Periodic Reporting for period 2 - SPICE (Social cohesion, Participation, and Inclusion through Cultural Engagement)
Periodo di rendicontazione: 2021-05-01 al 2023-04-30
The overall aim of the SPICE project is to develop tools and methods to support Citizen Curation, in which citizens actively engage in curatorial activities to learn more about themselves and develop a better understanding of, and empathy for, other communities.
# Social Inclusion: Supporting citizens in finding and articulating their own voice in an inclusive way. SPICE results provide citizens with data-centric, artificial intelligence-informed technologies, and methods that support them in interpreting culture for themselves. These are designed in an inclusive way by framing them as familiar or entertaining activities. Activities are accessible to people with varied skills and abilities.
# Personalized Interpretation: Supporting citizens in interpreting art in a way that is personally relevant and consistent with the interpretative guidance and intent of the museum and/or the artist. SPICE technologies and methods guide the interpretation process by, for example, prompting citizens to notice features of the artwork, or what the artist was trying to say. The technology guides the citizens in developing their own personal interpretations.
# Reflection: Making multiple voices enabled by Citizen Curation available to visitors, who can engage with them productively. The varied interpretations contributed by citizens are combined to help citizens understand the perspectives of others. For example, visualisations will be provided that show the viewpoints of citizen groups such as older people, children, people with disabilities, and people of distinct cultural and ethnic backgrounds.
# Data-driven Platform: Designing/applying technologies that enable (a) museums to control how their digital assets are used and (b) citizens to manage how their contributions and identity are shared. The underlying technology developed to support Citizen Curation enable citizens to manage and control any digital contributions they make, for example, easily accessing any stories or comments they have added, editing, or removing them.
# Operationalised Meaning in Citizen Curation: Putting existing data schemas, metadata standards, and informal theories/text in the context of a large network of knowledge graphs to capture multiple perspectives with an interoperable semantics. Museums are able to manage their digital resources and understand how citizens respond to their collections. Citizen Curation involves the development of new ways of describing artworks in terms of how they let citizens feel and respond to artworks that are e.g. scary, controversial, or comforting.
# Case studies: Five case studies address citizen communities, using participatory, co-design activities to the development of Citizen Curation methods. For example, Case Study Finland includes elderly citizens, rural dwellers, and asylum seekers: in that case, spatial distance is correlated to exclusion and disadvantage. For the elderly, spatial distance relates to a lack of physical mobility. For the rural dwellers, spatial distance relates to great geographical distances that limit frequent physical engagement and interaction. For asylum seekers, it is a lack of knowledge about the national culture that are going to encounter, thus limiting cultural and social engagement.
# Social Inclusion: Supporting citizens in finding and articulating their own voice in an inclusive way. SPICE results provide citizens with data-centric, artificial intelligence-informed technologies, and methods that support them in interpreting culture for themselves. These are designed in an inclusive way by framing them as familiar or entertaining activities. Activities are accessible to people with varied skills and abilities.
# Personalized Interpretation: Supporting citizens in interpreting art in a way that is personally relevant and consistent with the interpretative guidance and intent of the museum and/or the artist. SPICE technologies and methods guide the interpretation process by, for example, prompting citizens to notice features of the artwork, or what the artist was trying to say. The technology guides the citizens in developing their own personal interpretations.
# Reflection: Making multiple voices enabled by Citizen Curation available to visitors, who can engage with them productively. The varied interpretations contributed by citizens are combined to help citizens understand the perspectives of others. For example, visualisations will be provided that show the viewpoints of citizen groups such as older people, children, people with disabilities, and people of distinct cultural and ethnic backgrounds.
# Data-driven Platform: Designing/applying technologies that enable (a) museums to control how their digital assets are used and (b) citizens to manage how their contributions and identity are shared. The underlying technology developed to support Citizen Curation enable citizens to manage and control any digital contributions they make, for example, easily accessing any stories or comments they have added, editing, or removing them.
# Operationalised Meaning in Citizen Curation: Putting existing data schemas, metadata standards, and informal theories/text in the context of a large network of knowledge graphs to capture multiple perspectives with an interoperable semantics. Museums are able to manage their digital resources and understand how citizens respond to their collections. Citizen Curation involves the development of new ways of describing artworks in terms of how they let citizens feel and respond to artworks that are e.g. scary, controversial, or comforting.
# Case studies: Five case studies address citizen communities, using participatory, co-design activities to the development of Citizen Curation methods. For example, Case Study Finland includes elderly citizens, rural dwellers, and asylum seekers: in that case, spatial distance is correlated to exclusion and disadvantage. For the elderly, spatial distance relates to a lack of physical mobility. For the rural dwellers, spatial distance relates to great geographical distances that limit frequent physical engagement and interaction. For asylum seekers, it is a lack of knowledge about the national culture that are going to encounter, thus limiting cultural and social engagement.
As detailed in the Technical Report, the main achievements can be summarized as follows:
* Methods for activating the Interpretation-Reflection-Loop (IRL) have been selected (WP2, WP5), adapted to case studies (WP2, WP7, WP5), and partly implemented with prototypes (WP3, WP4, WP5) and co-design workshops (WP2, WP7).
* We have analysed (WP2) how richer representations of citizen groups can be derived from the data-driven modules of the platform: interfaces, user/community modelling, recommender, ontology-based querying/reasoning, etc. (WP3, WP4, WP5, WP6).
* Prototypes to configure, detect, visualize, understand, explain, and navigate through citizen communities have been developed (WP3, WP4, WP5, WP6).
* A semantic multilingual annotator has been implemented for English, Finnish, Hebrew, Italian and Spanish (WP3). It annotates opinion polarity (“sentiment”), emotion and public identity of entities. The reference system for annotation comes from the SON ontologies (WP6) and stores the generated RDF graph in the linked data hub (WP4).
* We designed the Architecture Layout of the technical ecosystem for the SPICE project (WP4). It currently consists of a novel universal transformer for arbitrary data (SPARQLAnything), a novel platform for linked data storage (the Linked Data Hub), and a Linked Data Intelligence layer to connect the User Model management component (WP3), the community model API (WP3), the Semantic Annotator (WP3), and the Ontology Reasoner (WP6) technical research infrastructure, developed to store, query, and reason on ontologies and ontology-based knowledge graphs.
* We developed the SPICE Ontology Network (SON) GitHub repository (WP6), including newly designed (or integrated) state-of-the-art ontology modules enabling the description of cultural objects and their sense-making aggregation during curatorial, interpretation, and reflection activities.
* WP2 and WP5 have identified the necessary interfaces for citizen curation within the case studies.
* Several events, including mini conferences, were co-designed, targeted at the case study members (WP7). We also conducted ethnographic interviews with members involved in technical WP to understand their background, approach, and perspectives towards the project.
* Methods for activating the Interpretation-Reflection-Loop (IRL) have been selected (WP2, WP5), adapted to case studies (WP2, WP7, WP5), and partly implemented with prototypes (WP3, WP4, WP5) and co-design workshops (WP2, WP7).
* We have analysed (WP2) how richer representations of citizen groups can be derived from the data-driven modules of the platform: interfaces, user/community modelling, recommender, ontology-based querying/reasoning, etc. (WP3, WP4, WP5, WP6).
* Prototypes to configure, detect, visualize, understand, explain, and navigate through citizen communities have been developed (WP3, WP4, WP5, WP6).
* A semantic multilingual annotator has been implemented for English, Finnish, Hebrew, Italian and Spanish (WP3). It annotates opinion polarity (“sentiment”), emotion and public identity of entities. The reference system for annotation comes from the SON ontologies (WP6) and stores the generated RDF graph in the linked data hub (WP4).
* We designed the Architecture Layout of the technical ecosystem for the SPICE project (WP4). It currently consists of a novel universal transformer for arbitrary data (SPARQLAnything), a novel platform for linked data storage (the Linked Data Hub), and a Linked Data Intelligence layer to connect the User Model management component (WP3), the community model API (WP3), the Semantic Annotator (WP3), and the Ontology Reasoner (WP6) technical research infrastructure, developed to store, query, and reason on ontologies and ontology-based knowledge graphs.
* We developed the SPICE Ontology Network (SON) GitHub repository (WP6), including newly designed (or integrated) state-of-the-art ontology modules enabling the description of cultural objects and their sense-making aggregation during curatorial, interpretation, and reflection activities.
* WP2 and WP5 have identified the necessary interfaces for citizen curation within the case studies.
* Several events, including mini conferences, were co-designed, targeted at the case study members (WP7). We also conducted ethnographic interviews with members involved in technical WP to understand their background, approach, and perspectives towards the project.
Concerning expected progress beyond state of the art and broader impact:
* The operational deployment of the IRL is an important result of SPICE, which gathers benefits beyond social inclusion through cultural heritage, e.g. in public discussion platforms, ethical, political and moral issues, social categorization, sustainable goal achievement, etc.
* The “rich citizen representation” paradigm emerging from methods, tools, annotations, ontologies, and co-design activities is one of the foundations of the SPICE ecosystem, providing ways to operationalize profiling, self-assessment, extended knowledge of citizen groups, hence supporting the powerful IRL vision of SPICE, with great potential for impact beyond the project.
* Annotation tools, based on deep cultural requirements and carefully designed ontologies, associated with a pragmatic computational attitude, offer an alternative to bulk, off-the-shelf generic tools for e.g. sentiment analysis and emotion detection, establishing a different milestone for scientific advancement in the field.
* The technical ecosystem including a new universal data transformer (SPARQLAnything) and a new streamlined linked data storage system (DataHub) including communication to multiple, loosely coupled, local applications, is a major software architecture result.
* The SPICE Ontology Network lines in the nouvelle vague of networked ontologies in the humanities, such as the reused ArCo network for cultural entities, but it delves into areas (emotions, moral values, narratives, action frames, etc.) that are not typically considered in mainstream projects.
* A novel paradigm of data- and semantics-centric human-machine interaction has been created in SPICE, integrating to the requirements, use cases, data layers, and ontologies emerging in the project.
* The operational deployment of the IRL is an important result of SPICE, which gathers benefits beyond social inclusion through cultural heritage, e.g. in public discussion platforms, ethical, political and moral issues, social categorization, sustainable goal achievement, etc.
* The “rich citizen representation” paradigm emerging from methods, tools, annotations, ontologies, and co-design activities is one of the foundations of the SPICE ecosystem, providing ways to operationalize profiling, self-assessment, extended knowledge of citizen groups, hence supporting the powerful IRL vision of SPICE, with great potential for impact beyond the project.
* Annotation tools, based on deep cultural requirements and carefully designed ontologies, associated with a pragmatic computational attitude, offer an alternative to bulk, off-the-shelf generic tools for e.g. sentiment analysis and emotion detection, establishing a different milestone for scientific advancement in the field.
* The technical ecosystem including a new universal data transformer (SPARQLAnything) and a new streamlined linked data storage system (DataHub) including communication to multiple, loosely coupled, local applications, is a major software architecture result.
* The SPICE Ontology Network lines in the nouvelle vague of networked ontologies in the humanities, such as the reused ArCo network for cultural entities, but it delves into areas (emotions, moral values, narratives, action frames, etc.) that are not typically considered in mainstream projects.
* A novel paradigm of data- and semantics-centric human-machine interaction has been created in SPICE, integrating to the requirements, use cases, data layers, and ontologies emerging in the project.