Investigating the Relation between Lexical Smells and Change- and Fault-proneness: An Empirical Study

Investigating the Relation between Lexical Smells and Change- and Fault-proneness: An Empirical Study. Guerrouj, L., Kermansaravi, Z., Arnaoudova, V., Fung, B. C. M., Khomh, F., Antoniol, G., & Gu�h�neuc, Y. Software Quality Journal (SQJ), 25(3):641–670, Springer, September, 2017. 30 pages.

Paper abstract bibtex

Past and recent studies have shown that design smells which are poor solutions to recurrent design problems make object-oriented systems difficult to maintain, and that they negatively impact the class change- and fault-proneness. More recently, lexical smells have been introduced to capture recurring poor practices in the naming, documentation, and choice of identifiers during the implementation of an entity. Although recent studies show that developers perceive lexical smells as impairing program understanding, no study has actually evaluated the relationship between lexical smells and software quality as well as their interaction with design smells. In this paper, we detect 29 smells consisting of 13 design smells and 16 lexical smells in 30 releases of three projects: ANT, ArgoUML, and Hibernate. We analyze to what extent classes containing lexical smells have higher (or lower) odds to change or to be subject to fault fixing than other classes containing design smells. Our results show and bring empirical evidence on the fact that lexical smells can make, in some cases, classes with design smells more fault-prone. In addition, we empirically demonstrate that classes containing design smells only are more change- and fault-prone than classes with lexical smells only.

@ARTICLE{Guerrouj17-SQJ-LexicalSmells,
   AUTHOR       = {Latifa Guerrouj and Zeinab Kermansaravi and 
      Venera Arnaoudova and Benjamin C. M. Fung and Foutse Khomh and 
      Giuliano Antoniol and Yann-Ga�l Gu�h�neuc},
   JOURNAL      = {Software Quality Journal (SQJ)},
   TITLE        = {Investigating the Relation between Lexical Smells and 
      Change- and Fault-proneness: An Empirical Study},
   YEAR         = {2017},
   MONTH        = {September},
   NOTE         = {30 pages.},
   NUMBER       = {3},
   PAGES        = {641--670},
   VOLUME       = {25},
   EDITOR       = {Rachel Harrison},
   KEYWORDS     = {Topic: <b>Code and design smells</b>, Venue: <b>SQJ</b>},
   PUBLISHER    = {Springer},
   URL          = {http://www.ptidej.net/publications/documents/SQJ17b.doc.pdf},
   ABSTRACT     = {Past and recent studies have shown that design smells 
      which are poor solutions to recurrent design problems make 
      object-oriented systems difficult to maintain, and that they 
      negatively impact the class change- and fault-proneness. More 
      recently, lexical smells have been introduced to capture recurring 
      poor practices in the naming, documentation, and choice of 
      identifiers during the implementation of an entity. Although recent 
      studies show that developers perceive lexical smells as impairing 
      program understanding, no study has actually evaluated the 
      relationship between lexical smells and software quality as well as 
      their interaction with design smells. In this paper, we detect 29 
      smells consisting of 13 design smells and 16 lexical smells in 30 
      releases of three projects: ANT, ArgoUML, and Hibernate. We analyze 
      to what extent classes containing lexical smells have higher (or 
      lower) odds to change or to be subject to fault fixing than other 
      classes containing design smells. Our results show and bring 
      empirical evidence on the fact that lexical smells can make, in some 
      cases, classes with design smells more fault-prone. In addition, we 
      empirically demonstrate that classes containing design smells only 
      are more change- and fault-prone than classes with lexical smells 
      only.}
}

Downloads: 0

{"_id":"WenwLWG7vfmMLxyMy","bibbaseid":"guerrouj-kermansaravi-arnaoudova-fung-khomh-antoniol-guhneuc-investigatingtherelationbetweenlexicalsmellsandchangeandfaultpronenessanempiricalstudy-2017","downloads":0,"creationDate":"2018-01-17T20:29:42.149Z","title":"Investigating the Relation between Lexical Smells and Change- and Fault-proneness: An Empirical Study","author_short":["Guerrouj, L.","Kermansaravi, Z.","Arnaoudova, V.","Fung, B. C. M.","Khomh, F.","Antoniol, G.","Gu�h�neuc, Y."],"year":2017,"bibtype":"article","biburl":"http://www.yann-gael.gueheneuc.net/Work/Publications/Biblio/complete-bibliography.bib","bibdata":{"bibtype":"article","type":"article","author":[{"firstnames":["Latifa"],"propositions":[],"lastnames":["Guerrouj"],"suffixes":[]},{"firstnames":["Zeinab"],"propositions":[],"lastnames":["Kermansaravi"],"suffixes":[]},{"firstnames":["Venera"],"propositions":[],"lastnames":["Arnaoudova"],"suffixes":[]},{"firstnames":["Benjamin","C.","M."],"propositions":[],"lastnames":["Fung"],"suffixes":[]},{"firstnames":["Foutse"],"propositions":[],"lastnames":["Khomh"],"suffixes":[]},{"firstnames":["Giuliano"],"propositions":[],"lastnames":["Antoniol"],"suffixes":[]},{"firstnames":["Yann-Ga�l"],"propositions":[],"lastnames":["Gu�h�neuc"],"suffixes":[]}],"journal":"Software Quality Journal (SQJ)","title":"Investigating the Relation between Lexical Smells and Change- and Fault-proneness: An Empirical Study","year":"2017","month":"September","note":"30 pages.","number":"3","pages":"641–670","volume":"25","editor":[{"firstnames":["Rachel"],"propositions":[],"lastnames":["Harrison"],"suffixes":[]}],"keywords":"Topic: Code and design smells, Venue: SQJ","publisher":"Springer","url":"http://www.ptidej.net/publications/documents/SQJ17b.doc.pdf","abstract":"Past and recent studies have shown that design smells which are poor solutions to recurrent design problems make object-oriented systems difficult to maintain, and that they negatively impact the class change- and fault-proneness. More recently, lexical smells have been introduced to capture recurring poor practices in the naming, documentation, and choice of identifiers during the implementation of an entity. Although recent studies show that developers perceive lexical smells as impairing program understanding, no study has actually evaluated the relationship between lexical smells and software quality as well as their interaction with design smells. In this paper, we detect 29 smells consisting of 13 design smells and 16 lexical smells in 30 releases of three projects: ANT, ArgoUML, and Hibernate. We analyze to what extent classes containing lexical smells have higher (or lower) odds to change or to be subject to fault fixing than other classes containing design smells. Our results show and bring empirical evidence on the fact that lexical smells can make, in some cases, classes with design smells more fault-prone. In addition, we empirically demonstrate that classes containing design smells only are more change- and fault-prone than classes with lexical smells only.","bibtex":"@ARTICLE{Guerrouj17-SQJ-LexicalSmells,\r\n AUTHOR = {Latifa Guerrouj and Zeinab Kermansaravi and \r\n Venera Arnaoudova and Benjamin C. M. Fung and Foutse Khomh and \r\n Giuliano Antoniol and Yann-Ga�l Gu�h�neuc},\r\n JOURNAL = {Software Quality Journal (SQJ)},\r\n TITLE = {Investigating the Relation between Lexical Smells and \r\n Change- and Fault-proneness: An Empirical Study},\r\n YEAR = {2017},\r\n MONTH = {September},\r\n NOTE = {30 pages.},\r\n NUMBER = {3},\r\n PAGES = {641--670},\r\n VOLUME = {25},\r\n EDITOR = {Rachel Harrison},\r\n KEYWORDS = {Topic: Code and design smells, Venue: SQJ},\r\n PUBLISHER = {Springer},\r\n URL = {http://www.ptidej.net/publications/documents/SQJ17b.doc.pdf},\r\n ABSTRACT = {Past and recent studies have shown that design smells \r\n which are poor solutions to recurrent design problems make \r\n object-oriented systems difficult to maintain, and that they \r\n negatively impact the class change- and fault-proneness. More \r\n recently, lexical smells have been introduced to capture recurring \r\n poor practices in the naming, documentation, and choice of \r\n identifiers during the implementation of an entity. Although recent \r\n studies show that developers perceive lexical smells as impairing \r\n program understanding, no study has actually evaluated the \r\n relationship between lexical smells and software quality as well as \r\n their interaction with design smells. In this paper, we detect 29 \r\n smells consisting of 13 design smells and 16 lexical smells in 30 \r\n releases of three projects: ANT, ArgoUML, and Hibernate. We analyze \r\n to what extent classes containing lexical smells have higher (or \r\n lower) odds to change or to be subject to fault fixing than other \r\n classes containing design smells. Our results show and bring \r\n empirical evidence on the fact that lexical smells can make, in some \r\n cases, classes with design smells more fault-prone. In addition, we \r\n empirically demonstrate that classes containing design smells only \r\n are more change- and fault-prone than classes with lexical smells \r\n only.}\r\n}\r\n\r\n","author_short":["Guerrouj, L.","Kermansaravi, Z.","Arnaoudova, V.","Fung, B. C. M.","Khomh, F.","Antoniol, G.","Gu�h�neuc, Y."],"editor_short":["Harrison, R."],"key":"Guerrouj17-SQJ-LexicalSmells","id":"Guerrouj17-SQJ-LexicalSmells","bibbaseid":"guerrouj-kermansaravi-arnaoudova-fung-khomh-antoniol-guhneuc-investigatingtherelationbetweenlexicalsmellsandchangeandfaultpronenessanempiricalstudy-2017","role":"author","urls":{"Paper":"http://www.ptidej.net/publications/documents/SQJ17b.doc.pdf"},"keyword":["Topic: Code and design smells","Venue: SQJ"],"metadata":{"authorlinks":{"gu�h�neuc, y":"https://bibbase.org/show?bib=http%3A%2F%2Fwww.yann-gael.gueheneuc.net%2FWork%2FPublications%2FBiblio%2Fcomplete-bibliography.bib&msg=embed","guéhéneuc, y":"http://www.yann-gael.gueheneuc.net/"}}},"search_terms":["investigating","relation","between","lexical","smells","change","fault","proneness","empirical","study","guerrouj","kermansaravi","arnaoudova","fung","khomh","antoniol","gu�h�neuc"],"keywords":["topic: code and design smells","venue: sqj"],"authorIDs":["2tFXMaTSHJKEB5ebi","2wY5eBcsYmbPNfmMS","36dm7jaw5EK5Wrr4D","3NxaNKic3nkXi568L","3S5Dkpx7DNefzJrnf","3afmfmoPr4SHa8B5F","3wmHB7JoQbQz2ujun","4YBWWbao6RKgiyGJE","4jZj9tB4SJ8zEEgHk","5CvA2hsaib2bPMaef","5TFJbxqRDGFj2P8Rg","5a5fb236a39f2c3645000032","5a8f17e006df23bc34000020","5cx79LBmaWcihgM4J","5de9a6425b51bcde01000042","5dee1197584fb4df010000fc","5df228a41e4fe9df0100012c","5df617f72b34d0de0100008b","5dfa14782e791dde010000ea","5dfe3d5e68d95dde01000080","5e02525b6ffa15df0100009f","5e0662c07da1d1de0100021a","5e093e8b934cacdf0100008b","5e0a61673eccf6e001000016","5e0b75b7e73cd6de010000f9","5e0d4ca6ae5827df0100007f","5e0ddf08552b25df01000137","5e0e5c41ac7d11df010000a3","5e1268e7a4cabfdf0100002c","5e12c45a70e2c4f201000043","5e157809f1f31adf01000006","5e162ca1df1bb4de01000123","5e185cff809b84f201000091","5e1a6c39b16ec5df0100000f","5e21b27e96aea7de01000084","5e22c57e49e2b4df0100000f","5e23c2aeb93b51de01000030","5e245835079bb2df0100007d","5e24fa3e2e79a1f201000027","5e26252f408641df01000161","5e26bfbd8535cedf0100005c","5e280fd1f860fcde0100006a","5e2a827f881468de01000080","5e2eb321b84405df01000128","5e2ef635e374eede0100001a","5e2fd6a74e91a9df01000010","5e3266bb5633c9de01000068","5e32ab0ee17accde0100012a","5e32bdec466076df010000d9","5e32d603150c84df01000068","5e34fb145978bef2010000a6","5e36bc8e7b975dde0100009a","5e389940030bcadf010001b4","5e39dd9a3687dddf010000a4","5e3ad173f2a00cdf01000206","5e3dcd50d51253de0100003d","5e3e8713666d79df010000a6","5e3ed80986a596de010000b9","5e3fefe1add5fbde01000087","5e409c79d668c6de010000c7","5e41795ed9f47bee01000194","5e41cd5be7c67ade010000eb","5e42ef1ca6f4a6f2010001eb","5e46dcb342fb31df01000113","5e46f12c461d04f201000078","5e478c9e27a0c8de010000ef","5e47fb06385298df010000b2","5e4add1941072bdf01000011","5e4c1c792dc400de0100011a","5e4c6262271596df010001b9","5e4f0360338acfde01000156","5e4f11b0e5389bde0100007e","5e530b976d68b8df010000a5","5e54ad6d929495df0100007c","5e57161b429006de0100005a","5e57839fcef9b7de0100003c","5e580f5a6a456fde0100004f","5e5afa78038583de010000f7","5e5b477174a3e7df010000b7","5e5d370173eb2edf01000038","5e5fca336b32b0f20100011b","5e60e7f0839e59df010000e8","5e6377cfae1c4dde0100011e","5e657007de41b9df0100017a","5e676f0910be53de0100001a","5gPbX6aQJFjpv2Na9","6eE2yRdMDQr2WGXuA","6iHE5tuM7yTfLd2pA","7BPWyvMr5e6bzbk7T","7RFwhpGkpZRsLwnmB","7amRA4ALcR2mksheF","7mkQL8eiftj5bGMzB","8jPjKehCMsj7ncvxN","8peLXfWtCSic5n7oz","95eRgTcabnJwF46f3","9Ba9JxkjQBCeGBZKg","9DjgvzQrx27uxbyJj","9HD56d3k5yrB9H9oq","9RtPuXNyeS3k8LM9J","9diLYpd8cMmjBh54T","9nx6Yv3XREwJDyRms","AfJhKcg96muyPdu7S","BGvchZsjW7Wejj9Cz","BYwdHpGr6xT5vmE5C","Bah6LM7GXdXTy8GGA","BmH2ytt7sXwPHcrse","CqJYxtqe6qBbtd5yz","D4kEZ2JcWCoMvRPy7","DFWW7D6Y7X57n4cbM","DSorPqHDfrFiNM5Ew","DWXisKXaQArvre3QL","DwBm6isMpKSHHkhAd","E88raoktD8ANF92Yu","EAjLox7ycbofcCXce","F8rzFhY9yWA7pBX4j","G3iynDKjz9BHJbrdg","GJw6mQETXADSCZuuk","GWK5669HLqPyYMQ5J","GibAXjj4xXdFT8qWh","HzFZpgGcfabjAp9x6","KJ4eYziy6hanF9kr9","Kcyu7uncEFiYzYP2D","N4zzhqcywSzDDYsdh","NCDg3xE2mPcNAu7LX","NvgbTAz3hZ9SevZvd","QbcDS3wK43sRASvgu","S3b7Bb9wwfpByQgbo","SXJaeFCgBDJ5HAHtj","T5nL8TGrggoLAF8Dj","W9vT8YcCNFEcp9mWQ","WZ5CpBEFNsb2ivfah","XxviSwRxhwgNwsraH","Z2Zs662GpXqKBEAMc","ZKYFgjHGm7PE4Y2kv","a5qpGirN3B5BLKdMh","ahGA65oGDChNYp7Mb","bA7pGCMS9AB2RBo2p","bTQb3TcrbBShtqFPS","cYnqisf4wzBsM7MF5","cjHpaYiWD5eX7btH4","ckrbesqi3pWqfF2nP","dH8EsWHZtCFuQk5bq","dS5kvBMnk3LMQe56w","eXsFRMzE7WfbHbBL4","fmmsBu4m6ayKtuopf","hdXr3PD8cHNWyAdCe","hgZxckC87u2A57teF","juvCjffHJaPQf44im","keQBT2Apb9yaev8AH","myHdF8zARwW5uGmFs","nJLfaznnYgFqWQQrv","onghitNWSvN2FpCaN","osgPwDW2y5KDXRa2i","pAWFMDHu5dNixqPAq","pLvmgrCjMeDYJiJxB","q4azvWakEjp2TQM7S","qBee6Md9YwRKwkeW3","qQky2Csek4mroLn2P","tJz4YBCqAzZAzek5d","tLtjttw8dEqF6YQ4s","uQ6jCrPijzAmZyfXz","vGEaFNt7mm92Z7GXc","vRkMmE65HSFpCk6FW","vsEsf8FR3Fxb6z7fJ","x5ejzvDeXCc89Dukv","xEQyC5shxpYySSJJm","xhwDdvQ7MYxa6keXm","xkviMnkrGBneANvMr","y64rFMcyp7tDsBrJQ","yBYJWSShoKkMG8aPE","yQPghCwQv22kf6dFq","yd5sCxaEiu5vWizTq"],"dataSources":["Sed98LbBeGaXxenrM","8vn5MSGYWB4fAx9Z4"]}