Vallecupola Ecosystem

Metadata Analysis

Theoretical Model

As a first step, it was necessary to shape the knowledge domain, connecting the selected items. In order to show the interconnections between the objects, it was created a graph made of nodes and edges labelled in natural language, explicitly describing the important links between the items. Of course, only the most meaningful features and relations were chosen. The graph resembles a spider web, making it a small dense "knowledge graph", representing real instances and their meaningful bonds.

For the reader, it will be sufficiently easy to follow the various paths that bring from one object to the other, understanding the density and relevance of the cultural framework built all around Vallecupola.

In this theoretical model, not only can we see the items themselves, but also some extra entities necessary to connect them (they are basically intermediate nodes).

In the following reproduction of the theoretical model:

• the green circle represents the main concept of the domain;
• the yellow circles represent the twenty chosen items composing the exhibition;
• the grey circles represent the extra entities (intermediate nodes).

Conceptual Model

Our conceptual model provides a structured approach to representing the complex relationships between people, places, objects, and events within the Vallecupola ecosystem. It serves as the foundation for all subsequent data modeling and representation efforts.

Starting from the theoretical model, containing the individuals of the knowledge domain, put in relation between each other, it was possibile to elaborate a "conceptual model" by converting the previous into a more abstract level of representation. We collapsed the various individuals in classes, grouping similar ones, also based on their common features.

The result is a schema made of 11 classes, of which:

• 7 representing the museum items:
• dcmitype:Text for collections, poems, stories;
• crm:E24_Physical_Human_Made_Thing for various objects and artifacts;
• schema:VisualArtwork for paintings and stamps;
• fabio:MovingImage for videos;
• schema:Sculpture for sculptures and installations;
• fabio:MusicalComposition for songs and musics;
• frbroo:F31_Performance for stage performances;
• 4 representing only the extra entities:
• crm:E53_Place for villages, hamlets and locations in them;
• crm:E21_Person for people, artists;
• schema:Museum for museums holding some items;
• skos:Concept for subjects represented or topics conveyed by the cultural items.

Actually, also some extra entities have classes dcmitype:Text (scenes of the story), crm:E24_Physical_Human_Made_Thing (objects mentioned in the story) and fabio:MovingImage, because they were useful to better express the connections.

This is a model on a really high level of abstraction. It resembles an ontology. Here - as in the previous theoretical model - only the most meaningful part of the knowledge domain has been shown. Indeed, we can only see classes and relations, not all the attributes associated to the classes.

In the following list, we recap all the predicates that we used for each class when describing their instances in detail during the CSV creation phase (in yellow the actual relations that you can see as edges in the conceptual model above):

• dcmitype:Text:
dcterms:type, dcterms:title, dcterms:creator, dcterms:publisher, dcterms:spatial, dcterms:issued, dcterms:isPartOf, dcterms:language, dcterms:coverage, dcterms:description, dcterms:identifier, dcterms:source, frbr:part, schema:award, foaf:page, schema:contentLocation, schema:mentions, schema:citation, prov:influenced;
• crm:E24_Physical_Human_Made_Thing:
crm:P2_has_type, dc:subject, dc:title, dc:description, dcterms:issued, dcterms:valid, dc:source, nmo:hasWeight, nmo:hasDiameter, nmo:hasThickness, nmo:hasShape, crm:P50_has_current_keeper, dc:rights, dcterms:isReferencedBy, dc:coverage, pico:materialAndTechnique, a-cd:hasDating, a-cd:hasUse, a-dd:hasTechnicalStatus, a-cd:hasDocumentation, dcterms:created, a-dd:hasDesignationInTime, a-dd:hasTechnique, dc:format, dc:temporal, dcterms:provenance, dc:creator, schema:material, schema:locationCreated, schema:artMedium, schema:image, schema:weight, schema:isSimilarTo;
• schema:VisualArtwork:
rdf:type, crm:P129_is_about, crm:P102_has_title, crm:P14_carried_out_by, crm:P4_has_time-span, crm:P138_shows, crm:P45_consists_of, crm:P70_documents, crm:P104_is_subject_to, crm:P43_has_dimension, crm:P50_has_current_keeper, crm:P55_has_current_location, crm:P53_has_former_location;
• fabio:MovingImage:
rdf:type, dcterms:title, dcterms:creator, dcterms:date, dcterms:publisher, crm:P50_has_current_keeper, dcterms:description, dcterms:extent, crm:P138_shows, foaf:page;
• schema:Sculpture:
crm:P2_has_type, crm:P102_has_title, crm:P14_carried_out_by, crm:P50_has_current_keeper, crm:P129_is_about, crm:P3_has_note, crm:P4_has_time-span, crm:P70_documents, crm:P55_has_current_location, crm:P45_consists_of, crm:P43_has_dimension;
• fabio:MusicalComposition:
rdf:type, mo:title, mo:iswc, mo:performer, mo:duration, mo:record, mo:release_date, mo:instrument, mo:recorded_in, dcterms:language, mo:genre, mo:publisher, dcterms:coverage, mo:producer, mo:lyricist, mo:composer, dcterms:contributor, dcterms:subject, frbr:part, fabio:hasManifestation;
• frbroo:F31_Performance:
rdf:type, event:place, dcterms:date, mo:performer, dcterms:contributor, frbr:exemplar;
• crm:E53_Place:
rdf:type, owl:sameAs, rdfs:label, schema:containedInPlace, prov:influenced, frbr:part;
• crm:E21_Person:
rdf:type, owl:sameAs, rdfs:label, schema:birthPlace;
• schema:Museum:
rdf:type, owl:sameAs, rdfs:label, dbpedia:namedAfter, schema:location;
• skos:Concept:
rdf:type, rdfs:label.

As shown, in order to describe cultural objects in a standard-compliant and interoperable way, Vallecupola Ecosystem uses many classes and predicates taken from already existing ontologies. Models cover simple to complex metadata, enabling rich, integrable knowledge.

Here, we list all the selected ontologies with prefix, full name, namespace and short description:

When possible, entities have been aligned with authority controls. In particular, the controlled vocabularies used by the project are:

CSV Descriptions of Items

Structured tabular data representations provide a straightforward and accessible format for describing exhibition items. The CSV format ensures compatibility with spreadsheet applications and database systems while maintaining human readability.

The project knowledge representation strategy is built around accessibility, reusability, and interoperability within the context of a Linked Open Data (LOD) project.

We adopted a multi-format approach to ensure that the cultural data can serve diverse audiences and technical environments.

To model the cultural heritage objects and their related entities, we created a comprehensive Excel file containing multiple sheets. Each sheet corresponds to a distinct conceptual table in the data model. The format of the tables follows a Subject–Predicate–Object structure, combining both:

• Human-readable descriptions (in natural language), and
• Machine-readable URIs (for semantic web compatibility).

This approach ensures that the metadata can be easily understood by non-experts while remaining fully compatible with LOD principles.

Given that Excel is a proprietary format, we exported each sheet into individual CSV files to enhance long-term accessibility and machine processing. CSV was chosen for its wide support across platforms, readability, and compatibility with both relational databases and RDF transformation workflows.

In total, we produced 22 CSV files:

• 20 CSVs describe the core exhibition objects.
• 1 CSV contains metadata for additional entities (those referenced but not primary objects — often displayed in grey in our conceptual map).
• 1 CSV provides a prefix–namespace mapping table, listing all ontological prefixes used in the dataset alongside their corresponding full URIs.

All the files produced for the project are:

• Stored in the GitHub repository, serving as both a version-controlled archival platform and a public access point.
• Downloadable in the Download Section of this website.

XML-TEI Encoding of the Story

The Text Encoding Initiative (TEI) standard has been employed to encode the narrative elements of the Vallecupola story (one of our exhibition items). This approach ensures scholarly rigor and enables sophisticated textual analysis and presentation options.

After a minimal teiHeader, the content of the story has been encoded only with TEI elements that were functional both for:

• the following conversion from XML to HTML, because we used ref to link some text portions to the exhibition webpages describing the items. This HTML version serves as a narrative guide for visitors, letting them read the story while exploring collection items in the order they appear in the narrative;
• the following conversion from XML to RDF, because we used xml:id global attribute to uniquely identify entities connected to the story node (shown in the theoretical model).

Other meaningful TEI elements we used are the following:

• div1, marking the structural division of the text in six main blocks (using indeed the attribute type="block");
• div2, marking the relevant scenes shown in the theoretical model. Since XML's tree structure required us to also mark other content with div2, we added type="scene" only to actual scenes to distinguish them from other content;
• rs, marking the strings referring to some mentioned objects (that are items of our exhibition);
• quote, marking the citation of music verses.

HTML Conversion of the Story

Web-ready HTML versions of the encoded texts provide immediate accessibility for online audiences. The conversion process maintains the structural integrity of the original TEI encoding while optimizing for web presentation.

For this stage, the goal was the development of the XSLT stylesheet and the generation of the final HTML output from the TEI-encoded short story. The primary objective was to ensure that the semantically marked-up XML could be rendered as a clean, accessible HTML document, while preserving the narrative structure and interpretive elements encoded in TEI.

The TEI file was based on the TEI P5 guidelines and included a variety of structural and semantic elements, such as <teiHeader>, <div1>, <p>, <ref>, <lg>, and <milestone>. As the TEI namespace (http://www.tei-c.org/ns/1.0) was used as the default, a prefix (tei) had to be declared and applied within all XPath expressions in the XSLT.

In the transformation process, the following tasks were carried out:

• The title and author information were extracted from the <teiHeader> and inserted into the <head> and <body> sections of the HTML using XPath.
• Elements such as <div1>, <div2>, and <p> were converted into corresponding HTML <div> and <p> tags to maintain semantic and structural clarity.
• Tags used for quotations, hyperlinks, and poetic or lyrical formatting—including <quote>, <ref>, <lg>, and <l>—were matched to semantic HTML equivalents: <blockquote>, <a>, and styled <div> elements.
• Special handling was done for visual breaks, with <milestone rend="*" /> being transformed into <hr /> elements.

The transformation was done using the online tool FreeFormatter.com, which allowed the TEI XML and XSLT files to be processed together and the resulting HTML to be previewed and downloaded. This approach was chosen for its accessibility and convenience, as a local XML/XSLT processing environment is not available.

RDF Conversion from CSV and XML

Target Audiences

The website is designed to serve multiple target audiences, including academic researchers, cultural heritage professionals, tourists, and local community members. Each audience has specific needs and expectations that inform the design and functionality decisions.

HTML Structure

The HTML structure follows semantic markup principles and web accessibility guidelines. The markup provides a solid foundation for styling and interactive functionality while ensuring compatibility across different browsers and assistive technologies.

CSS Theme 1

The first theme emphasizes clean, minimal design with a focus on readability and accessibility. It uses a neutral color palette and traditional typography to create a professional, scholarly appearance.

CSS Theme 2

The second theme incorporates warmer colors and more dynamic visual elements to create a more engaging and inviting experience. It balances visual appeal with functional requirements.

CSS Theme 3

The third theme explores contemporary design trends with bold typography and vibrant colors. It aims to attract younger audiences while maintaining professional standards.

CSS Theme 4

The fourth theme focuses on high contrast and accessibility features, ensuring that the content is accessible to users with various visual needs and preferences.

Chronological Order

The chronological narrative presents the story of Vallecupola through time, allowing users to understand the historical development of the village and its cultural significance.

Story

The story-based narrative provides a more literary approach to exploring the content, weaving together historical facts, cultural insights, and personal experiences into a cohesive narrative.