Collaborative ORDnance Data Repository (CORD)

by Dionysia Kontotasiou and Olivier Cottray [ Geneva International Centre for Humanitarian Demining ] - view pdf

Until recently, the mine action and broader humanitarian disarmament community have relied on ORDATA for ordnance information. However, the Geneva International Centre for Humanitarian Demining, James Madison University and software-development company Ripple Design developed a novel aid to using ORDATA: the Collaborative ORDnance data repository (CORD). Representing a significant update to the data set, CORD is an ordnance-identification system featuring Web 2.0 concepts that allow individual users to contribute updates to the data.

Figure 1. The CORD website.All figures courtesy of GICHD.
Figure 1. The CORD website.
All figures courtesy of GICHD.

The humanitarian demining and conventional weapons destruction communities previously relied on a single open source for ordnance data: ORDATA, which had not been updated for many years. In general, maintaining this system was costly; subject-matter experts were not hired on a regular basis to add new and update existing content to the database. ORDATA was significantly out of date, and important applications that relied on it as a reference source for ordnance data, like the Information Management System for Mine Action (IMSMA), did not produce accurate statistics.1 Countries remedied this issue by building and integrating their own personalized copies of ORDATA into their IMSMA installations. However, this created heterogeneity between databases that prevented data set comparisons.

In response, the Geneva International Centre for Humanitarian Demining (GICHD), James Madison University (JMU) and software-development company Ripple Design developed a novel approach for ordnance data management: the Collaborative ORDnance data repository (CORD).2,3,4

CORD is a comprehensive approach to solving the aforementioned issues in mine action as well as the broader areas of humanitarian disarmament and human rights domains. CORD is a collaborative ordnance data repository for the extended humanitarian disarmament community built on Web 2.0 concepts, where data users contribute by keeping the data up to date. This approach could bring exciting new possibilities to the humanitarian disarmament field while significantly cutting maintenance costs. CORD further innovates through the use of semantic technologies that move away from relational databases and facilitate querying in human language as well as offer some built-in analytical capacity to answer questions, such as “in which countries is ordnance X used?”

Figure 2. Detailed search result from CORD.
Figure 2. Detailed search result from CORD.

CORD’s main role is to act as an interface for heterogeneous resources, which can be found in several databases. These ordnance data resources are structured in a semantic taxonomy or ontology, which allows for resource querying in order to provide a global collaborative ordnance data repository to any user requesting dynamic content.

Typical CORD uses

Use Case 1: IMSMA. The IMSMA application references ORDATA for ordnance data. However, due to ORDATA’s high maintenance costs, ORDATA is not up to date, nor is ordnance data inside IMSMA. By building their own personalized copies of the ORDATA databases and integrating them into their IMSMA installations, countries have avoided issues of obsolete data. Alternatively, this creates heterogeneity that prevents countries from comparing statistics. By using CORD, IMSMA and other applications relying on ORDATA, mine action users can link data across database boundaries and seamlessly enable more intuitive queries, searches and navigation. CORD will also facilitate more advanced data analysis and integrative knowledge discovery based on the huge web of data, thus improving the quality of statistics.

Use Case 2: Mine action users. The professional mine action community often needs information on specific items, namely technical information concerning various explosive remnants of war (ERW). Currently, that information is decentralized and not easily accessible in a comprehensive way. By using CORD, mine action users can find necessary information in an intuitive format.

Use Case 3: Academic users. Academic institutions can also access an online repository of gathered information for ordnance data to support various research initiatives studying the impact of landmines and ERW contamination. CORD can be seen as an important data source for attempting to answer questions, such as “What are the specific impacts of specific types of munitions?” While the impact data would originate from other sources, ordnance characteristics could be extracted from CORD.

Desk Study

Several ongoing projects outside of CORD approached the same topic from different angles. A conceptual desk study was performed prior to CORD implementation with the following objectives:

Methodology. In order to get an overview of existing ordnance systems, the following methodologies were used:

Outcome. The desk study determined no common agreement in the basic terminology for ordnance classification. This is reflected in the following points:

Finding a common agreement within the community on a standardized ordnance categorization was challenging. Even if such an agreement could be reached, standards would be impossible to implement with the prominence of relational databases (the common technology used in existing systems today).

Analysis and Design

System architecture. The architecture of CORD can be seen in Figure 3.

The main architectural diagram of Figure 1 depicts the interoperations occurring between the basic modules, external interacting entities and end users of CORD. The architecture includes the following components:

Figure 3. CORD architecture.
Figure 3. CORD architecture.

Ordnance data representation (back end). Since the core functionality of the framework relies on an ordnance-characteristics combination, a major requirement was the representation and concurrent communication with numerous heterogeneous data platforms and databases. Two main data models are used for representing knowledge and information in computer systems. Database models, especially relational databases, have been the leader in the last few decades, enabling efficient storage and querying of information. However, ontologies have appeared as an alternative to databases in applications that require a more enriched meaning.

A database model is a representation mechanism designed mainly to meet the requirements of a particular application or corporation. When these requirements change, the viewpoint and the schema also need modification, which is one of the main reasons that another solution for representing ordnance data was required for ORDATA.

In contrast, ontologies are the result of a collective effort and should therefore be shared among the community. Ontologies provide a restriction-free framework to represent a machine-readable reality on the Web. This framework assumes an open world in which information can be explicitly defined, shared, reused or distributed. Moreover, information can be interchanged and used to make deductions or queries. Thus, databases result from teamwork, and ontologies require coordination among several work groups.

For CORD to be feasible, a common, modular, shareable ontological framework is necessary to deal with representation issues and provide the required infrastructure for decision support. This infrastructure was provided in the form of an ontology that interconnects ordnance resources available in several databases and Extensible Markup Language (XML) structures.

Figure 4. Ontology architecture.
Figure 4. Ontology architecture.

CORD moves forward from databases to ontologies for representing ordnance data, because ontology:

CORD ontology includes, but is not limited to, the following categories of data:

Collaborative repository (front end). Since an ontology was selected to represent ordnance data, an open-source interface was developed to make these data available.

The main functionalities of CORD front end are
Collaborative authoring: operations that allow easy and expressive additions of data

Sequence diagram. A sequence diagram of CORD is presented in Figure 5. The basic purpose of using this kind of interaction diagram is to show how processes operate with one another and in what order. The important processes are inside the sequence diagram, which describes completely the main interactive functionality of the system.

Starting from the top of the diagram, an administrator requests authentication (message 1) from the server, which receives the signal and responds with a success signal (message 2) if the authentication process was successful. The server continuously awaits for user signals (messages 3, 5, 7) in order to authenticate them (registered user 1-N). Registered user 1 decides to create a new ordnance category and requests creation from the server (message 9). Immediately, the administrator accepts the ordnance category (message 10). The new ordnance category is viewable from all lifelines, so even a simple user can browse it. Finally, a simple user makes a SPARQL query to the system (message 15), and the system responds with the appropriate results (message 16).

Figure 5. Sequence diagram.
Figure 5. Sequence diagram.

Expected Benefits

Generic benefits. CORD will offer a more cost-effective, accessible, reusable and completely expansionistic system.

Technological benefits. CORD’s main innovation is the homogenization of ordnance resources under a common ontology in order to deliver reliable information to the end user in various and combined ways. The collaborative data repository is critical, since individuals may have a common resource access point for all ordnance data. Major benefits include centralization of resources from different stand-alone databases and fusion of ordnance data through the collaborative data repository, which will be easily accessed by all.

Analytics

3,343 users from 124 countries have accessed CORD (September 2014–September 2015).14

Conclusions

The main innovations of CORD are the use of semantic technologies and collaborative Web 2.0 concepts to structure and represent ordnance data. Semantic technologies were used to represent and provide dynamic searches of multiple stand-alone ordnance data. Semantic technologies and ontologies promise a more flexible representation than XML-based technologies and flat databases. Data do not need to conform to a tree structure but can follow an arbitrary graph shape. Given the benefits of these technologies and the trends of developments in the Web space, semantic technologies were adopted to represent ordnance data in a structured way, interfacing them via a collaborative website.

Some important constraints with the approach of making an open system is that classified or otherwise sensitive information will not be included in the central system, such as render-safe procedures on how to disarm ordnance. Should an organization using CORD want to have additional information, it can set up private databases extending the data yet still linking to CORD. This architecture also allows organizations to have additional data-verification processes.

For more information, the tool can be found at http://bit.ly/1Lbcpvp. c

 

Biography

Dionysia KontotasiouDionysia Kontotasiou joined Geneva International Centre for Humanitarian Demining (GICHD) in March 2013 as a MediaWiki specialist. She is the project manager of CORD. In addition, Kontotasiou is the project manager of the Information Management System for Mine Action (IMSMA), and administrator of the in-house MediaWiki website that serves as IMSMA’s documentation and support site. Prior to joining GICHD, she was a research assistant in the Informatics and Telematics Institute in Thessaloniki, Greece. Kontotasiou earned a diploma in electrical and computer engineering from Aristotle University of Thessaloniki (Greece).

Olivier Cottray Olivier Cottray joined Geneva International Centre for Humanitarian Demining (GICHD) in January 2012, initially as the information-services coordinator. As head of the Information Management Division, he is in charge of managing the team that provides information-management capacity development and technical support to the mine action community. Prior to joining GICHD, Cottray ran geographic information-systems (GIS) support cells in the U.N. and nongovernmental field operations. Cottray helped found the NGO, Map Action, and has run emergency GIS cells in a number of countries. Cottray earned a Bachelor of Science in geography and economics at the London School of Economics (U.K.), and a master’s degree in GIS and remote sensing at the University of Cambridge (U.K.).


Contact Information

Dionysia Kontotasiou
Collaborative ORDnance Data Repository Project Manager
Geneva International Centre for Humanitarian Demining (GICHD)
Email: d.kontotasiou@gichd.org
Website: http://www.gichd.org

Olivier Cottray
Head of Information Management
Geneva International Centre for Humanitarian Demining (GICHD)
Tel: +41 22 730 93 43
Email: o.cottray@gichd.org
Website: http://www.gichd.org

 

Endnotes

  1. “Welcome to IMSMA Wiki.” GICHD IMSMA. Accessed 6 October 2015 http://bit.ly/1KwUr4o.
  2. “Mine Action.” GICHD. Accessed 6 October 2015 http://bit.ly/1P8xsMO.
  3. James Madison University. Accessed 6 October 2015 http://bit.ly/1bSkFgp.
  4. Ripple Design. Accessed 6 October 2015 http://bit.ly/1LmQuQl.
  5. Fenix Insight Ltd. http://bit.ly/1Mhg87D. Accessed 6 October 2015.
  6. Golden West Humanitarian Foundation. Accessed 6 October 2015 http://bit.ly/1NBkY1o.
  7. Spinator. http://bit.ly/1FlgquD. Accessed Accessed 6 October 2015.
  8. NATO Support and Procurement Agency. Accessed 6 October 2015 http://bit.ly/1P8QnHg.
  9. Weapons Law Encyclopedia. Geneva Academy of International Humanitarian Law and Human Rights. Accessed 6 October 2015 http://bit.ly/1iVlKuB.
  10. Federations of American Scientists. http://fas.org/. 22 October 2015.
  11. Collective Awareness to Unexploded Ordnance. http://www.cat-uxo.com. 22 October 2015.
  12. Protégé. http://stanford.io/1KW4tOz. 6 October 2015.
  13. SPARQL Query Language for RDF. http://bit.ly/1C1UoJ5. Accessed 6 October 2015.
  14. Numbers were extracted from Google Analytics.

TOP OF PAGE