SCAN: An Approach to Label and Relate Execution Trace Segments

SCAN: An Approach to Label and Relate Execution Trace Segments. Medini, S., Arnaoudova, V., Di Penta, M., Antoniol, G., Gu�h�neuc, Y., & Tonella, P. Journal of Software: Evolution and Process (JSEP), 26(11):962–995, Wiley, November, 2014. 33 pages.

Paper abstract bibtex

Program comprehension is a prerequisite to any maintenance and evolution task. In particular, when performing feature location, developers perform program comprehension by abstracting software features and identifying the links between high-level abstractions (features) and program elements. We present Segment Concept AssigNer (SCAN), an approach to support developers in feature location. SCAN uses a search-based approach to split execution traces into cohesive segments. Then, it labels the segments with relevant keywords and, finally, uses formal concept analysis to identify relations among segments. In a first study, we evaluate the performances of SCAN on six Java programs by 31 participants. We report an average precision of 69% and a recall of 63% when comparing the manual and automatic labels and a precision of 63% regarding the relations among segments identified by SCAN. After that, we evaluate the usefulness of SCAN for the purpose of feature location on two Java programs. We provide evidence that SCAN (i) identifies 69% of the gold set methods and (ii) is effective in reducing the quantity of information that developers must process to locate features—reducing the number of methods to understand by an average of 43% compared to the entire execution traces.

@ARTICLE{Medini14-JSEP-SCAN,
   AUTHOR       = {Soumaya Medini and Venera Arnaoudova and 
      Di Penta, Massimiliano and Giulian Antoniol and Yann-Ga�l Gu�h�neuc and 
      Paolo Tonella},
   JOURNAL      = {Journal of Software: Evolution and Process (JSEP)},
   TITLE        = {SCAN: An Approach to Label and Relate Execution Trace 
      Segments},
   YEAR         = {2014},
   MONTH        = {November},
   NOTE         = {33 pages.},
   NUMBER       = {11},
   PAGES        = {962&ndash;995},
   VOLUME       = {26},
   EDITOR       = {Rocco Oliveto and Denys Poshyvanyk},
   KEYWORDS     = {Topic: <b>Requirements and features</b>, 
      Rubrique : <b>besoins et fonctionalit�s</b>, Journal: <b>JSEP</b>},
   PUBLISHER    = {Wiley},
   URL          = {http://www.ptidej.net/publications/documents/JSEP14.doc.pdf},
   ABSTRACT     = {Program comprehension is a prerequisite to any 
      maintenance and evolution task. In particular, when performing 
      feature location, developers perform program comprehension by 
      abstracting software features and identifying the links between 
      high-level abstractions (features) and program elements. We present 
      Segment Concept AssigNer (SCAN), an approach to support developers in 
      feature location. SCAN uses a search-based approach to split 
      execution traces into cohesive segments. Then, it labels the segments 
      with relevant keywords and, finally, uses formal concept analysis to 
      identify relations among segments. In a first study, we evaluate the 
      performances of SCAN on six Java programs by 31 participants. We 
      report an average precision of 69% and a recall of 63% when comparing 
      the manual and automatic labels and a precision of 63% regarding the 
      relations among segments identified by SCAN. After that, we evaluate 
      the usefulness of SCAN for the purpose of feature location on two 
      Java programs. We provide evidence that SCAN (i) identifies 69% of 
      the gold set methods and (ii) is effective in reducing the quantity 
      of information that developers must process to locate 
      features&mdash;reducing the number of methods to understand by an 
      average of 43% compared to the entire execution traces.}
}

Downloads: 0

{"_id":"42c4sbLvRCxv6QYB4","bibbaseid":"medini-arnaoudova-dipenta-antoniol-guhneuc-tonella-scananapproachtolabelandrelateexecutiontracesegments-2014","downloads":0,"creationDate":"2018-01-17T20:29:42.269Z","title":"SCAN: An Approach to Label and Relate Execution Trace Segments","author_short":["Medini, S.","Arnaoudova, V.","Di Penta, M.","Antoniol, G.","Gu�h�neuc, Y.","Tonella, P."],"year":2014,"bibtype":"article","biburl":"http://www.yann-gael.gueheneuc.net/Work/Publications/Biblio/complete-bibliography.bib","bibdata":{"bibtype":"article","type":"article","author":[{"firstnames":["Soumaya"],"propositions":[],"lastnames":["Medini"],"suffixes":[]},{"firstnames":["Venera"],"propositions":[],"lastnames":["Arnaoudova"],"suffixes":[]},{"propositions":[],"lastnames":["Di","Penta"],"firstnames":["Massimiliano"],"suffixes":[]},{"firstnames":["Giulian"],"propositions":[],"lastnames":["Antoniol"],"suffixes":[]},{"firstnames":["Yann-Ga�l"],"propositions":[],"lastnames":["Gu�h�neuc"],"suffixes":[]},{"firstnames":["Paolo"],"propositions":[],"lastnames":["Tonella"],"suffixes":[]}],"journal":"Journal of Software: Evolution and Process (JSEP)","title":"SCAN: An Approach to Label and Relate Execution Trace Segments","year":"2014","month":"November","note":"33 pages.","number":"11","pages":"962–995","volume":"26","editor":[{"firstnames":["Rocco"],"propositions":[],"lastnames":["Oliveto"],"suffixes":[]},{"firstnames":["Denys"],"propositions":[],"lastnames":["Poshyvanyk"],"suffixes":[]}],"keywords":"Topic: Requirements and features, Rubrique : besoins et fonctionalit�s, Journal: JSEP","publisher":"Wiley","url":"http://www.ptidej.net/publications/documents/JSEP14.doc.pdf","abstract":"Program comprehension is a prerequisite to any maintenance and evolution task. In particular, when performing feature location, developers perform program comprehension by abstracting software features and identifying the links between high-level abstractions (features) and program elements. We present Segment Concept AssigNer (SCAN), an approach to support developers in feature location. SCAN uses a search-based approach to split execution traces into cohesive segments. Then, it labels the segments with relevant keywords and, finally, uses formal concept analysis to identify relations among segments. In a first study, we evaluate the performances of SCAN on six Java programs by 31 participants. We report an average precision of 69% and a recall of 63% when comparing the manual and automatic labels and a precision of 63% regarding the relations among segments identified by SCAN. After that, we evaluate the usefulness of SCAN for the purpose of feature location on two Java programs. We provide evidence that SCAN (i) identifies 69% of the gold set methods and (ii) is effective in reducing the quantity of information that developers must process to locate features—reducing the number of methods to understand by an average of 43% compared to the entire execution traces.","bibtex":"@ARTICLE{Medini14-JSEP-SCAN,\r\n AUTHOR = {Soumaya Medini and Venera Arnaoudova and \r\n Di Penta, Massimiliano and Giulian Antoniol and Yann-Ga�l Gu�h�neuc and \r\n Paolo Tonella},\r\n JOURNAL = {Journal of Software: Evolution and Process (JSEP)},\r\n TITLE = {SCAN: An Approach to Label and Relate Execution Trace \r\n Segments},\r\n YEAR = {2014},\r\n MONTH = {November},\r\n NOTE = {33 pages.},\r\n NUMBER = {11},\r\n PAGES = {962–995},\r\n VOLUME = {26},\r\n EDITOR = {Rocco Oliveto and Denys Poshyvanyk},\r\n KEYWORDS = {Topic: Requirements and features, \r\n Rubrique : besoins et fonctionalit�s, Journal: JSEP},\r\n PUBLISHER = {Wiley},\r\n URL = {http://www.ptidej.net/publications/documents/JSEP14.doc.pdf},\r\n ABSTRACT = {Program comprehension is a prerequisite to any \r\n maintenance and evolution task. In particular, when performing \r\n feature location, developers perform program comprehension by \r\n abstracting software features and identifying the links between \r\n high-level abstractions (features) and program elements. We present \r\n Segment Concept AssigNer (SCAN), an approach to support developers in \r\n feature location. SCAN uses a search-based approach to split \r\n execution traces into cohesive segments. Then, it labels the segments \r\n with relevant keywords and, finally, uses formal concept analysis to \r\n identify relations among segments. In a first study, we evaluate the \r\n performances of SCAN on six Java programs by 31 participants. We \r\n report an average precision of 69% and a recall of 63% when comparing \r\n the manual and automatic labels and a precision of 63% regarding the \r\n relations among segments identified by SCAN. After that, we evaluate \r\n the usefulness of SCAN for the purpose of feature location on two \r\n Java programs. We provide evidence that SCAN (i) identifies 69% of \r\n the gold set methods and (ii) is effective in reducing the quantity \r\n of information that developers must process to locate \r\n features—reducing the number of methods to understand by an \r\n average of 43% compared to the entire execution traces.}\r\n}\r\n\r\n","author_short":["Medini, S.","Arnaoudova, V.","Di Penta, M.","Antoniol, G.","Gu�h�neuc, Y.","Tonella, P."],"editor_short":["Oliveto, R.","Poshyvanyk, D."],"key":"Medini14-JSEP-SCAN","id":"Medini14-JSEP-SCAN","bibbaseid":"medini-arnaoudova-dipenta-antoniol-guhneuc-tonella-scananapproachtolabelandrelateexecutiontracesegments-2014","role":"author","urls":{"Paper":"http://www.ptidej.net/publications/documents/JSEP14.doc.pdf"},"keyword":["Topic: Requirements and features","Rubrique : besoins et fonctionalit�s","Journal: JSEP"],"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":"https://bibbase.org/show?bib=http://www.yann-gael.gueheneuc.net/Work/BibBase/guehene%20(automatically%20cleaned).bib"}},"downloads":0},"search_terms":["scan","approach","label","relate","execution","trace","segments","medini","arnaoudova","di penta","antoniol","gu�h�neuc","tonella"],"keywords":["topic: requirements and features","rubrique : besoins et fonctionalit�s","journal: jsep"],"authorIDs":["AfJhKcg96muyPdu7S","xkviMnkrGBneANvMr"],"dataSources":["Sed98LbBeGaXxenrM","8vn5MSGYWB4fAx9Z4"]}