METICOS Data Model Development
METICOS project is developing a platform that integrates information systems and networks of data sources in order to validate the social and societal acceptance of Automated Border Control solutions. In order to achieve this, it leverages heterogenous data, from various locations, types and formats. Moreover, one of the aims of METICOS project is to introduce Big Data Analytics (WP7) in order to predict the acceptance and perception of Smart Border Control Technologies.
In order to capture the data needed to assess (in both an objective and subjective way) the Smart Border Control Technologies (SBCT) in a harmonized manner, it is necessary to develop a flexible data model. All services and modules that are planned to operate within the METICOS system will use the proposed structures defined in the METICOS Data Model in order to exchange data. The responsible partner for developing the METICOS Data Model is ICCS.
In order to achieve the above objectives, we have followed a hybrid approach, both data-driven and knowledge-based. This approach is based upon the literature, for instance  , ,  , ,  and , and adapted to METICOS requirements and needs. Regarding the data-driven approach, after creating a corpus of relevant literature, we employ Natural Language Processing techniques in order to capture terms and relationships between them which could be used in the construction of the METICOS data model. These techniques include Keyword Extraction and Topic Modelling, and some of the algorithms and tools employed include Latent Semantic Analysis (LSA), Latent Dirichlet Allocation (LDA), keyBERT – a keyword extraction technique that leverages BERT embeddings to create keywords and keyphrases. Then, experts’ knowledge is harnessed to confirm the obtained results and build the data model entities, their attributes, metadata and the relationships amongst them, whilst enriching the terms with additional ones whenever needed. An example of this domain expert knowledge which was taken into account when developing this data model is the METICOS Technology Acceptance Model, which aims to model the factors having an impact on the acceptance of SBCT. The output of this hybrid approach is the METICOS data model, which leverages the domain expert knowledge as well as data-driven knowledge extraction and combines the complementary advantages of both approaches.
Finally, the main entities which have been identified by this hybrid approach include the type of technology, the subjective feedback and objective performance indicators as the influencing factors, as well as the various types of users of the technologies. Moreover, METICOS data model will allow the assessment of the acceptance of SBCT in other settings, configurations and Border Control Points (BCPs) under the condition that the same data entities and their attributes are captured.
|||M. N. Asim, M. Wasim, M. U. G. Khan, W. Mahmood and H. MahnoorAbbasi, “A survey of ontology learning techniques and applications,” Database: The journal of Biological Databases and Curation, vol. 2018, 2018.|
|||V. Presutti, E. Daga, A. Gangemi and E. Blomqvist, “eXtreme Design with Content Ontology Design Patterns.,” in Proceedings of WOP 2009, 2009.|
|||A. De Nicola and M. A. Missikoff, “Lightweight Methodology for Rapid Ontology Engineering,” Communications of the ACM, vol. 59, no. 3, 2016.|
|||K. .M. and L. A., ” Test-Driven Development of Ontologies,” in Proceedings of ESWC 2016: 642-657., 2016.|
|||M. Suárez-Figueroa, A. Gómez-Pérez and “. M. Fernández-López, “The NeOn Methodology for Ontology Engineering,” Ontology Engineering in a Networked World, pp. 9-34, 2012.|
|||S. John, N. Shah and L. Smalov, “Incremental and Iterative Agile Methodology (IIAM): Hybrid Approach for Ontology Design towards Semantic web Based Educational Systems Development,” ”, International Journal of Knowledge Engineering, vol. 2, 2016.|