The Triquetra Knowledge Base Platform

THE TRIQUETRA KNOWLEDGE BASE PLATFORM Anastasia Anastasiou1, Efthymios Magkoufis1, Christos Kontopoulos1, Vasiliki (Betty) Charalampopoulou1 1. Geosystems Hellas; [email protected] (A.A.); [email protected] (E.M.); [email protected] (C.K.); [email protected] (V.C.) ABSTRACT The TRIQUETRA Knowledge Base Platform (KBP), which is part of the TRIQUETRA Decision Support System (DSS) Platform comprises two distinct components, a database containing the outputs of the literature review conducted in the context of the project and a WebGIS platform containing all relevant data to the pilot Cultural Heritage (CH) sites of the project, namely Aegina, Choirokoitia, Epidaurus, Kalapodi, Les Argilliez, Roseninsel, Smuszewo and Ventotene. Essentially, the KBP acts as an electronic repository covering a broad spectrum of data, featuring advanced search tools, enabling users to efficiently search and discover relevant information within the stored datasets. The main objective of this manuscript is to outline the capabilities and main functionalities, technological stacks and interdependencies (e.g., architecture, data type) of the KBP, as well as its potential exploitation and expansion beyond the context of the project. Index Terms— cultural heritage, database, geospatial, map, platform 1. INTRODUCTION Nowadays, in this rapidly evolving business landscape, efficient data management as well as the accessibility to organizational knowledge are an integral part of fostering innovation and streamlining operations. To this end, the implementation of robust knowledge base platforms, hosting a variety of features with advanced search functionalities, has emerged as a fundamental strategy for data management. Consequently, KBPs serve as valuable tools in a series of industries and fields, among which are cultural heritage, health, agriculture, transportation etc. For instance, PopHR is a KBP for the integration, analysis and visualisation of population health data [1], whereas PDSIDES is a KBP for the design of engineering systems [2]. Moreover, a series of up-and-coming commercial solutions of KBPs aims to assist all relevant stakeholders in generating and managing their data using these knowledge repositories. At the same time, the evolution of WebGIS has followed a consistent trajectory, bringing transformative impacts, including interactive accessibility to geospatial data, realtime integration of data, as well as the availability of GIS analysis tools [3][4]. In recent years, a series of studies have explored the applications of WebGIS across various domains, with notable examples, a WebGIS tailored for the visualisation of the outcomes in assessing seismic risk for structures within Italy’s infrastructures [5], as well as a WebGIS developed for the holistic monitoring of critical infrastructures combining earth observation and in-situ data for the area of west Peloponnese, Greece [6]. The TRIQUETRA project is an EU funded project (GA number 101094818) aiming at creating an evidence-based assessment platform that allows precise risk stratification via the creation of a database of available mitigation measures and strategies, acting as a Decision Support Tool towards efficient risk mitigation and site remediation. The overall approach of the TRIQUETRA project is based on three distinct steps: (i) Risk Identification, (ii) Risk Quantification and (iii) Risk Mitigation. In this context, a novel KPB has been developed, constituting an electronic repository with advanced search tools and capabilities. It includes information on Climate Change (CC), geological, historical, and site-specific data, risks & mitigation measures for CH sites etc., based on the verified data, geographical identification and results obtained from the outputs of the project. The main objective of the KBP is to comprehensively integrate and visualize all the data shared within the project, leveraging both a searchable database of literature and a sophisticated WebGIS platform, adhering to the Open Geospatial Consortium (OGC) [7] and INSPIRE [8] standards. The combination of the available datasets for each pilot site leads to the creation of a data cube, a multidimensional data structure used for the efficient representation and analysis of data in multiple dimensions. These dimensions include, among others, different variables and attributes of the data such as time, location etc. This manuscript aims to give an overview of the KBP, detailing its function and current contents. It is important to note that the platform will continue to evolve and adapt to the upcoming outputs of the project. Additionally, it aims to give insight into the prospect of further exploitation of the platform, as a rich database for fusing different types of data and enabling efficient research of each CH site, fostering comprehensive knowledge in the field of CH monitoring and facilitating the optimal preservation and risk mitigation actions. 2. GENERAL DESCRIPTION OF THE PLATFORM The TRIQUETRA KBP is a digital repository equipped with a set of advanced search tools. The platform incorporates various features for the deployment, cataloguing and categorisation of the data produced and shared within the project, to enhance the discoverability of the data. The KBP is structured based on two distinct key components, namely the Bibliography database and the WebGIS (Figure 1). These distinct sections of the platform gather all the data and information related to the pilot CH sites of the project. Figure 1. The homepage of the TRIQUETRA KBP. More specifically, the Bibliography database contains various types of scientific reports, papers and literature thoroughly reviewed as part of the project. Consequently, a classification system was established based on a series of criteria regarding key information deriving from each report such as title, country, type of risk, monitoring method, mitigation measure, period etc. In parallel, the WebGIS component of the KBP is hosting all the geospatial data related to the pilot CH sites, constituting the outcomes of other complementary tasks undertaken within the project. Additionally, the WebGIS seamlessly incorporates EU web services, such as various Copernicus Services [9] and on-site observation open services which are leveraged throughout the project. Aiming to provide integrable and interoperable open data services, the open OGC web services are utilised for data exchange and integration adhere to unified standards, including the Web Map Service (WMS) for serving collections of layers as map images and the Web Feature Service (WFS) for serving data as vector features. Furthermore, the Web Server operates as a gateway server receiving client requests, such as a specific inquiry to access geospatial data stored within the database, and then redirects these requests to an open source server for sharing geospatial data, which in turn handles it appropriately, generating the corresponding response. Finally, the request is transmitted back to the Web Server, which redirects the response to the client. This streamlined communication between the components ensures the efficient performance of the overall framework, enabling the simultaneous management of incoming client requests. Each pilot site undergoes a similar classification process as the one followed in the Bibliography component for the available related data, based on their type and characteristics. Figure 2. Structure of layer folders and data types of the WebGIS component. Historical/archaeological data, topographic data (including aerial photographs, orthophotos, DEMs and point clouds), geophysical data, climatology data, mitigation measures, risk models etc., are only a few examples of the categories regarding each pilot site (Figure 2). 3. DESIGN AND IMPLEMENTATION OF THE TRIQUETRA KNOWLEDGE BASE PLATFORM The TRIQUETRA KBP employs a dual-front-end architectural approach, dividing the User Interface (UI) layer into two distinct components, the Bibliography and the WebGIS, as already mentioned. This decision stems from the necessity to cater to different user needs and preferences in data interpretation and visualisation, while also taking into account that each component handles different kinds of datasets referring to different kinds of activities. 3.1. Bibliography component Regarding the Bibliography component, the data is presented in a tabular format resembling traditional databases, while interaction with the data is enabled through table management and filtering options, as well as a map tool. Moreover, the map tool serves as a distinct, simpler and clearer map search option for the user, updating the map and tabular information based on the current map extend. A data table is an effective format for both data analysis and visualisation. When dealing with large datasets, it becomes crucial to employ a tool that not only visualises the data but also conveys a sense of scale of the different kinds of data. The bibliography page presents a tabular view of the data and incorporates an interactive map enhancing their visualisation. The table provides in-depth descriptions of each paper and its variations, while also grouping them based on their geographical significance (Figure 3). Figure 3. The Bibliography component. 3.2. WebGIS component A WebGIS module provides users with the essential tools for generating insights and conducting predictive modeling to aid decision-making policies, including risk assessment and mitigation. The TRIQUETRA WebGIS component utilises MapBox GL to represent geospatial data and layers collected and created within the project in a 3D environment with advanced visualization capabilities and significantly less, compared to other alternatives, network bandwidth [10]. The platform transforms data into nested components simplifying the navigation of layers and point clouds for users, while also maintaining a uniform appearance without compromising the integrity of the data. The WebGIS component includes the same navigation tools as the Bibliography component, such as move, zoom, tilt, and search, but it also offers a variety of additional tools beyond what is available in the Bibliography component. Users can activate multiple layers simultaneously, for a more detailed view of the selected area, mass control all layers, select different map styles, view 3D versions of the CH sites via the available point clouds etc. leveraging the state-of-theart capabilities of Deck gl library for 3D tiles rendering [11] (Figure 4). Figure 4. The WebGIS component. Additionally, in the WebGIS component, the data is presented in a geographical context, georeferenced to their accurate position, allowing the user to view overlapping datasets, while also using mild data management tools such as opacity sliders for better understanding, comparing and combining the different datasets. To establish a data exchange framework aligned with European directives such as INSPIRE, a Geoserver v.2.23.1 Docker image has been successfully installed. Concurrently, to facilitate the storage of vector data, a PostgreSQL relational database has been deployed. At the same time, the activation of the PostGIS extension enhances its capabilities, enabling the storage, indexing, and querying of spatial data within specific geographic coordinate systems (EPSG), as well as the execution of specialised spatial searches. The connection between the data storage environment and the GeoServer is established through a custom-built processing engine written in Python3, which leverages wellestablished Python libraries such as GeoPandas, GeoServerrest, SQLAlchemy, etc. to perform a recurring scheduled workflow, ensuring seamless integration. 3.3. TRIQUETRA Knowledge Base Platform architectural diagram The optimal performance of the Knowledge Base Platform is strongly dependent both on the user’s interface and the established interconnection with the offered data of the TRIQUETRA project. Therefore, it is imperative to develop each component in a manner that ensures the successful and continuous operation of the whole system [6]. The following diagram (Figure 5) thoroughly explains the complete architecture of the KBP and its contents. Figure 5. TRIQUETRA KBP Architecture. 3.4. Front end To ensure the optimal performance and scalability of the KBP, the development of the front-end leverages the enhanced capabilities of the React JS framework [12]. Regarding the front-end design, the objective is to closely adhere to the design vision while ensuring cohesion with the parent website of the platform. This requires attention to details such as colours, fonts, structure, and the overall aesthetic of the platform since harmonising these elements helps maintain a consistent visual identity across the project. Simultaneously, the design process integrates fundamental principles of front-end design [13], encompassing responsiveness, performance optimisation, user-centred design, and progressive enhancement. This translates into the ability of the platform to adapt to various screen sizes and resolutions to ensure the platform appears the same by eliminating pixel count and opting for a more dynamic view. 3.5. Back end The back-end functions as a crucial link, connecting the Geoserver with the SQL database generated from the received data for both the Bibliography and WebGIS components. To this end, it is essential to prioritise straightforward navigation of the server and the relationships between data cells to avoid data conflicts in case of more complex requests. For diverse categories of data, such as time periods and pilot areas, separate tables were created, each linked to a central paper contents table through a unique ID. This design choice enables efficient management of unique values accessible to all scientific papers, providing a comprehensive view of any research paper. Additionally, it simplifies data manipulation on the front-end of the webpage by granting access to unique, non-repeating values. Notably, the data subsections (e.g., pilot area, period, site context) operate independently, lacking communication and shared foreign keys. This deliberate isolation ensures that data have a limited number of pathways, reinforcing the database structure. This approach adds a layer of assurance to the database, contributing to its resilience and reliability. 4. FUTURE INSIGHTS The maintenance and updating of the KBP continue throughout the TRIQUETRA project, incorporating new datasets from ongoing and upcoming tasks. Given that the project’s duration exceeds one additional year, additional data shall be further acquired and new architectural components will potentially be developed within the WebGIS of the TRIQUETRA KBP. Considering the differentiation in sharing rights of the project’s various datasets, with some intended solely for distribution among project partners, initial access to the KBP via the TRIQUETRA project website will be restricted and require a secured user login. The goal is to categorise data based on their sharing rights using Dataset Identification Templates (DIT) provided by partners. Subsequently, different access levels to the KBP will be established, providing full access to consortium members and limited access to external users. To ensure the optimal operation and accessibility of the end-users, further efforts shall be made to enhance accessibility for diverse end-users, including those with hearing or visual impairments. The KBP is designed to accommodate feedback from end-users, addressing queries and incorporating valuable comments to continually improve the platform. The encompassing of a great variety of geospatial and non-geospatial data, in combination with the implementation of advanced architecture that enables the functionalities for multiparametric search and the augmented visualisation of the data onto a 3D virtual earth environment, positions the introduced technical solution as a valuable asset for the CH community and the future monitoring of the CH sites. The KBP platform has been developed within the context of the TRIQUETRA project, however the potential commercial exploitation of the KBP either as a whole or as two distinct components can lead to the improvement in the long-term preservation of CH sites in the manner of efficiency and cost savings, while also serving in the facilitation of decision-making and management of CH sites. Using the variety of tools provided by the KBP beyond the context of the TRIQUETRA project on a larger scale, researchers will be able to deploy and combine all available data to reach useful conclusions and optimal solutions for the protection of CH. 5. CONCLUSIONS The monitoring of Cultural Heritage sites and the efficiency level of the applied solutions for their maintenance and longterm preservation constitute a multidimensional issue. The Knowledge Base Platform, being developed under the research project TRIQUETRA, aims to provide the most upto-date knowledge around the current situation of CH sites as well as the risks they face and all available mitigation measures, via a user-friendly web interface that leverages state-of-the-art technologies (e.g., ReactJS, Mapbox gl js, Deck gl, 3D-tiles, etc.) for enhancing the platform’s performance and the overall user experience. The nature of the provided data, as well as the complexity of the objective, led to the development of a platform consisting of two individual components; the Bibliography and the WebGIS, each encompassing all necessary tools and technologies for any future development requirements. In conclusion, the KBP serves as a useful electronic repository where all available information and data regarding a series of CH sites are stored. End-users can effectively utilise for assessing their current state and the associated risks, while also providing useful insights on how to tackle these risks, decisively aiding their effective protection. 6. REFERENCES [1] A. Shaban-Nejad, M. Lavigne, A. Okhmatovskaia, D. L. Buckeridge, "PopHR: a knowledge-based platform to support integration analysis and visualization of population health data", Annals of the New York Academy of Sciences, 1387(1), 44–53. https://doi.org/10.1111/nyas.13271, 2016. [2] Z. Ming, A.B. Nellippallil, Y. Yan, G. Wang, C.H. Goh, J.K. Allen and F. Mistree, PDSIDES—A Knowledge-Based platform for decision support in the design of engineering systems. Journal of Computing and Information Science in Engineering, 18(4). https://doi.org/10.1115/1.4040461, 2018. [3] Y. Su, J. Slottow, A. Mozes, "Distributing proprietary geographic data on the World Wide Web—UCLA GIS database and map server", Comput Geosci 26(7):741–749. https://doi.org/10.1016/S0098-3004(99)00130-2, 2000. [4] H.C. Karnatak, R. Shukla, V.K. Sharma, Y.V.S. Murthy, V. Bhanumurthy, "Spatial mashup technology and real-time data integration in geo-web application using open source GIS—a case study for disaster management", Geocarto Int 27(6):499–514. https://doi.org/10.1080/10106049.2011.650651, 2012. [5] M. Mazzei, D. Quaroni, "Development of a 3D WebGIS Application for the Visualization of Seismic Risk on Infrastructural Work", ISPRS Int. J. Geo-Inf., 11, 22. https://doi.org/10.3390/ijgi11010022, 2022. [6] E. Magkoufis, S. Chlorokostas, L. Valdesera, A. Kyriou, and K. Nikolakopoulos "A web GIS platform for critical infrastructure monitoring within PROION project", Proc. SPIE 12734, Earth Resources and Environmental Remote Sensing/GIS Applications XIV, 1273415, https://doi.org/10.1117/12.2679457, 19 October 2023. [7] Open Geospatial Consortium (OGC): https://www.ogc.org/standards/ [8] INSPIRE Knowledge base, https://knowledgebase.inspire.ec.europa.eu/index_en, July 2023. [9] Copernicus services .xml file, https://viewer.globalland.vgt.vito.be/geoserver/ows?service=WMS &request=GetCapabilities. [10] A. Zunino, G. Velázquez, J. Celemín, C. Mateos, M. Hirsch and J. Rodriguez, "Evaluating the performance of three popular Web mapping libraries: A case study using Argentina’s life quality index", ISPRS Int. J. Geoinf. 9(10), 563, 2020. [11] K. Woo, A. Onsen, & W. Kim, “Implementation of a 3D WebGIS for dynamic Geo-referencing of 3D tiles on the virtual globe”, Journal of Geographic Information System, 15(05), 440– 457. https://doi.org/10.4236/jgis.2023.155022, 2023. [12] A. Javeed, "Performance Optimization Techniques for ReactJS", 2019 IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT), IEEE, 2019. [13] P. Kashfi, R. Feldt and A. Nilsson, "Integrating UX principles and practices into software development organizations: A case study of influencing events", J. Syst. Softw. 154, 37–58, 2019.