2026-09-22T09:18:18Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/120357 2023-09-05 col_2142_5131 col_2142_10761 com_2142_5130 com_2142_10755 com_2142_234

Han, Jiawei 2023-05 application/pdf en text Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2025-05-01 The student, Tanay Komarlu, accepted the attached license on 2023-04-11 at 14:14. The student, Tanay Komarlu, submitted this Thesis for approval on 2023-04-11 at 14:21. This Thesis was approved for publication on 2023-04-12 at 16:35. DSpace SAF Submission Ingestion Package generated from Vireo submission #18953 on 2023-09-01 at 17:13:17 Ontology-guided annotation-free fine-grained entity typing Komarlu, Tanay Manjunath 2023-04-12 Fine-grained Entity Typing Annotation-free Entity Typing Zero-shot Entity Typing Information Extraction Fine-grained entity typing (FET), which assigns entities in text with context-sensitive, fine-grained semantic types, will play an important role in natural language understanding. FET has been studied extensively in natural language processing and typically relies on human-annotated corpora for training, which is costly and difficult to scale. Recent studies leverage pre-trained language models (PLMs) to generate rich and context-aware weak supervision for FET. However, the weak supervision based on masked language model (MLM) prompting still has two challenges: (1) a prompt often generates a set of tokens unsuitable for typing, and includes a mixture of rough and fine-grained types or (2) it does not incorporate the rich structural information in a given fine-grained type ontology. In this study, we vision that an ontology provides a semantic-rich, hierarchical structure, which will help select the best results generated by multiple PLM models and head words. Specifically, we propose a novel annotation-free, ontology-guided FET method, OntoType, which follows a type ontological structure, from coarse to fine, ensembles multiple PLM prompting results to generate a set of type candidates, and refines its type resolution, under the local context with a natural language inference model. Our experiments on the Ontonotes, FIGER, and NYT datasets using their associated ontological structures demonstrate that our method outperforms the state-of-the-art zero-shot fine-grained entity typing methods. Our error analysis shows that refinement of the existing ontology structures will further improve fine-grained entity typing. Thesis eng https://hdl.handle.net/2142/120357 Copyright 2023 Tanay Komarlu M.S. Thesis Computer Science University of Illinois at Urbana-Champaign Computer Science