Reflections on the VIVO project


VIVO is a semantic web tool that allows interdisciplinary networking. My project aimed to explore the use of VIVO within NeSI as a means to both network researchers together for collaboration and also link them through common use of NeSI and its associated tools and software. This will facilitate increased productivity in all areas of research and in particular the sciences and information technology disciplines.
A more detailed outline of the project can be found here.

The road so far

After having introductions and an outline of what was meant to be done during the project, the first thing to do was to get VIVO running. A virtual machine was created to host the program and accessed through command line using MobaXterm. Following the VIVO instructions proved difficult and we kept running into problems including permissions & ownership issues, version problems,  build failures. Even when it was finally hosted and we were able to login, the content would not show because of a problem connecting to solr which appeared in the startup tests. After several attempts to correct this problem, including emailing the vivo mailing list, we decided it would be better to just run an already set-up instance on VirtualBox for our purposes. This was successful and allowed the creation of several user accounts at different levels to see their functionality. The data ingest guide was followed and after a few attempts, some data was successfully loaded from a model csv file. A php script that alters and amends the raw csv data output from HighRise (Human Resource Datanase) was also created so the relevant data can be selected and easily converted into a format that VIVO ingest systems will accept. The jargon used for the hierarchy of classes, properties and their connections has been established, described and documented. A broad overview of the potential use of VIVO in NeSI was created and after a meeting to discuss this, a more refined and specific chart base on use-cases was created.

The immediate Future

The immediate goals are to learn SPARQL methods and syntax to aid the data entry/extraction process, to learn how to delete content and if possible amend content. After this is completed, to enter the actual converted data from HighRise into the system and see how it functions.

In the future

Access the other databases including: the NeSI Wiki, GOLD and Drupal. Produce scripts to convert their data sources into appropriate csv format. Design a custom class and property ontology based on the structure of VIVO and using information from current VIVO ontologies,the use-case diagram and what is available in the databses to produce a plan. Create a full diagram with connections between class groups, classes, property groups, data properties and namespace pointers. Implement this within VIVO and trial with HighRise data then add data from the other sources. Assess the usefulness of VIVO in its current configuration for its purpose and make suitable alterations to the created custom model. Describe the potential for its expansion and use elsewhere.

Submitted by Brendan Tonson-Older on


What's SPARQL?

Hey Brendan, thanks for that. What resources have you been using to learn SPARQL?