Large-scale language models (LLMs) dominate tasks such as natural language processing and computer vision, but using their spatial and temporal predictive power still needs to be improved. The main focus of the task described in this paper is to offer a technical solution that implements the Time Yards Model (TYM), which means putting in evidence time events and their evolution in a novel book or large text in an unsupervised manner. This article has a ground foundation and uses a manually annotated novel to train multiple state-of-the-art machine learning techniques to divide time segments. The aim is to identify specific text chunks based on several features (part-of-speech and part-of-sentence) to branch off the temporal tracks. This manuscript uses TimeML specification language to mark relevant words in time and space to achieve tempo-spatial investigation. Based on the observations, the proposed algorithm extensively analyses various novels presented for comparison, underlining their main strengths and weaknesses. Comprehensive experiments on diversified datasets show that our proposed solution successfully unlocks the potential for spatial and temporal forecasting with reasonable accuracy. Remarkably, our approach has achieved competitive performance compared to relevant papers from this field. Various text-processing methods are described, and different natural language processing frameworks and libraries are compared.
| Published in | International Journal of Information and Communication Sciences (Volume 9, Issue 2) |
| DOI | 10.11648/j.ijics.20240902.11 |
| Page(s) | 24-32 |
| Creative Commons |
This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited. |
| Copyright |
Copyright © The Author(s), 2024. Published by Science Publishing Group |
Formalism, Time, Temporality, Named Entity Recognition, Geoparsing, Recurrent Neural Network, Natural Text Processing
| [1] | Pustejovsky, James & Ingria, Robert & Saurí, Roser & Castaño, José & Moszkowicz, Jessica & Katz, Graham. (2005): The Specification Language TimeML, TimeML Project. |
| [2] | James F. Allen, George Ferguson (1994): Actions and Events in Interval Temporal Logic, Journal of Logic and Computation, Volume 4, Issue 5, October 1994, pp. 531–579, |
| [3] | Cristea, Dan and Macovei, Andreea (2017): Why Time Resembles Rail Yards? A Way to Look at Time in Textbooks, in Romanian Journal of Information Science and Technology, Vol. 19, 1-2, 2016, pp. 31-43. |
| [4] | Macovei, Andreea (2021): Deciphering temporal and semantic relations in texts, Ph.D. thesis, “Alexandru Ioan Cuza” University of Iași, Faculty of Computer Science. |
| [5] | Boghiu, Șerban (2020): Spatio-Temporal and Semantic Information Extraction, internal Ph.D. report, “Alexandru Ioan Cuza” University of Iași, Faculty of Computer Science. |
| [6] | Cunningham, Hamish & Maynard, Diana & Bontcheva, Kalina & Tablan, Valentin (2002): GATE: A Framework and Graphical Development Environment for Robust NLP Tools and Applications. Proceedings of the 40th Anniversary Meeting of the Association for Computational Linguistics (ACL'02). |
| [7] | Tash Aw (2009): Map of the Invisible World, https://booksvooks.com/map-of-the-invisible-world-pdf-tash-aw.html |
| [8] | Amit, Kumar Sharma & Sandeep, Chaurasia & Srivastava, Devesh Kumar (2020): Sentimental Short Sentences Classification by Using CNN Deep Learning Model with Fine Tuned Word2Vec, Procedia Computer Science, Volume 167, 2020, Pages 1139-1147, ISSN 1877-0509. |
| [9] | Kozhevnikov, Vadim & Pankratova, Evgeniya (2020): Research of the Text Data Vectorization and Classification Algorithms of Machine Learning. Theoretical & Applied Science. |
| [10] | Mingyong Liu, Jiangang Yang (2012): International Conference on Computer Technology and Science (ICCTS 2012). |
| [11] | Ying Xie, Linh Le, Yiyun Zhou, Vijay V. Raghavan (2018): Deep Learning for Natural Language Processing, Handbook of Statistics, Elsevier, Volume 38, 2018, Pages 317-328, ISSN 0169-7161, ISBN 9780444640420. |
APA Style
Paul, B. Ș. (2024). Practical Results for General Models of Temporal, Spatial and Semantic Text Processing. International Journal of Information and Communication Sciences, 9(2), 24-32. https://doi.org/10.11648/j.ijics.20240902.11
ACS Style
Paul, B. Ș. Practical Results for General Models of Temporal, Spatial and Semantic Text Processing. Int. J. Inf. Commun. Sci. 2024, 9(2), 24-32. doi: 10.11648/j.ijics.20240902.11
Share This Article
Twitter
Linked In
Facebook
Copy
Download@article{10.11648/j.ijics.20240902.11,
author = {Boghiu Șerban Paul},
title = {Practical Results for General Models of Temporal, Spatial and Semantic Text Processing
},
journal = {International Journal of Information and Communication Sciences},
volume = {9},
number = {2},
pages = {24-32},
doi = {10.11648/j.ijics.20240902.11},
url = {https://doi.org/10.11648/j.ijics.20240902.11},
eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijics.20240902.11},
abstract = {Large-scale language models (LLMs) dominate tasks such as natural language processing and computer vision, but using their spatial and temporal predictive power still needs to be improved. The main focus of the task described in this paper is to offer a technical solution that implements the Time Yards Model (TYM), which means putting in evidence time events and their evolution in a novel book or large text in an unsupervised manner. This article has a ground foundation and uses a manually annotated novel to train multiple state-of-the-art machine learning techniques to divide time segments. The aim is to identify specific text chunks based on several features (part-of-speech and part-of-sentence) to branch off the temporal tracks. This manuscript uses TimeML specification language to mark relevant words in time and space to achieve tempo-spatial investigation. Based on the observations, the proposed algorithm extensively analyses various novels presented for comparison, underlining their main strengths and weaknesses. Comprehensive experiments on diversified datasets show that our proposed solution successfully unlocks the potential for spatial and temporal forecasting with reasonable accuracy. Remarkably, our approach has achieved competitive performance compared to relevant papers from this field. Various text-processing methods are described, and different natural language processing frameworks and libraries are compared.
},
year = {2024}
}
TY - JOUR T1 - Practical Results for General Models of Temporal, Spatial and Semantic Text Processing AU - Boghiu Șerban Paul Y1 - 2024/12/25 PY - 2024 N1 - https://doi.org/10.11648/j.ijics.20240902.11 DO - 10.11648/j.ijics.20240902.11 T2 - International Journal of Information and Communication Sciences JF - International Journal of Information and Communication Sciences JO - International Journal of Information and Communication Sciences SP - 24 EP - 32 PB - Science Publishing Group SN - 2575-1719 UR - https://doi.org/10.11648/j.ijics.20240902.11 AB - Large-scale language models (LLMs) dominate tasks such as natural language processing and computer vision, but using their spatial and temporal predictive power still needs to be improved. The main focus of the task described in this paper is to offer a technical solution that implements the Time Yards Model (TYM), which means putting in evidence time events and their evolution in a novel book or large text in an unsupervised manner. This article has a ground foundation and uses a manually annotated novel to train multiple state-of-the-art machine learning techniques to divide time segments. The aim is to identify specific text chunks based on several features (part-of-speech and part-of-sentence) to branch off the temporal tracks. This manuscript uses TimeML specification language to mark relevant words in time and space to achieve tempo-spatial investigation. Based on the observations, the proposed algorithm extensively analyses various novels presented for comparison, underlining their main strengths and weaknesses. Comprehensive experiments on diversified datasets show that our proposed solution successfully unlocks the potential for spatial and temporal forecasting with reasonable accuracy. Remarkably, our approach has achieved competitive performance compared to relevant papers from this field. Various text-processing methods are described, and different natural language processing frameworks and libraries are compared. VL - 9 IS - 2 ER -
Faculty of Computer Science, “Alexandru Ioan Cuza” the University of Iași, Iași, Romania
