@ARTICLE{Hayes16-REEN-AnalystEffort,
   AUTHOR       = {Jane Huffman Hayes and Alexander Dekhtyar and 
      Jody Larsen and Yann-Ga�l Gu�h�neuc},
   JOURNAL      = {Requirements Engineering (REEN)},
   TITLE        = {Effective Use of Analysts' Effort in Automated Tracing},
   YEAR         = {2018},
   MONTH        = {March},
   NOTE         = {26 pages.},
   NUMBER       = {1},
   PAGES        = {119--143},
   VOLUME       = {23},
   EDITOR       = {Pericles Loucopoulos},
   KEYWORDS     = {Topic: <b>Requirements and features</b>, 
      Venue: <b>REEN</b>},
   PUBLISHER    = {Springer},
   URL          = {http://www.ptidej.net/publications/documents/REEN16.doc.pdf},
   ABSTRACT     = {Because of the large amount of effort it takes to 
      manually trace requirements, automated traceability methods for 
      mapping textual software engineering artifacts to each other and 
      generating candidate links have received increased attention over the 
      past 15 years. Automatically generated links, however, are viewed as 
      candidates until human analysts confirm/reject them for the final 
      requirements traceability matrix. Studies have shown that analysts 
      are a fallible, but necessary, participant in the tracing process. 
      There are two key measures guiding analyst work on the evaluation of 
      candidate links: accuracy of analyst decision and efficiency of their 
      work. Intuitively, it is expected that the more effort the analyst 
      spends on candidate link validation, the more accurate the final 
      traceability matrix is likely to be, although the exact nature of 
      this relationship may be difficult to gauge outright. To assist 
      analysts in making the best use of their time when reviewing 
      candidate links, prior work simulated four possible behaviors and 
      showed that more structured approaches save the analysts' time/effort 
      required to achieve certain levels of accuracy. However, these 
      behavioral simulations are complex to run and their results difficult 
      to interpret and use in practice. In this paper, we present a 
      mathematical model for evaluating analyst effort during requirements 
      tracing tasks. We apply this model to a simulation study of 12 
      candidate link validation approaches. The simulation study is 
      conducted on a number of different datasets. In each study, we assume 
      perfect analyst behavior (i.e., analyst always being correct when 
      making a determination about a link). Under this assumption, we 
      evaluate the estimated effort for the analyst and plot it against the 
      accuracy of the recovered traceability matrix. The effort estimation 
      model is guided by a parameter specifying the relationship between 
      the time it takes an analyst to evaluate a presented link and the 
      time it takes an analyst to discover a link not presented to her. We 
      construct a series of effort estimations based on different values of 
      the model parameter. We found that the analysts' approach to 
      candidate link validation---essentially the order in which the 
      analyst examines presented candidate links---does impact the effort. 
      We also found that the lowest ratio of the cost of finding a correct 
      link from scratch over the cost of recognizing a correct link yields 
      the lowest effort for all datasets, but that the lowest effort does 
      not always yield the highest quality matrix. We finally observed that 
      effort varies by dataset. We conclude that the link evaluation 
      approach we call ``Top 1 Not Yet Examined Feedback Pruning'' was the 
      overall winner in terms of effort and highest quality and, thus, 
      should be followed by human analysts if possible.}
}

{"_id":"rff8psKdeBwPDkJdj","bibbaseid":"hayes-dekhtyar-larsen-guhneuc-effectiveuseofanalystseffortinautomatedtracing-2018","downloads":0,"creationDate":"2018-10-22T00:13:29.334Z","title":"Effective Use of Analysts' Effort in Automated Tracing","author_short":["Hayes, J. H.","Dekhtyar, A.","Larsen, J.","Gu�h�neuc, Y."],"year":2018,"bibtype":"article","biburl":"http://www.yann-gael.gueheneuc.net/Work/Publications/Biblio/complete-bibliography.bib","bibdata":{"bibtype":"article","type":"article","author":[{"firstnames":["Jane","Huffman"],"propositions":[],"lastnames":["Hayes"],"suffixes":[]},{"firstnames":["Alexander"],"propositions":[],"lastnames":["Dekhtyar"],"suffixes":[]},{"firstnames":["Jody"],"propositions":[],"lastnames":["Larsen"],"suffixes":[]},{"firstnames":["Yann-Ga�l"],"propositions":[],"lastnames":["Gu�h�neuc"],"suffixes":[]}],"journal":"Requirements Engineering (REEN)","title":"Effective Use of Analysts' Effort in Automated Tracing","year":"2018","month":"March","note":"26 pages.","number":"1","pages":"119–143","volume":"23","editor":[{"firstnames":["Pericles"],"propositions":[],"lastnames":["Loucopoulos"],"suffixes":[]}],"keywords":"Topic: <b>Requirements and features</b>, Venue: <b>REEN</b>","publisher":"Springer","url":"http://www.ptidej.net/publications/documents/REEN16.doc.pdf","abstract":"Because of the large amount of effort it takes to manually trace requirements, automated traceability methods for mapping textual software engineering artifacts to each other and generating candidate links have received increased attention over the past 15 years. Automatically generated links, however, are viewed as candidates until human analysts confirm/reject them for the final requirements traceability matrix. Studies have shown that analysts are a fallible, but necessary, participant in the tracing process. There are two key measures guiding analyst work on the evaluation of candidate links: accuracy of analyst decision and efficiency of their work. Intuitively, it is expected that the more effort the analyst spends on candidate link validation, the more accurate the final traceability matrix is likely to be, although the exact nature of this relationship may be difficult to gauge outright. To assist analysts in making the best use of their time when reviewing candidate links, prior work simulated four possible behaviors and showed that more structured approaches save the analysts' time/effort required to achieve certain levels of accuracy. However, these behavioral simulations are complex to run and their results difficult to interpret and use in practice. In this paper, we present a mathematical model for evaluating analyst effort during requirements tracing tasks. We apply this model to a simulation study of 12 candidate link validation approaches. The simulation study is conducted on a number of different datasets. In each study, we assume perfect analyst behavior (i.e., analyst always being correct when making a determination about a link). Under this assumption, we evaluate the estimated effort for the analyst and plot it against the accuracy of the recovered traceability matrix. The effort estimation model is guided by a parameter specifying the relationship between the time it takes an analyst to evaluate a presented link and the time it takes an analyst to discover a link not presented to her. We construct a series of effort estimations based on different values of the model parameter. We found that the analysts' approach to candidate link validation—essentially the order in which the analyst examines presented candidate links—does impact the effort. We also found that the lowest ratio of the cost of finding a correct link from scratch over the cost of recognizing a correct link yields the lowest effort for all datasets, but that the lowest effort does not always yield the highest quality matrix. We finally observed that effort varies by dataset. We conclude that the link evaluation approach we call ``Top 1 Not Yet Examined Feedback Pruning'' was the overall winner in terms of effort and highest quality and, thus, should be followed by human analysts if possible.","bibtex":"@ARTICLE{Hayes16-REEN-AnalystEffort,\r\n AUTHOR = {Jane Huffman Hayes and Alexander Dekhtyar and \r\n Jody Larsen and Yann-Ga�l Gu�h�neuc},\r\n JOURNAL = {Requirements Engineering (REEN)},\r\n TITLE = {Effective Use of Analysts' Effort in Automated Tracing},\r\n YEAR = {2018},\r\n MONTH = {March},\r\n NOTE = {26 pages.},\r\n NUMBER = {1},\r\n PAGES = {119--143},\r\n VOLUME = {23},\r\n EDITOR = {Pericles Loucopoulos},\r\n KEYWORDS = {Topic: <b>Requirements and features</b>, \r\n Venue: <b>REEN</b>},\r\n PUBLISHER = {Springer},\r\n URL = {http://www.ptidej.net/publications/documents/REEN16.doc.pdf},\r\n ABSTRACT = {Because of the large amount of effort it takes to \r\n manually trace requirements, automated traceability methods for \r\n mapping textual software engineering artifacts to each other and \r\n generating candidate links have received increased attention over the \r\n past 15 years. Automatically generated links, however, are viewed as \r\n candidates until human analysts confirm/reject them for the final \r\n requirements traceability matrix. Studies have shown that analysts \r\n are a fallible, but necessary, participant in the tracing process. \r\n There are two key measures guiding analyst work on the evaluation of \r\n candidate links: accuracy of analyst decision and efficiency of their \r\n work. Intuitively, it is expected that the more effort the analyst \r\n spends on candidate link validation, the more accurate the final \r\n traceability matrix is likely to be, although the exact nature of \r\n this relationship may be difficult to gauge outright. To assist \r\n analysts in making the best use of their time when reviewing \r\n candidate links, prior work simulated four possible behaviors and \r\n showed that more structured approaches save the analysts' time/effort \r\n required to achieve certain levels of accuracy. However, these \r\n behavioral simulations are complex to run and their results difficult \r\n to interpret and use in practice. In this paper, we present a \r\n mathematical model for evaluating analyst effort during requirements \r\n tracing tasks. We apply this model to a simulation study of 12 \r\n candidate link validation approaches. The simulation study is \r\n conducted on a number of different datasets. In each study, we assume \r\n perfect analyst behavior (i.e., analyst always being correct when \r\n making a determination about a link). Under this assumption, we \r\n evaluate the estimated effort for the analyst and plot it against the \r\n accuracy of the recovered traceability matrix. The effort estimation \r\n model is guided by a parameter specifying the relationship between \r\n the time it takes an analyst to evaluate a presented link and the \r\n time it takes an analyst to discover a link not presented to her. We \r\n construct a series of effort estimations based on different values of \r\n the model parameter. We found that the analysts' approach to \r\n candidate link validation---essentially the order in which the \r\n analyst examines presented candidate links---does impact the effort. \r\n We also found that the lowest ratio of the cost of finding a correct \r\n link from scratch over the cost of recognizing a correct link yields \r\n the lowest effort for all datasets, but that the lowest effort does \r\n not always yield the highest quality matrix. We finally observed that \r\n effort varies by dataset. We conclude that the link evaluation \r\n approach we call ``Top 1 Not Yet Examined Feedback Pruning'' was the \r\n overall winner in terms of effort and highest quality and, thus, \r\n should be followed by human analysts if possible.}\r\n}\r\n\r\n","author_short":["Hayes, J. H.","Dekhtyar, A.","Larsen, J.","Gu�h�neuc, Y."],"editor_short":["Loucopoulos, P."],"key":"Hayes16-REEN-AnalystEffort","id":"Hayes16-REEN-AnalystEffort","bibbaseid":"hayes-dekhtyar-larsen-guhneuc-effectiveuseofanalystseffortinautomatedtracing-2018","role":"author","urls":{"Paper":"http://www.ptidej.net/publications/documents/REEN16.doc.pdf"},"keyword":["Topic: <b>Requirements and features</b>","Venue: <b>REEN</b>"],"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"}}},"search_terms":["effective","use","analysts","effort","automated","tracing","hayes","dekhtyar","larsen","gu�h�neuc"],"keywords":["topic: <b>requirements and features</b>","venue: <b>reen</b>"],"authorIDs":["5a5fb236a39f2c3645000032","5e60e7f0839e59df010000e8","AfJhKcg96muyPdu7S","ahGA65oGDChNYp7Mb"],"dataSources":["8vn5MSGYWB4fAx9Z4"]}