Permalink
Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
54 lines (30 sloc) 3.72 KB

Final GSoC 2018 submission

Work done

  • Bash scripts and other resources that help to generate RDF graphs to load in Virtuoso
  • Refinement of code from previous Mihindukulasooriya's work et al. to generate pairs of annotation from DBpedia mappings and SPARQL queries:
    • The code has been tested and updated, to ensure it still works.
  • Creation of a Java API that allows storing annotations, users can vote them and can be classified as correct or wrong by using Random Forest's Weka classifier.
  • Creation of a Web application that consumes the API and exposes an easy interface for users to help in the annotation and mapping process.
  • A weekly blog can be consulted. It contains the work done week by week.
  • Some Docker images and Docker-compose files have been created, to help in the deployment process.

Work to be done

This is a finished product, as it can be used for the purpose it was conceived from the beginning. Nevertheless, every work always can be improved.

In this case, the part that has more room to be improved is the web application. The time for the Google Summer of code is limited, so it is not possible to perform a broad study of user interfaces and possible alternatives.

In the same way, it could be a good idea to make an in-depth study of AI techniques and finding which would be a better alternative to Random Forest classifier, if any. The classification result might be improved by using a better classifier, maybe using a Neural Network classifier or similar.

Delivered products

In this section, you can access the links to each product. Each product itself should contain a more detailed explanation of how it works.

License

Unless other licenses are specified on each document, all the source code and derived work are released under Apache 2.0 License. This is a free software license approved by Free Software Foundation.