Europeana Fashion project overview @Carare workshop


Published on

Published in: Lifestyle, Business
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide
  • [ Scattered access ] Access to fashion digital collections is often “on-site” only. Databases are usually kept locally in closed systems, and even if they’re on-line, there are no integrated access points. [ Lack of interoperability ] There are no common metadata standards to properly describe Fashion objects. And institutions haven’t implemented protocols to exchange data and to offer integrated access to their users. Interoperability at the semantic level is also lacking. [ Rights issues ] Different IPR legislations in different countries across Europe and worldwide, and increased expectations regarding access to digital content by the public made rights issues an important point to tackle when offering access to digital collections. [ Lack of exploitation strategies ] There are few case studies regarding possible ways of exploitation of Fashion digital content for academic and scholarly access and for the creative industry at large.
  • With the ambition to address all these issues we setup a consortium made of 23 partners from 12 European countries…, which aims at building a thematic fashion aggregator.
  • Fashion content in fact is currently not well represented there, despite fashion is a popular theme that potentially could attract a large audience. So building such an aggregator will represent a great value for Europeana itself but also for the fashion community, allowing to establish best practices, tools and solutions for accessing and integrating heterogeneous content in the fashion domain. Europeana Fashion addresses the ICT PSP Objective 2.1 “Aggregating content for Europeana”.
  • Gathering and harmonising fashion digital content coming form the most important European fashion collections (both public and private). (of content holders) Developing and promoting a common metadata model for fashion content, based on the EDM, and developing also tools for semantic interoperability like the Fashion Thesaurus. Setting up the Europeana Fashion portal, that will offer an integrated access to all the aggregated fashion collections, offering to users added value services. (on-line) Stimulating synergies between fashion content holders and the creative community for the reuse of on-line fashion content (starting also a specific collaboration with the Wikipedia community). In the fashion community on digitisation, IPR issues and interoperability, with the aim also to attract new fashion content providers in the BPN.
  • Since we are aggregating a huge amount of interesting fashion content, we would like to relate it with the outside world of the web and offer something to users that allows them and us to better exploit pur aggregated content.
  • This idea can be applied in different domains, but in fashion makes much more sense…
  • Europeana Fashion project overview @Carare workshop

    1. 1. Europeana Fashion CARARE workshop 24-25 May 2012 Project overview Marco Rendina Fondazione Rinascimento Digitale Co-funded by the EuroSupport peanConnected to Commission within the ICT Policy Programme
    2. 2. Europeana Fashion in a nutshell The Context the scenario in which the project operates The Consortium the composition of our project consortium Objectives and expected outcomes aims and results of the project …Some fun stuff we’re going to do Wikipedia collaboration and topic & media analysis tool
    3. 3. The ContextThe scenario in which the project operates Fashion is an important part of our European Cultural Heritage. A growing interest in Fashion content is developing in a broader audience (not only professionals and domain specialists). Digitisation and on-line access is fundamental in allowing museums and archives to exploit their holdings. But there are still some remaining challenges…
    4. 4. Challenges and open issuesMany Fashion institutions, both public and private, reconsidered the role oftheir archives as an important resource. This led to new investments inconservation and digitisation of their heritage, but still some issues remains: Scattered access Lack of interoperability Rights issues Lack of exploitation strategiesEuropeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    5. 5. The ConsortiumThe composition of our projectconsortiumThe Europeana Fashion Best PracticeNetwork will encompass 23 partnersfrom 12 European countries, whichrepresent the leading Europeaninstitutions and collections in thefashion domain.
    6. 6. The Europeana Fashion ConsortiumApart from Fondazione Rinascimento Digitale, which is coordinating theproject, the consortium is made of: 20 Content ProvidersRepresenting the leading European institutions in the domain, including both privateand public archives and museums: - ModeMuseum - Victoria & Albert Museum - KMKG - Netherland Institute for Sound & Vision - Stiftung Preussischer Kulturbesitz - Catwalk Pictures - Stockholms University - Belgrade Museum of Applied Arts - Les Arts Décoratifs - MUDE - Peloponnesian Folklore Foundation - Roger Dean Photography - Emilio Pucci Archive - Pitti Immagine - Centraal Museum Utrecht - Nordiska Museet - Rossimoda Shoe Museum - Museo del Traje - Wien Museum - Archivio Missoni 2 Technical PartnersRepresenting the state-of-the-art in the domain of semantic interoperability and on-lineaccess: - National Technical University of Athens - Internet ArchitectsEuropeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    7. 7. The Europeana Fashion ConsortiumIn the consortium there are also: 3 Associated PartnersRepresenting the state-of-the-art in the domain of semantic interoperability and on-lineaccess: - Berg Publishers - Central Museum of Textiles, Lods - Museo Salvatore FerragamoEuropeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    8. 8. Objectives and expected outcomesAims and results of the projectThe main objective of the EuropeanaFashion Best Practice Network is tobuild a thematic aggregator onFashion, putting emphasis on thequality, the variety and granularity ofcontent that will be delivered toEuropeana, and producing asignificant increase of the fashionrelated content in it.
    9. 9. Project Objectives and expected outcomesEuropeana Fashion objectives aim to: aggregate 700.000+ digital items of fashion content toEuropeana improve interoperability in the fashion community create a specialised access point for fashion content develop case studies for the exploitation of fashionmaterial build consensus and raise awareness on best practicesEuropeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    10. 10. Some fun stuff we’re going to do…
    11. 11. Setting up the collaboration with Wikipedia We will set up a direct collaboration with Wikimedia Chapters, the independent organizations founded to support and promote the Wikimedia projects within a specified geographical region. The aim is to invite consortium members to contribute sample content to Wikimedia Commons and subsequently host a number of so-called GLAM-WIKI events
    12. 12. Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    13. 13. Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    14. 14. Winter picture of Chiajna Monastery. The monastery is situated on the outskirts of Bucharest, Romania. by Mihai PetreEuropeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    15. 15. • dgdsEuropeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    16. 16. Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    17. 17. We plan to host events in 5countries starting from May2013.
    18. 18. Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    19. 19. …and now something completely different! Topic and Social media analysis •what we would like to do: • understand how a wide range of media & data sources relate to each other • get a synthesis of meaningful geo-located information • merge these sources to derive insights & analytics •in a nutshell, measure the buzz and aggregate media surrounding interesting topics (codename: Flux of Meme)
    20. 20. The ideaMeme: a postulated unit or element of cultural ideas transmitted from onemind to another through speech or similar phenomena.Zeitgeist: German language expression referring to "the spirit of the times"Semantic Web: an evolving development of the World Wide Web in whichthe meaning (semantics) of information on the web is defined, making itpossible for machines to process itFlux of Meme: analysis of the web Zeitgeist through geo-localized Memes, updated and shared on social media mainlyvia mobile networks Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    21. 21. The Fashion use case• $1,000 billion industry• examples of how social media events affect the real world: • British fashion designers have pound signs in their eyes and ka-chings in their ears • Whatever she [Kate Middleton] is wearing sells out instantly — both in store and online [...] The Kate Effect that her influence has been calculated to be worth an astonishing £2 billion to the British economy• lack of effective tools for topic & media analysis Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    22. 22. The Fashion use case• key technologies: topic extraction and image analysis• input type: blog posts, RSS or Twitter feeds, image databases, etc.• user stereotype: fashionistas / trend/fashion researcher / market• need: analyze online presence and social media buzz related to brands or chosen topics around specific locations, evaluate style trends through image analysis Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    23. 23. Algorithm steps1. fetch data 2. create clusters 3. extract topics 4. analyze stats Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    24. 24. The implementation1. fetch big data: tweets, google+, blogs, images & videos. all data is sanitized and stored2. enrich data crawling the web to find correlations from additional sources (e.g. related twitter profiles, instagram, pinterest, public facebook timelines, news, etc.)3. create geo-located clusters and time series from the mash-up of data sources4. extract topics, entities (i.e. brands, organizations, names, etc.), locations and image similarities using machine learning models (Named Entity Detection, Latent Dirichlet Allocation, ...) and Scale Invariant Feature Transform (SIFT) for feature detection5. aggregate & synthesize data for final presentation Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    25. 25. 1. content crawled from the internet is sent to storage & analyze engine 2. unstructured data is extracted from content and stored for further inference4. users can access analytics datathrough rich media query systems 3. data is semantically enriched with additional sources Europeana Fashion @ CARARE workshop Mykonos, 24-25 May 2012
    26. 26. or simply browse clusters’ content to search for image similarity, brand, analyze trends through their images & year & season, free text social media buzzadvanced searchshows clever fields,not only the usualsuspects
    27. 27. Thank you for your attention! Contact me at or follow me @mrendinaConnected to Co-funded by the European Commission within the ICT Policy Support Programme