Publications

Neural Symbolic Models for Overcoming Deep Learning Limitations and Korean Dependency Parsing

Feb 2023 — Master's Thesis

Domestic Conferences & Theses

Authors

Joongmin Shin*

Abstract

Proposed and analyzed a novel neural-symbolic model that controls final probability values based on additional linguistic knowledge to overcome the limitations of overfitting and data scarcity issues arising from deep learning's dependence on datasets.

Key Contribution

Neural-symbolic parser integrating linguistic constraints to overcome deep learning limitations in dependency parsing.