• Home
  • Search
  • Exploring Earth Science Applications using Word Embeddings
  • https://doi.org/10.5194/egusphere-egu2020-9966Copy DOI Icon

Exploring Earth Science Applications using Word Embeddings

  • Mar 23, 2020
  • Derek Koehl +4 more
Show More
  • Abstract
  • Literature Map
  • Similar Papers
Abstract

<p>Word embedding are numeric representations of text which capture meanings and semantic relationships in text. Embeddings can be constructed using different methods such as One Hot encoding, Frequency-based or Prediction-based approaches. Prediction-based approaches such as  Word2Vec, can be used to generate word embeddings that can capture the underlying semantics and word relationships in a corpus. Word2Vec embeddings generated from domain specific corpus have been shown in studies to both predict relationships and augment word vectors to improve classifications. We describe results from two different experiments utilizing word embeddings for Earth science constructed from a corpus of over 20,000 journal papers using Word2Vec. </p><p>The first experiment explores the analogy prediction performance of word embeddings built from the Earth science journal corpus and trained using domain-specific vocabulary. Our results demonstrate that the accuracy of domain-specific word embeddings in predicting Earth science analogy questions outperforms the ability of general corpus embedding to predict general analogy questions. While the results are as anticipated,  the substantial increase in accuracy, particularly in the lexicographical domain was encouraging. The results point to the need for developing a comprehensive Earth science analogy test set that covers the full breadth of lexicographical and encyclopedic categories for validating word embeddings.</p><p>The second experiment utilizes the word embeddings to augment metadata keyword classifications. Metadata describing NASA datasets have science keywords that are manually assigned which can lead to errors and inconsistencies. These science keywords are controlled vocabulary and are used to aid data discovery via faceted search and relevancy ranking. Given the small size of the number of metadata records with proper description and keywords, word embeddings were used for augmentation. A fully connected neural network was trained to suggest keywords given a description text. This approach provided the best accuracy at ~76% as compared to other methods tested.</p>

Similar Papers
  • Conference Article
  • Citations2

An exploration of semantic relations in neural word embeddings using extrinsic knowledge

  • Nov 01, 2017
  • Zhiwei Chen +3
  • Conference Article

Research of knowledge mapping construction based on word vector

  • May 01, 2018
  • Li Ji +2
  • Research Article

Cword2vec: a novel morphological rule-based word embedding approach for Urdu text sentiment analysis

  • Jul 15, 2025
  • PeerJ Computer Science
  • Saquib Khushhal +5
  • Research Article
  • Citations539

AraVec: A set of Arabic Word Embedding Models for use in Arabic NLP

  • Jan 01, 2017
  • Procedia Computer Science
  • Abu Bakr Soliman +2
  • Conference Article
  • Citations24

Word Embedding Comparison for Indonesian Language Sentiment Analysis

  • Mar 01, 2019
  • Helmi Imaduddin +2
  • Conference Article
  • Citations4

Study on Semantic Transparency of Chinese Compounds Based on Word Embedding

  • Dec 04, 2020
  • Xuemei Tang +1
  • PDF
  • Research Article
  • Citations25

The Accuracy Comparison Between Word2Vec and FastText On Sentiment Analysis of Hotel Reviews

  • Jun 30, 2022
  • Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
  • Siti Khomsah +2
  • Book Chapter
  • Citations50

A Systematic Literature Review on Word Embeddings

  • Sep 27, 2018
  • Luis Gutiérrez +1
  • Research Article
  • Citations153

Enhancing Topic Modeling for Short Texts with Auxiliary Word Embeddings

  • Aug 21, 2017
  • ACM Transactions on Information Systems
  • Chenliang Li +5
  • Research Article
  • Citations14

Context-Aware Dynamic Word Embeddings for Aspect Term Extraction

  • Jan 01, 2024
  • IEEE Transactions on Affective Computing
  • Jingyun Xu +6
  • Research Article
  • Citations19

Analysis of Sentiment on Movie Reviews Using Word Embedding Self-Attentive LSTM

  • Apr 01, 2021
  • International Journal of Ambient Computing and Intelligence
  • Soubraylu Sivakumar +1
  • Research Article
  • Citations2

Code‐mixed Hindi‐English text correction using fuzzy graph and word embedding

  • May 14, 2023
  • Expert Systems
  • Minni Jain +2
  • PDF
  • Conference Article
  • Citations26

Corpus Specificity in LSA and Word2vec: The Role of Out-of-Domain Documents

  • Jan 01, 2018
  • Edgar Altszyler +2
  • Conference Article
  • Citations30

Automatic Anonymization of Textual Documents: Detecting Sensitive Information via Word Embeddings

  • Aug 01, 2019
  • Fadi Hassan +3
  • Research Article
  • Citations4

Assessment of Mutation Susceptibility in DNA Sequences with Word Vectors

  • Mar 20, 2020
  • Journal of Intelligent Systems: Theory and Applications
  • Alper Yilmaz
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.