution: Vilnius University www.vu.lt • Project partner: Vilnius University Hospital Santariškių Klinikos (Santariškės Clinics) santa.lt • Project participants: 13 institutions of science and studies, and medical institutions
  • 4. MII 4 1. Introductory facts about MIDAS project (2) • Funded by: EU Structural Funds and national budget • Project budget: ~ € 4.34M (i.e., almost 15M LTL) • Duration: 40 months (start date: January 1, 2012 , end date: June 30, 2014  April 30, 2015) • Current status: – technical infrastructure: not installed yet; – software development: beginning of 2nd iteration.
  • 5. MII 5 2. Implementation aims and principles of MIDAS MIDAS implementation purpose • to establish the infrastructure that enables collection, organizing and storage of empirical and research data (with corresponding metadata), ensuring free, convenient, interactive search, access and analysis of data.
  • 6. MII 6 Prospective MIDAS users • Researchers, lecturers, professors, students; • Science and studies institutions [and/or their representatives]; • Institutions which present research data (e.g., hospitals), • Research and development (R&D) enterprises; • Public administration institutions which use R&D statistical data; • other interested physical and judicial persons.
  • 7. MII 7 Development principles • privacy and security (i.e., information confidentiality, integrity and non-repudiation) • usability • accessibility (functioning 24 hours per day, 7 days per week) • extensibility (i.e., software architecture scaling in cases of incorporation of additional hardware)
  • 8. MII 8 MIDAS compatibility • MIDAS archive will be based on usage of open code software, XML format and other open metadata, bibliographic, information retrieval standards (CERIF, CERIF for Datasets, CIF, DICOM, Dublin Core, MARC21, ISO/IEC 11179-1:2004, OAI-PMH, etc.). • That will ensure compatibility with other information systems, data archives and registries in Lithuania and internationally (e.g., Data Citation Index of Thomson Reuters http://thomsonreuters.com/data-citation-index/ ).
  • 9. MII 9 Integration with other data archives and registers • Lithuanian Academic E-Library eLABa www.elaba.lt • Lithuanian Data Archive for Social Sciences and Humanities LiDA www.lidata.eu/en • Lithuanian Networked Digital Library of Theses and Dissertations Lit-ETD etd.elaba.lt • National Medical Picture Archiving and Information Exchange System MedVAIS http://www.epractice.eu/en/news/5364871 • etc.
  • 10. MII 10 3. Planned MIDAS outcomes and peculiarities MIDAS outcomes (1) • The infrastructure that enables collection, organizing and storage of empirical and research data (with corresponding metadata), ensuring free, convenient, interactive search, access and analysis of data;
  • 11. MII 11 MIDAS outcomes (2) • National united research data archive with analytical software tools; • Infrastructure for collection and transferring of biomedical research data, consisting of DICOM (for collecting data from medical equipment), ECG (for collecting electrical cardiogram data from medical devices), content management, data depersonalisation, and data archiving modules; • Public interactive e-service “Search, Delivery and Analysis of Research Data”.
  • 12. MII 12 MIDAS implementation advantages • Guaranteed safety and effective sharing of research data • Increased quality of research outputs • Preventing duplication of effort in research data collection • Increased variety of research outputs
  • 13. 4. Data mining tool DAMIS (slides by Olga Kurasova <......................................> )
  • 14. Functionalities of DAMIS • DAMIS is a tool for analysis of the MIDAS data; • The following data mining methods are implemented: • preprocessing (cleaning, filtering, splitting, transposing, norming, feature selecting); • statistical primitives (min, max, mean, standard deviation, median); • dimensionality reduction (multidimensional data visualization); • classification and clustering.
  • 15. Functionalities of DAMIS • DAMIS is a web-based system http://dev.damis.lt (user name/password: demo/demo , 1234/1234 ); • The web interface does not require any software installation; a web browser is enough for its usage; • There is a possibility to choose high performance computing resources (VU MII cluster – VU MIF supercomputer); • The usage is based on creation of scientific workflows; • The results obtained can be saved in MIDAS and in a user computer.
  • 16. A sample of multidimensional data (breast cancer data) C 5 1 1 1 2 1 3 1 1 b 5 4 4 5 7 10 3 2 1 b 3 1 1 1 2 2 3 1 1 b 6 8 8 1 3 4 3 7 1 b 4 1 1 3 2 1 3 1 1 b 1 1 1 1 2 10 3 1 1 b 2 1 2 1 2 1 3 1 1 b 2 1 1 1 2 1 1 1 5 b 4 2 1 1 2 1 2 1 1 b ... ... ... ... ... ... ... ... ... ... ... 8 10 10 8 7 10 9 7 1 m 5 3 3 3 2 3 4 4 1 m 8 7 5 10 7 9 5 5 4 m 7 4 6 4 6 1 4 3 1 m 10 7 7 6 4 10 4 1 2 m 7 3 2 10 5 10 5 4 4 m 10 5 5 3 6 7 7 10 1 m ... ... ... ... ... ... ... ... ... ... ... 4 8 8 5 4 5 10 4 1 m
  • 17. DAMIS GUI
  • 18. Data upload
  • 19. Data preprocessing
  • 20. Experiments
  • 21. Statistical primitives
  • 22. Dimensionality reduction
  • 23. Data classification and clustering
  • 24. Matrix view of Iris after dimensionality reduction by PCA
  • 25. Iris graphical representation
  • 26. MII 26 5. Conclusions (1) • MIDAS will provide virtual services for researchers and other participants in research and education that can lead to more efficient, effective and higher quality research; • Users will have the possibilities to: – register, find and cite research data, – search for and use other infrastructures and tools (which provide data archiving services), – share or integrate data and tools to other science and studies infrastructures;
  • 27. MII 27 5. Conclusions (2) • National Research Data Archive MIDAS will increase research cooperation possibilities, because of simpler, more convenient, unified, advanced possibilities of research data collection, analysis, application and sharing.
  • 28. MII 28 6. Demonstration of MIDAS http://midas.insoft.lt:8888/web/ User name / password: 101/101
  • 29. MII 29 7. Demonstration of DAMIS http://dev.damis.lt User name / password: demo/demo
  • 30. Thanks for Your Attention ! Questions ?...

Related Documents