Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

J2N -- Nominal Adjective Identification and its Application

Lemeng Qi, Yang Han, Zhuotong Xie

TL;DR

This study shows that reclassifying NAs can improve the accuracy of syntactic analysis and structural understanding in NLP, and present experimental results using Hidden Markov Models, Maximum Entropy models, and Spacy.

Abstract

This paper explores the challenges posed by nominal adjectives (NAs) in natural language processing (NLP) tasks, particularly in part-of-speech (POS) tagging. We propose treating NAs as a distinct POS tag, "JN," and investigate its impact on POS tagging, BIO chunking, and coreference resolution. Our study shows that reclassifying NAs can improve the accuracy of syntactic analysis and structural understanding in NLP. We present experimental results using Hidden Markov Models (HMMs), Maximum Entropy (MaxEnt) models, and Spacy, demonstrating the feasibility and potential benefits of this approach. Additionally we finetuned a bert model to identify the NA in untagged text.

J2N -- Nominal Adjective Identification and its Application

TL;DR

This study shows that reclassifying NAs can improve the accuracy of syntactic analysis and structural understanding in NLP, and present experimental results using Hidden Markov Models, Maximum Entropy models, and Spacy.

Abstract

This paper explores the challenges posed by nominal adjectives (NAs) in natural language processing (NLP) tasks, particularly in part-of-speech (POS) tagging. We propose treating NAs as a distinct POS tag, "JN," and investigate its impact on POS tagging, BIO chunking, and coreference resolution. Our study shows that reclassifying NAs can improve the accuracy of syntactic analysis and structural understanding in NLP. We present experimental results using Hidden Markov Models (HMMs), Maximum Entropy (MaxEnt) models, and Spacy, demonstrating the feasibility and potential benefits of this approach. Additionally we finetuned a bert model to identify the NA in untagged text.
Paper Structure (38 sections, 4 figures, 2 tables)

This paper contains 38 sections, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Problem Statement
  4. Objectives
  5. Related Work
  6. Distinctive Adjectival Constructions
  7. Computational Approaches to Word Class Conversion
  8. Extension to Nominal Adjectives
  9. Give nominal adjectives JN tags
  10. Reservations on tagging Adjectives as Nouns
  11. Investigation of Nominal Adjectives in the WSJ Corpus
  12. Comparison and Analysis of Screened Nominal Adjectives
  13. Results and Discussion
  14. General Methodology
  15. Dataset Preparation
  16. ...and 23 more sections

Figures (4)

  • Figure 1: substantive adjectives are marked as noun phrases (NP) together with the preceding determiner
  • Figure 2: comparison of previous word pos tag probability of NN&JN, JJ&JN, JJS&JN and the consine similarities of them
  • Figure 3: comparison of next word pos tag probability of NN&JN, JJ&JN, JJS&JN and the consine similarities of them
  • Figure 4: An example sentence containing nominal adjective coref. When "gifted" is labeled JJ by the default POS tagger, the coref chain cannot be recognized.