國立虎尾科技大學 |

A Probabilistic-Based Approach for Expanding Abbreviations in Source Code.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	A Probabilistic-Based Approach for Expanding Abbreviations in Source Code./
Author:	Alatawi, Abdulrahman M.
Description:	1 online resource (143 pages)
Notes:	Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.
Contained By:	Dissertation Abstracts International79-09B(E).
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9780355934397

A Probabilistic-Based Approach for Expanding Abbreviations in Source Code.
Alatawi, Abdulrahman M.

A Probabilistic-Based Approach for Expanding Abbreviations in Source Code. - 1 online resource (143 pages)

Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.

Thesis (D.S.)--Bowie State University, 2018.

Includes bibliographical references

Software developers often use abbreviations for naming identifiers in source code. However, when the source code is maintained by other developers, they sometimes cannot easily expand abbreviations to their correct expansion. The difficulty to retrieve the correct expansions of abbreviations during the maintenance phase makes the source code more problematic to comprehend and increases the likelihood of introducing code defects. There are exist many approaches that automatically expand abbreviations to their original words, unfortunately, they are based on predefined patterns and single words dictionaries providing low rates of expansion accuracy. In this study, we describe a probabilistic-based approach to automatically expand abbreviations to their original expansions to enhance source code comprehension. Our approach automatically extracts a list of potential candidates from the source code for a given abbreviation and uses a probabilistic model based on Bayesian inference to estimate the likelihood of a candidate being the correct expansion. The probabilistic model relies on the unigram and bigram language models and utilizes statistical abbreviations' patterns as evidence to estimate the likelihood of a candidate being the correct expansion. We evaluated our unigram-based approach on a set of 638 abbreviations randomly picked from eight open source projects and found that our approach correctly expands 70.69% of the set. Our unigram-based approach provides an improvement of 22.06% in abbreviation expansion accuracy in comparison to the state-of-the-art approaches. In addition, we evaluated our bigram-based approach on a set of 100 abbreviated phrases, and found that it correctly retrieved 78% of the set. The bigram-based is 9% more accurate than the unigram-based approach and much better than other existing approaches.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355934397Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

A Probabilistic-Based Approach for Expanding Abbreviations in Source Code.
LDR:03060ntm a2200325Ki 4500 001 920918
005 20181227095853.5
006 m o u
007 cr mn||||a|a||
008 190606s2018 xx obm 000 0 eng d
020 $a 9780355934397
035 $a (MiAaPQ)AAI10816390
035 $a (MiAaPQ)bowiestate:10113
035 $a AAI10816390
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Alatawi, Abdulrahman M. $3 1195858
245 1 2 $a A Probabilistic-Based Approach for Expanding Abbreviations in Source Code.
264 0 $c 2018
300 $a 1 online resource (143 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.
500 $a Adviser: Weifeng Xu.
502 $a Thesis (D.S.)--Bowie State University, 2018.
504 $a Includes bibliographical references
520 $a Software developers often use abbreviations for naming identifiers in source code. However, when the source code is maintained by other developers, they sometimes cannot easily expand abbreviations to their correct expansion. The difficulty to retrieve the correct expansions of abbreviations during the maintenance phase makes the source code more problematic to comprehend and increases the likelihood of introducing code defects. There are exist many approaches that automatically expand abbreviations to their original words, unfortunately, they are based on predefined patterns and single words dictionaries providing low rates of expansion accuracy. In this study, we describe a probabilistic-based approach to automatically expand abbreviations to their original expansions to enhance source code comprehension. Our approach automatically extracts a list of potential candidates from the source code for a given abbreviation and uses a probabilistic model based on Bayesian inference to estimate the likelihood of a candidate being the correct expansion. The probabilistic model relies on the unigram and bigram language models and utilizes statistical abbreviations' patterns as evidence to estimate the likelihood of a candidate being the correct expansion. We evaluated our unigram-based approach on a set of 638 abbreviations randomly picked from eight open source projects and found that our approach correctly expands 70.69% of the set. Our unigram-based approach provides an improvement of 22.06% in abbreviation expansion accuracy in comparison to the state-of-the-art approaches. In addition, we evaluated our bigram-based approach on a set of 100 abbreviated phrases, and found that it correctly retrieved 78% of the set. The bigram-based is 9% more accurate than the unigram-based approach and much better than other existing approaches.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Bowie State University. $b Computer Science. $3 1195859
773 0 $t Dissertation Abstracts International $g 79-09B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10816390 $z click for full text (PQDT)