I want to build this project in as a Python wrapper that enhances the knowledge base embeddings in DBpedia to provide more accurate semantic information by using generative models to provide schemes for link prediction and entity recognition systems. As we aim to get the embeddings of these documents in the DBpedia Knowledge Base, I want to first perform the paper implementation Yuxing Zhang: Recurrent Neural Network Embedding for Knowledge-base Completion. As knowledge base completion is a task of inferring the missing triples, or out-of-vocabulary triples, from existing triples in the knowledge base, this will help build a model to solve this problem.
Goals
With this project, I aim to make the DBPedia corpus of KB embeddings more complete by using this wrapper that will further allow the extension of the knowledge base, add missing links, etc.
Warm up tasks
This project has a direct impact on the DBPedia corpus as the models used in this task will be used to find missing links in the DBpedia Knowledge base. It will also allow a better representation of the embeddings.
Mentors
Keywords
Python, NLP, Machine Learning, Deep Learning, Knowledge Graph, Knowledge Base, Knowledge Base Embeddings
Student Proposal
Description
I want to build this project in as a Python wrapper that enhances the knowledge base embeddings in DBpedia to provide more accurate semantic information by using generative models to provide schemes for link prediction and entity recognition systems. As we aim to get the embeddings of these documents in the DBpedia Knowledge Base, I want to first perform the paper implementation Yuxing Zhang: Recurrent Neural Network Embedding for Knowledge-base Completion. As knowledge base completion is a task of inferring the missing triples, or out-of-vocabulary triples, from existing triples in the knowledge base, this will help build a model to solve this problem.
Goals
With this project, I aim to make the DBPedia corpus of KB embeddings more complete by using this wrapper that will further allow the extension of the knowledge base, add missing links, etc.
Warm up tasks
This project has a direct impact on the DBPedia corpus as the models used in this task will be used to find missing links in the DBpedia Knowledge base. It will also allow a better representation of the embeddings.
Mentors
Keywords
Python, NLP, Machine Learning, Deep Learning, Knowledge Graph, Knowledge Base, Knowledge Base Embeddings