RDF, RDFS代写、代写CS/python语言、代做dbpedia留学生、代写C/C++设计

日期：2018-11-27 10:59

Project: From knowledge extraction to knowledge

representation

There are different tools and techniques that allow to:

Represent knowledge using RDF, RDFS and OWL

Extract knowledge from text using Native Bayes and SVM

In this project, the objective is to build a simple end to end pipeline in order to extract

knowledge from a set of documents, represent this knowledge using rdf, query the created rdf

graph and eventually use reasoning to enrich the extracted knowledge.

Before starting the implementation of your pipeline. You should collect 30 news articles

about different topics.

In order to build the whole pipeline, you will need four subparts in your project :

1. A program that takes as input a document and output the document class

2. A program that takes as input the document, the document ID and the label predicted

by the learned model. This program should output a rdf file.

a. Optional : You can do some topic modelling to extract more knowledge about

the documents

b. Optional : You can also perform Named Entity Recognition using

some pretrained models in NLTK or Spacy

3. A program that contains some generic sparql queries that will be executed on the rdf file.

4. A program that will takes an ontology modeling some class hierarchies about the

labels you have, the rdf file and will perform reasoning in order to enrich the extracted

knowledge

5. Optional : You can run some queries over the dbpedia (or any other Linked Open

Dataset) to get more knowledge about the topics you extract from document.