ESplora Research Project in Emilia-Romagna

ECOSISTER Programme: code ECS00000033

University of Ferrara

Cascade Call for Businesses SPOKE 5 – Theme 4: Development of innovative and sustainable solutions and services for the protection of cultural, maritime and coastal heritage. Regeneration and development of tourism systems, creation of new models, tools and IT services to improve the quality of the tourism offering.

Decree of the Director General, ref. no. 1517/2024, protocol no. 231800/2024 dated 31 July 2024

CUP: F78H22000410006

ESplora l’Emilia Romagna.

The system for the production, distribution and access to digital content relating to the region’s cultural, artistic, landscape, maritime and coastal heritage

The ESplora project was created with the aim of providing visitors of all nationalities with an innovative and easy-to-use tool, capable of ensuring immediate access to a vast database of cultural content. Through a multi-device app, ESplora provides digital information on historical, artistic and cultural heritage, enhanced with augmented reality features and supported by sophisticated web crawling and data mining techniques. These technologies make it possible to retrieve, aggregate and enhance data available on the web and from open data sources, transforming them into content that is simple and intuitive to access.

The platform is designed to meet the needs of public bodies, industry operators, citizens and tourists, providing targeted tools to inform, guide and enhance the visitor experience in the local area.

Thanks to its flexible and easily integrated architecture, ESplora can connect with existing initiatives, fostering collaboration, dissemination and active participation at local level, thereby helping to enhance the promotion and understanding of cultural heritage.

The project
The ESplora project had an overall duration of 12 months, starting on 1 August 2024 and concluding on 31 July 2025. During this period, research and development activities were guided by clear and concrete objectives aimed at creating an innovative and flexible platform capable of enhancing cultural, anthropological and natural heritage through state-of-the-art digital tools.
The main objectives of the project can be summarised as follows:

1. Database. Create a streamlined, efficient and optimised data structure capable of storing content that is easily accessible to tourists and visitors. The database was designed to ensure fast consultation and simple management, avoiding unnecessary redundancies and complexity, in order to focus on the quality of the user experience.
2. Content. Define an innovative methodology for populating the database with a wide range of content distributed extensively throughout the territory, while minimising creation and maintenance times and costs. The aim was to create a rich and constantly updated archive capable of enhancing places, heritage assets and cultural routes.
3. Crawling. Develop and implement Crawling and Scraping algorithms, essential for the automated acquisition of information content already available on websites and public databases. This approach makes it possible to systematically retrieve digital resources and transform them into structured information ready to be integrated into the system.
4. Data Mining. Evolve and enhance the Data Mining engine already developed in previous projects, with the aim of continuously enriching the database and supporting new ways of aggregating Points of Interest (POIs). In particular, the system was designed to support both traditional groupings based on the geographical location of heritage assets and thematic routes independent of their physical location, thereby offering new perspectives for cultural exploration.
5. Editorial Portal. Develop a complete editorial portal designed for the management and maintenance of the ESplora system’s information content. This tool integrates GIS features, assisted translation and structured editorial workflow management processes: from content drafting, to version history management, through to content validation and publication.
6. App. Develop a mobile application, available for download from the main digital stores, capable of directly accessing the information stored in the centralised repository. The app was designed to ensure ease of use and provide an immersive and comprehensive visitor experience, integrating maps, multimedia content and innovative interaction tools.

Objectives achieved
1 – Database. The database was implemented using the Elasticsearch NoSQL DBMS, chosen for its ability to store, index and retrieve large volumes of information quickly and efficiently. The production environment was configured as a cluster, currently consisting of three nodes, a solution that ensures high performance both in searches and data management, as well as improved system reliability and scalability.
Particular attention was paid to simplicity and a lightweight structure when designing the database. A streamlined data model was therefore adopted, free from unnecessary layers and focused on storing only genuinely essential data. This approach made it possible to optimise query times, reduce complexity and focus on the specific objectives of the project.
ESplora’s record structures do not follow the ICCD cataloguing standards typically used by museum or archival platforms, but were specifically designed according to a different criterion: prioritising performance while ensuring effective management of multilingual content, a fundamental requirement for the system’s international use.
Geographic data management was also integrated directly into Elasticsearch’s native modules for spatial data processing. Thanks to this solution, Points of Interest can be georeferenced efficiently using native support, making both coordinate-based searches and the cartographic representation of information more immediate.
Client applications access the data through an HTTP/REST API layer, which provides the information in JSON format. This approach ensures simple and standardised integration with the various components of the platform and with any external applications. The effectiveness of the design choices can be clearly seen in the day-to-day use of ESplora services: queries return responses within more than satisfactory times, demonstrating that the architecture has been correctly sized to meet the intended objectives. Furthermore, the system is scalable: should it become necessary in the future to increase processing or storage capacity, the infrastructure can be easily expanded by adding further nodes to both the Application Server cluster and the Data Server cluster, without requiring invasive interventions or major changes.

2 – Content. The ESplora system was designed to ensure maximum flexibility in information management and provides two different methods for entering content: manual and automated.
The manual method corresponds to traditional data entry carried out by specialised operators. Although more demanding in terms of time and resources, this method is particularly useful in specific contexts, for example when launching content enrichment projects developed in partnership with cultural institutions, universities or tourism operators. This method makes it possible to ensure a high level of attention to quality, accuracy of information and content customisation, all of which are essential in projects requiring a high degree of scientific or documentary reliability.
The automated method, on the other hand, represents the true innovation introduced by the research project and was one of its central objectives from the outset.
During the project, advanced algorithms and procedures were implemented to automatically acquire large amounts of data and information from the web, process and structure them, and finally store them in the ESplora database. These processes make it possible to systematically retrieve content relating to historical, artistic and cultural heritage (both images and descriptive texts), associate it with heritage assets already catalogued within the system and make it immediately available to end users.
Thanks to the integration of the manual and automated methods, the project benefited both from the accuracy and quality ensured by human work and from the speed and scalability made possible by automated procedures. By the end of the research project, this hybrid approach had made it possible to store and make available over 2,500 Points of Interest, extensively distributed throughout the Emilia-Romagna region, thereby creating a broad and structured digital heritage resource available to both residents and visitors.

3 – Crawling. The automated acquisition process represents one of the key elements within the ESplora project’s automated content acquisition system. Its implementation was designed to ensure a structured and scalable workflow capable of transforming the large amount of information available on the web into organised data that can be used within the platform. The procedure is divided into several stages. Initially, general-purpose search engines (Search Engines) are used to identify the so-called web targets: pages containing one or more pieces of content potentially associated with a specific point of interest. This preliminary stage makes it possible to narrow the scope of analysis by selecting only websites and pages that are genuinely relevant to the historical, artistic and cultural heritage that the system aims to catalogue. The process then moves into its core phase with the application of Crawling and Scraping techniques.

• Crawling can be thought of as a form of “systematic exploration”: a set of procedures that thoroughly scan the various pages of a website, following internal links and thereby reconstructing the overall structure of the domain. This makes it possible to collect not only the initial page identified as the target, but also any secondary pages containing useful information.
• Scraping, on the other hand, is the complementary process responsible for “extracting” the information that is genuinely relevant to the project from page content. It is a targeted activity in which specific data – such as descriptive texts, images or metadata – are isolated and then transformed into a structured format consistent with the requirements of the ESplora database.

Throughout the acquisition process, the identified content is not immediately entered into the final system, but is initially stored in temporary repositories. This methodological choice serves two purposes: on the one hand, it provides a “working copy” on which further checks can be carried out; on the other, it prevents incomplete or unvalidated data from flowing directly into the main database.
At this stage, additional analysis and filtering procedures come into play. Their purpose is to examine the collected data, discard redundant or low-quality information and prepare the content deemed valid for the subsequent validation stage. It is during this phase that raw data are transformed into resources ready to be integrated into the system and made available for final use.
The crawling process, therefore, is not limited to the simple mechanical acquisition of content, but represents a structured workflow in which search, exploration and extraction techniques work together to ensure that only relevant, reliable and high-quality information becomes part of ESplora’s information assets.

4 – Data Mining – The data mining engine integrated into the ESplora system represents one of the platform’s most significant components. It was not developed from scratch, but originated from solutions previously developed by ES Progetti e Sistemi as part of several earlier projects. With ESplora, this engine was significantly enhanced, updated and adapted to specific requirements, with the aim of ensuring the effective processing of large quantities of heterogeneous data from different sources, almost all of which are retrieved from the web. The information processed mainly comes from websites, online services and other digital resources and, less frequently, from static files. Once acquired, this information undergoes a complex processing workflow, which may include several operations:

• data filtering to eliminate redundancies, inconsistent elements or irrelevant content;
• translation of texts into multiple languages, to ensure the system can be used internationally;
• transformation and normalisation of information, with the aim of adapting it to the data model defined for the ESplora system.

At the end of this processing stage, the content is converted into a coherent structure compatible with the ESplora database, ready to be stored and subsequently made available to end users.
The level of process automation depends on the reliability and type of web resource from which the data originate. When sources are considered reliable and the quality of the content meets specific criteria, the workflow can be completed entirely automatically, allowing direct publication and immediate access to the information through the platform. In other cases, human intervention is required to verify the validity and consistency of the data before they are made available.
This hybrid approach, combining automation with human supervision, ensures both the scalability and efficiency required to manage large volumes of data and the quality and reliability essential for a system such as ESplora, which is designed to enhance cultural and tourism heritage.

5 – Editorial Portal. The editorial portal is a central and essential tool for managing content within the ESplora platform. It is a web app specifically designed to facilitate and improve the efficiency of Point of Interest management by authorised users, providing a complete, intuitive and highly functional working environment.
The application was developed with particular attention to usability and operational efficiency, making interaction simple and fast even when handling large amounts of data. Through the portal, users can: search for POIs already available in the system, view them on interactive maps, edit them and enrich them with new content. The interface, designed to be clear and immediate, makes it easy to navigate through the data, update information in real time and use advanced tools for adding images, textual content and automated translations, thereby making POIs more accessible in an international context.
The portal is not an isolated module, but is closely integrated with the ESplora platform, serving as its main point of interaction for editorial content management. Through it, users can access several key features, including:

• POI publication – once the editorial review stage has been completed, content can be published and becomes immediately available through the ESplora App.
• Targeted crawling – the system allows crawling procedures to be launched for individual Points of Interest, in order to enrich existing data with new information automatically acquired from the web.
• Data versioning – every change made to the content is versioned and stored in the history, with the option to view previous versions at any time and, if necessary, restore them. This feature ensures traceability, transparency and security throughout the management process.
• Assisted translation – textual content can be automatically translated into multiple languages, with the option of further manual review by operators, ensuring multilingual availability of the POIs.
• Creation of thematic itineraries – users can design and structure cultural or tourist itineraries, adding a dedicated description and associating relevant POIs with the routes, thereby turning the portal into a useful tool not only for data management, but also for designing complete visitor experiences.

In this way, the editorial portal is not merely an administration panel, but becomes a fully collaborative working environment, capable of combining technical efficiency with the enhancement of cultural content.

6 – App. The ESplora App represents the final link in the technological chain and is the main tool through which the platform’s services are delivered, drawing directly on the information stored in the central database. Its purpose is to provide end users with simple, immediate and interactive access to content, turning the database into a tool that can be used on the move. The application is compatible with Android and Apple iOS operating systems and is already available on digital stores, although access is currently limited to a restricted test group, which can only be joined by explicit invitation. This pre-release phase is intended to validate functionality, collect user feedback and optimise the overall experience. The public release of the application is scheduled for December 2025, from which point the services will be available to a wider audience.
From a functional point of view, the App was designed to provide practical support to users while visiting places of tourist and cultural interest. Its potential is best expressed in outdoor spaces, thanks to the use of GPS technology, which makes it possible to accurately detect the user’s location and identify nearby Points of Interest.
Access to content has been designed to be as simple and natural as possible: texts, audio content and images are automatically presented to visitors when they are in the vicinity of a POI, minimising the need for manual searches or complex interactions. In this way, the experience is smooth and immediate, enriching the visit with contextualised and easily accessible information.
POIs can be identified in two main ways:

• through an interactive map, which allows users to find their way and plan their routes;
• through augmented reality tools, which provide an immersive experience by allowing Points of Interest to be displayed in real time directly overlaid on the device camera view.

This combination of technologies makes the ESplora App not only a consultation tool, but also an intelligent travel companion capable of enriching the discovery of tourist and cultural heritage through a dynamic, personalised and experience-oriented approach.

Conclusions
The key aspect demonstrating the full achievement of the ESplora project’s objectives is the replicability and transferability of the processes developed. The automated content acquisition methods (crawling, scraping and data mining), the streamlined and efficient database structure, the integrated editorial portal and the multi-device App together constitute a solid and versatile model. Not only have the results achieved been concretely measurable – with thousands of Points of Interest catalogued, organised and made available to users – but the entire approach can also be applied to other territories, at both regional and national level, and even in cultural and thematic contexts different from the original one. The combination of automation and human supervision, the flexibility of the software architectures adopted and the possibility of integration with external sources and initiatives make ESplora a scalable and sustainable project, capable of adapting to new requirements and promoting the dissemination, enhancement and accessibility of cultural heritage in a wide range of contexts. In this sense, the project represents not only an achievement, but also a starting point for further development and future applications.

Future Directions for Research and Development
Looking to the future, ESplora aims to make a significant leap forward by focusing on the integration of artificial intelligence technologies.
The introduction of these tools will make it possible to automate content validation processes, significantly reducing operating times and costs and making the distribution of information across increasingly extensive territories more efficient.
One of the main development priorities is the creation of content with high historical and cultural value, capable of further enriching the database and offering visitors increasingly complete and engaging experiences. At the same time, one of the most ambitious technological challenges concerns the automatic recognition of cultural assets through the camera of a mobile device: this feature will allow users to point their device at a cultural asset and immediately receive related contextual and multimedia information.
Alongside these developments, targeted dissemination and commercialisation strategies are also planned, with the aim of consolidating and expanding ESplora’s presence on the market. These strategies will be pursued in close collaboration with cultural organisations and local institutions, in order to ensure not only the technological growth of the platform, but also its integration within local contexts, fostering synergies, new opportunities for heritage enhancement and increasingly participatory access to cultural heritage.