Keyphrase extraction methodology from short abstracts of medical documents

Eslam Amer; Khaled M. Fouad

doi:10.1109/CIBEC.2016.7836091

Keyphrase extraction methodology from short abstracts of medical documents

Eslam Amer^*
, Khaled M. Fouad

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Keyphrases are the keywords that provide the semantic metadata about document that represents the top concepts inside an article. Number of natural language processing (NLP) methodologies; such as text summarization, text mining and information retrieval applications, consider Keyphrase Extraction as a critical step to accomplish its tasks. This paper presents a method to extract keyphrases from abstracts of medical research papers. The proposed method utilizes DBpedia to gain access to terms that may be of interest to the candidate keyphrases extracted from the original document. Experimental results showed that the proposed method outperforms other related methods in terms of higher precision and F-measure values.

Original language	English
Title of host publication	2016 8th Cairo International Biomedical Engineering Conference, CIBEC 2016
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	23-26
Number of pages	4
ISBN (Electronic)	9781509029877
DOIs	https://doi.org/10.1109/CIBEC.2016.7836091
Publication status	Published - 30 Jan 2017
Event	8th Cairo International Biomedical Engineering Conference, CIBEC 2016 - Cairo, Egypt Duration: 15 Dec 2016 → 17 Dec 2016

Conference

Conference	8th Cairo International Biomedical Engineering Conference, CIBEC 2016
Country/Territory	Egypt
City	Cairo
Period	15/12/16 → 17/12/16

Keywords

Feature extraction
data mining
natural language processing
computer science
semanitics
encyclopedias

Access to Document

10.1109/CIBEC.2016.7836091

Cite this

@inproceedings{1597e70aafdb46adbc0b1b70f67907fd,

title = "Keyphrase extraction methodology from short abstracts of medical documents",

abstract = "Keyphrases are the keywords that provide the semantic metadata about document that represents the top concepts inside an article. Number of natural language processing (NLP) methodologies; such as text summarization, text mining and information retrieval applications, consider Keyphrase Extraction as a critical step to accomplish its tasks. This paper presents a method to extract keyphrases from abstracts of medical research papers. The proposed method utilizes DBpedia to gain access to terms that may be of interest to the candidate keyphrases extracted from the original document. Experimental results showed that the proposed method outperforms other related methods in terms of higher precision and F-measure values.",

keywords = "Feature extraction, data mining, natural language processing, computer science, semanitics, encyclopedias",

author = "Eslam Amer and Fouad, \{Khaled M.\}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE. Noissn.; 8th Cairo International Biomedical Engineering Conference, CIBEC 2016 ; Conference date: 15-12-2016 Through 17-12-2016",

year = "2017",

month = jan,

day = "30",

doi = "10.1109/CIBEC.2016.7836091",

language = "English",

pages = "23--26",

booktitle = "2016 8th Cairo International Biomedical Engineering Conference, CIBEC 2016",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

}

Amer, E & Fouad, KM 2017, Keyphrase extraction methodology from short abstracts of medical documents. in 2016 8th Cairo International Biomedical Engineering Conference, CIBEC 2016., 7836091, Institute of Electrical and Electronics Engineers Inc., pp. 23-26, 8th Cairo International Biomedical Engineering Conference, CIBEC 2016, Cairo, Egypt, 15/12/16. https://doi.org/10.1109/CIBEC.2016.7836091

TY - GEN

T1 - Keyphrase extraction methodology from short abstracts of medical documents

AU - Amer, Eslam

AU - Fouad, Khaled M.

PY - 2017/1/30

Y1 - 2017/1/30

N2 - Keyphrases are the keywords that provide the semantic metadata about document that represents the top concepts inside an article. Number of natural language processing (NLP) methodologies; such as text summarization, text mining and information retrieval applications, consider Keyphrase Extraction as a critical step to accomplish its tasks. This paper presents a method to extract keyphrases from abstracts of medical research papers. The proposed method utilizes DBpedia to gain access to terms that may be of interest to the candidate keyphrases extracted from the original document. Experimental results showed that the proposed method outperforms other related methods in terms of higher precision and F-measure values.

AB - Keyphrases are the keywords that provide the semantic metadata about document that represents the top concepts inside an article. Number of natural language processing (NLP) methodologies; such as text summarization, text mining and information retrieval applications, consider Keyphrase Extraction as a critical step to accomplish its tasks. This paper presents a method to extract keyphrases from abstracts of medical research papers. The proposed method utilizes DBpedia to gain access to terms that may be of interest to the candidate keyphrases extracted from the original document. Experimental results showed that the proposed method outperforms other related methods in terms of higher precision and F-measure values.

KW - Feature extraction

KW - data mining

KW - natural language processing

KW - computer science

KW - semanitics

KW - encyclopedias

UR - https://www.scopus.com/pages/publications/85015315103

U2 - 10.1109/CIBEC.2016.7836091

DO - 10.1109/CIBEC.2016.7836091

M3 - Conference contribution

AN - SCOPUS:85015315103

SP - 23

EP - 26

BT - 2016 8th Cairo International Biomedical Engineering Conference, CIBEC 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 8th Cairo International Biomedical Engineering Conference, CIBEC 2016

Y2 - 15 December 2016 through 17 December 2016

ER -

Keyphrase extraction methodology from short abstracts of medical documents

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this