Term based semantic clusters for very short text classification

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Very short texts, such as tweets and invoices, present challenges in classification. Although term occurrences are strong indicators of content, in very short texts, the sparsity of these texts makes it difficult to capture important semantic relationships. A solution calls for a method that not only considers term occurrence, but also handles sparseness well. In this work, we introduce such an approach, the Term Based Semantic Clusters (TBSeC) that employs terms to create distinctive semantic concept clusters. These clusters are ranked using a semantic similarity function which in turn defines a semantic feature space that can be used for text classification. Our method is evaluated in an invoice classification task. Compared to well-known content representation methods the proposed method performs competitively.
Original languageEnglish
Title of host publicationProceedings of the International Conference Recent Advances in Natural Language Processing 2019
PublisherAssociation for Computational Linguistics (ACL)
Number of pages10
Publication statusAccepted/In press - 2019
EventInternational Conference Recent Advances in Natural Language Processing - Varna, Bulgaria
Duration: 2 Sep 20194 Sep 2019
http://lml.bas.bg/ranlp2019/start.php

Conference

ConferenceInternational Conference Recent Advances in Natural Language Processing
Abbreviated titleRANLP
CountryBulgaria
CityVarna
Period2/09/194/09/19
Internet address

Fingerprint

Semantics

Keywords

  • text classification
  • term extraction
  • character word embeddings
  • invoice classification

Cite this

Paalman, J., Mullick, S., Zervanou, K., & Zhang, Y. (Accepted/In press). Term based semantic clusters for very short text classification. In Proceedings of the International Conference Recent Advances in Natural Language Processing 2019 Association for Computational Linguistics (ACL).
Paalman, Jasper ; Mullick, Shantanu ; Zervanou, Kalliopi ; Zhang, Yingqian. / Term based semantic clusters for very short text classification. Proceedings of the International Conference Recent Advances in Natural Language Processing 2019. Association for Computational Linguistics (ACL), 2019.
@inproceedings{70768a1425ca438b9cf0bfd2d5db726d,
title = "Term based semantic clusters for very short text classification",
abstract = "Very short texts, such as tweets and invoices, present challenges in classification. Although term occurrences are strong indicators of content, in very short texts, the sparsity of these texts makes it difficult to capture important semantic relationships. A solution calls for a method that not only considers term occurrence, but also handles sparseness well. In this work, we introduce such an approach, the Term Based Semantic Clusters (TBSeC) that employs terms to create distinctive semantic concept clusters. These clusters are ranked using a semantic similarity function which in turn defines a semantic feature space that can be used for text classification. Our method is evaluated in an invoice classification task. Compared to well-known content representation methods the proposed method performs competitively.",
keywords = "text classification, term extraction, character word embeddings, invoice classification",
author = "Jasper Paalman and Shantanu Mullick and Kalliopi Zervanou and Yingqian Zhang",
year = "2019",
language = "English",
booktitle = "Proceedings of the International Conference Recent Advances in Natural Language Processing 2019",
publisher = "Association for Computational Linguistics (ACL)",

}

Paalman, J, Mullick, S, Zervanou, K & Zhang, Y 2019, Term based semantic clusters for very short text classification. in Proceedings of the International Conference Recent Advances in Natural Language Processing 2019. Association for Computational Linguistics (ACL), International Conference Recent Advances in Natural Language Processing, Varna, Bulgaria, 2/09/19.

Term based semantic clusters for very short text classification. / Paalman, Jasper; Mullick, Shantanu; Zervanou, Kalliopi; Zhang, Yingqian.

Proceedings of the International Conference Recent Advances in Natural Language Processing 2019. Association for Computational Linguistics (ACL), 2019.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Term based semantic clusters for very short text classification

AU - Paalman, Jasper

AU - Mullick, Shantanu

AU - Zervanou, Kalliopi

AU - Zhang, Yingqian

PY - 2019

Y1 - 2019

N2 - Very short texts, such as tweets and invoices, present challenges in classification. Although term occurrences are strong indicators of content, in very short texts, the sparsity of these texts makes it difficult to capture important semantic relationships. A solution calls for a method that not only considers term occurrence, but also handles sparseness well. In this work, we introduce such an approach, the Term Based Semantic Clusters (TBSeC) that employs terms to create distinctive semantic concept clusters. These clusters are ranked using a semantic similarity function which in turn defines a semantic feature space that can be used for text classification. Our method is evaluated in an invoice classification task. Compared to well-known content representation methods the proposed method performs competitively.

AB - Very short texts, such as tweets and invoices, present challenges in classification. Although term occurrences are strong indicators of content, in very short texts, the sparsity of these texts makes it difficult to capture important semantic relationships. A solution calls for a method that not only considers term occurrence, but also handles sparseness well. In this work, we introduce such an approach, the Term Based Semantic Clusters (TBSeC) that employs terms to create distinctive semantic concept clusters. These clusters are ranked using a semantic similarity function which in turn defines a semantic feature space that can be used for text classification. Our method is evaluated in an invoice classification task. Compared to well-known content representation methods the proposed method performs competitively.

KW - text classification

KW - term extraction

KW - character word embeddings

KW - invoice classification

M3 - Conference contribution

BT - Proceedings of the International Conference Recent Advances in Natural Language Processing 2019

PB - Association for Computational Linguistics (ACL)

ER -

Paalman J, Mullick S, Zervanou K, Zhang Y. Term based semantic clusters for very short text classification. In Proceedings of the International Conference Recent Advances in Natural Language Processing 2019. Association for Computational Linguistics (ACL). 2019