Developing NLP Solutions for Automatic Knowledge Graph Construction - Part 2

Description

Knowledge graphs are increasingly becoming important in the AI world as an enabling technology for data integration and analytics, semantic search and question answering, and other cognitive applications.

However, developing and maintaining large knowledge graphs in a manual way is too expensive and time consuming. To accelerate and scale the process, methods and techniques from the areas of information extraction and natural language processing (NLP) can be very helpful.

In this masterclass we will take a deep dive into the tasks required for mining a knowledge graph from text, and we will implement basic and advanced NLP techniques for each of them, using open-source tools and resources.

Key Topics

• Knowledge Graphs
• Natural Language Processing

Target Audience

• NLP practitioners who want to learn how to apply their craft for constructing knowledge graphs
• Data Scientists and Machine Learning Engineers

Goal

By the end of this masterclass, the attendees will be able to: - Identify the main components of a knowledge graph and the stages involved in its development process. - Identify and formulate the NLP tasks involved in a knowledge graph mining project - Develop and evaluate methods for mining entities and relations using open-source software

Session Outline

Session 3: Mining Relations (60mins)

• Task formulation and approaches
• Special relations: synonyms and hyponyms
• Hands-on: Getting relations from external semantic models
• Hands-on: Mining relations using rules and patterns
• Hands-on: Mining relations using embeddings
• Q&A

Break (5 mins)

Session 4: Evaluation and Wrapping Up (60mins)

• Knowledge graph quality
• Text mining quality
• Sensitivity analysis and adversarial evaluation
• Hands-On: Adversarial evaluation of a relation extraction system
• Mitigation strategies
• Masterclass wrapping up
  • Take-Aways
  • Additional resources
  • Q&A

Format

The masterclass will be delivered in four 1-hour sessions, with one long break in between.

Each session will comprise a slide-based presentation of key topics and techniques, a hands-on demonstration of these techniques with relevant software, and a Q&A.

Level

Intermediate - Advanced

Prerequisite Knowledge

• Basic coding skills in Python and Jupyter notebooks
• Familiarity with RDF/OWL or Labeled Property Graphs (e.g. Neo4J).

You need an access pass to attend this session: Diversity Access Pass or Free Access Pass apply