The Semantic demonstrator demonstrates the use of semantic technologies for classification and discovery of Europeana objects related to Food and Drink; it aims to provide semantic enrichment, i.e. extract references to Food and Drink topics from free text in object metadata to strengthen the relevance of database search results on the theme of Food and Drink.
The application does this by applying a Europeana Food and Drink Classification scheme to Cultural Heritage Objects, such as those stored in Europeana to yield more precise groupings, relationships and categorisations for database items.
While semantic enrichment and semantic search are technically complex, simple keyword search often returns irrelevant objects because of the ambiguity of the words searched. When you add multilingual search results, the problem is exacerbated. For example:
- ‘Recipe’ can be related to food or to Medicine; ‘fork’ could mean cutlery, an agricultural instrument (pitch fork) or a musical device (tuning fork).
- If you search for “poison” in the collections provided by Swiss institutions, you may find photographs from India and Indian movie covers, this is because in Latvian “poison” is “Inde”, and “Inde” in French means “India”, so when the data was enriched to include multilingual equivalents, it made the search results less accurate.
- If you search for Beer in Europeana and go to a random page, you may find that only 1 of 24 objects are relevant to your query. The rest are images of bears, objects related to de Beer (a very popular Dutch surname), etc.
The Europeana Food and Drink Semantic Application seeks to overcome these ambiguities and make content easier to find and exploit. By providing semantic enrichment and enabling the discovery, classification and exploration of Food and Drink cultural heritage objects, the application could enable a number of valuable and practical uses, for example:
- A member of the Gallery, Library, Archive and Museum (GLAM) community can use it to find interesting objects to use in virtual exhibitions
- A teacher can draw useful primary sources to assist him/her in developing teaching resources for the classroom.
- A chef or gastronomical start-up can source recipes from a particular era to develop a menu for their niche restaurant or cook book or foodie app
- For content providers, it will significantly add to their semantic skills and enhance the documentation of their collections
- For end-users, it will offer an interesting tool for exploring concepts relating to food and drink. It will allow users to explore relationships between human history, society, living conditions, migration, agriculture and commerce and food and drink, highlighting how diverse content under a universal theme can be explored multi-dimensionally
- For Europeana, it will provide the beginnings of a Food and Drink Channel, and develop new technology for Topical Discovery that can be used in other domains as well, e.g. History
The Food and Drink Classification, enrichment results, and the semantic application source will be published for open re-use (not including proprietary Ontotext machine learning components).
For more details, see the Semantic Demonstrator description: please click here.
For more information or if you are interested in exploring semantic approaches for your Cultural Heritage collection, please contact Vladimir Alexiev.