Please note: This master’s thesis presentation will be given online.
Alireza
Vezvaei,
Master’s
candidate
David
R.
Cheriton
School
of
Computer
Science
Supervisor: Professor Lukasz Golab
Knowledge Graphs (KGs) have many industrial and academic applications, specifically in information retrieval and question answering. Prominent community projects are conducted for building large-scale KGs, but building KGs with crowdsourcing is costly. Considering the rapidly growing amount of unstructured text on the Web, systems for automatic KG generation are highly required.
We propose KG-Pipeline, a general-purpose end-to-end pipeline designed for automatically constructing KGs from unstructured text documents. State-of-the-art NLP models are leveraged in implementing various components of the pipeline. Following the algorithm introduced by recent work, we utilize our generated KGs in Question Answering (QA) and evaluate the performance of our system on a QA benchmark, comparing it to previous work and a baseline model.
To join this master’s thesis presentation on Zoom, please go to https://uwaterloo.zoom.us/j/97286628421.