Article Preview
TopIntroduction
The description of cultural heritage resource has become more complex and internationalized since the appearance of the Internet. Web applications are increasingly providing increased visibility of content and enhancing the contextualization of information produced by museum or research institutions, Interoperability between various highly specialized systems, integrated information access and information integration becomes an increasing demand to support research, professional heritage administration, preservation, public curiosity and education (Doerr, Martin, 2009). Before these applications were mainly intended for "human" users who had to guess the indirect relations between cultural heritage, through the use of search engines offered by the human interface machine. Except that if the human being is able to deduce the implicit relations between several information relating to the same object, the machines need an intelligent system that can relate different relations to make cultural resources accessible. It does requires rich metadata structures, able to cover the variety of material held in memory institutions (such as archives, bibliographic and electronic material) (Stasinopoulou, Bountouri, Kakali, Lourdi, Papatheodorou, Doerr, & Gergatsoulis, 2007) This is the very work of metadata in the digital environment, their purpose being to describe the data to improve its indexing and enhance it with search engines but also to make possible the interoperability between different systems. Interoperability is a key concept in what is now called the "data web" because it is by freeing data from the straitjacket of applications that it is possible to navigate "with the flair" and bounce back on additional information to the initial resource.
But by releasing web applications, the information can be described with so many different ways that it becomes redundant and loses its contextualization . It is then essential to document it by adding metadata that will accompany it between the systems via the http protocol. Standards play a vital role in this context of interoperability because they guarantee the uniqueness of syntax and vocabulary in exchanges between systems. Whether XML or RDF or others, depending on the technologies used, they unify the way of expressing the data.
We therefore aim to work on these cultural heritage documents to relate theme with the adequate information that can be either a human resource , another cultural heritage document or simply basic information, in our work that is what we will call in the rest of the paper a ICHD (Intelligent Cultural Heritage Document).
Knowing also the huge amount of data (Volume) , the variety of data (image , text , geographical information ,etc …) and the speed that is required to process the amount of cultural heritage documents that we have to manage guides us into a big data context requiring efficient algorithms and NoSQL databases. That’s why we choose MongoDB as a document- oriented NoSQL database, MapReduce for distributed treatment ,microservices as architecture for our platform with asynchronous communication.
The goal of our system is put the entire cultural heritage (tangible and intangible ) on a searchable application with a map to geolocate the different cultural entities constituting the wealth of the city. In order to set up a good implementation we have to specify a good data model for every type of cultural heritage, so we can describe every cultural heritage document in the best way possible and index them geographically and most importantly relate it to the different human resources who interacted with.