Knowledge Graph Extraction from Unstructured Data and Semantic Role Labeling

Vivek Khetan, AI Researcher, Accenture Labs

ABSTRACT:
Maintaining regulatory compliance is challenging for a range of businesses due to the volume of regulations and the rapid rate of change. Given the sheer volume, it is difficult for an enterprise to maintain a clear picture of the state of regulations that govern them, or to stay abreast of the changes.

In contrast to expert analysis or the development of domain-specific ontology and taxonomies, this talk will discuss how a task-based approach for fulfilling specific information needs within a new domain can be helpful.

This presentation will discuss various techniques for knowledge graph extraction and completion, domain-specific schema creation, custom bi-LSTM-CRF model for entity extractors and attention-based deep Semantic Role Labeling. We will walk through each of these algorithms in detail and their need for a specific use case.

SPEAKER BIO:
Vivek Khetan is an artificial intelligence researcher at Accenture Labs, San Francisco. He is currently focusing on semantic role labelling, entity, and relationship extraction, close, and open-domain knowledge graph creation. He has scholarly work published in the ECIR and Information Retrieval Journal. Currently, he is working on “Common Sense Reasoning” and also organizing the Knowledge Graph for Social Good (KGSK) workshop in collaboration with the United Nations. The KGSG workshop will be part of the Knowledge Graph Conference happening at Columbia University.

In previous roles, he has worked as a dialogue system researcher at SparkCognition. He has experience in the application of diverse machine learning methods, including information retrieval, survival analysis, and anomaly detection.

#KnowledgeGraphExtraction #UnstructuredData #SemanticRoleLabeling