VIVO as a Linked Open Data Enabler for the Université du Québec Network
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 22 | |
| Author | ||
| Contributors | ||
| License | CC Attribution - NonCommercial 3.0 Germany: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/47957 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
20
00:00
Computer animation
00:15
Computer animation
00:33
Computer animation
00:47
Computer animation
01:15
Computer animation
01:39
Computer animation
02:14
Computer animation
02:36
Computer animation
03:01
Computer animation
03:17
Computer animation
04:36
Computer animation
05:03
Diagram
05:57
Computer animation
07:14
Computer animation
07:39
Computer animation
09:09
Computer animation
Transcript: English(auto-generated)
00:05
Welcome to this session where we will talk about Linked Open Data and the Vivo merging. So, I will pass over this picture. It's just explained in the context of the University of Quebec where we are many, many students and many, many institutions inside,
00:30
distributed around all of Quebec. And the objective of this presentation is to present an architecture where Vivo is inside the Linked Open Data project.
00:47
So, what is the Linked Open Data? It's the main use case of the Semantic Web where the ID is putting all the data on the web in an open format,
01:05
which is the RDF notation and that's it. So, what we explain here, everybody knows that, is what is a Semantic Web Server.
01:23
You have a triple store, sparkle endpoint, using HTTP, RDF data annotation, URI, and what we observe is a Vivo server is ready to be a part of the Linked Open Data. So, a main functionality inside the Open Data is the Love, is the Linked Open Data vocabulary.
01:50
It's a type of dataset hub where you can put in vocabularies. So, inside the University of Quebec project, we would like to use this hub to retrieve some expertise ontology
02:10
or put in on them some expertise ontology. Another important point to put data on the Linked, on the LOD, is using public URL.
02:24
So, with Perl.org, we can use this platform to standardize the Perl we will use in the project. So, I will take more time on this aspect, is the LOD Capability Map Explorer.
02:47
It's a proof of concept we built just to evaluate a situation of using multiple instances of Vivo to make a map explorer in the LOD.
03:02
This proposal of this section is a demonstration that is possible to publish on the web the Capability Map Explorer that federates expertise navigations between UQ Network and UCam researchers. The diagrams below present the vision associated with the proof of concept.
03:22
To achieve the goal of the POC, the first step is to extract unlinked data from a different data source of different format and notation and transform them into RDF notation to be stored into the Vivo triple store. In the present example, three expert data sources are converted to stores in Vivo that are on institutions and Tranet.
03:44
The datasets are dust-linked but not open data. Some data come from the UCam data source, others come from the UQ Network, while others may come from other institutions. As far as four data sources, they come from the expertise vocabulary standardized by the Canadian government
04:03
and directly disseminated in a Vivo that is in the LOD. Even though it is not part of the proof of concept, the Canadian vocabulary could be also included in the LOD. The last component of the proof of concept is the one of the right-hand column. The application populator is an application that is in fact a federated SPARQL query that links data from different local Vivo
04:28
to extract open data and aggregate it into a Vivo that will be federated in a LOD application. Let's talk about standardized vocabulary. The Canadian government is proposing a standardization vocabulary
04:42
called Canadian Research and Development Classification. The government agency opened this vocabulary with its store in the Excel file and is not usable as it is in the LOD. To step the transformation of the Excel file into the RDF dataset is therefore necessary to complete in order to publish the vocabulary in the LOD.
05:05
Quickly, the process of transforming local data into open data is carried out in three major steps, which are represented here by ovals. The first step is to extract the local data and replace the organizational vocabulary skill with the standardized skills that are in the CRDC.
05:24
Once transformed, the data is translated into a turtle notation and ready to process by the next step. Step two is to publish the turtle file in the Vivo triple star, which are in each organizational institutional intranet. Process three is a sparkle query that federates the data contained in the
05:44
set of organizational triple stars to extract the open data about skill and expertise to push them into the UCAM capability map triple star, which then becomes a node that is part of the LOD. We will focus here on the representation of the CRDC vocabulary.
06:03
To present this vocabulary, we will use the UCAM Dev, a development environment that allows us to develop the Vivo components. Note that the UCAM Dev will present in more details in a future presentation in this conference. As a development tool, UCAM Dev incorporates another tool called StopBread Composer Free Edition, which is in fact an ontology editor.
06:26
The presentation in point one is divided into several views. The point two, we see the taxonomy of the vocabulary that's come from the transformation of the Excel files offered by the CRDC.
06:40
We see a taxonomy of expertise class contained in the vocabulary. View three presents the properties associated with the CRDC vocabulary, including the HasLevels property. View four presents the set of individual contents in the class selected in view two. The selected individual is presented to inform in the view five where the different attributes associated with this vocabulary element can be seen.
07:08
One can see the label of expertise, which is both French and English. We will pass over this one. This is a representation of the data we published from UCAM.
07:22
We also include inside this representation the Vivo ontology, the friend of a friend taxology, and the Oboe ontology. So we see a contextualized individual with inside his ontology.
07:40
Here is now the query that allowed to federate the data from the different institutional servers in order to populate the Vivo that will be in the log. Without going into the detail of this query explanation, let's simply note that using the service command that query can delegate the search to a client server,
08:03
and that it is therefore a grouping of data coming from a cluster of data servers. In this query, we will refer to the CRDC expertise vocabulary server, a UCAM and UQ expertise ontology server. The server with the namespace expertserv is in fact the one that holds the update command, meaning that is updated on data in its own triple store.
08:28
This view shows the result of the UQ-UCAM capability map application. In fact, it represents an expertise graph around the ontology expertise. The application identified three professionals with this expertise.
08:42
On the view on the right, there are two professionals who belong to the University of Quebec network and one professional who belongs to the University of Quebec in Montreal. This is therefore an application that federates an expertise of professionals from several institutions whose data
09:01
sources are distributed in several Vivo's and that use a standardized open vocabulary to identify expertise. In conclusion, that linked open data is a cloud of RDF triples that make accessible dataset vocabularies, ontologies, and can cross-reference the specific application.
09:23
A Vivo instance in the semantic web can also be a node in the log cloud. The domain data can be distributed in several Vivo instances and be federated by a SPARQL services query. A Vivo instance on an institution's intranet is also a useful graph data source that can be used by the enterprise.
09:46
Thank you.