{"id":"https://openalex.org/W6968468993","doi":"https://doi.org/10.5281/zenodo.16789964","title":"Replication Package of the TOSEM Journal Article: \"What Were You Thinking? An LLM-Driven Large-Scale Study of Refactoring Motivations in Open-Source Projects\"","display_name":"Replication Package of the TOSEM Journal Article: \"What Were You Thinking? An LLM-Driven Large-Scale Study of Refactoring Motivations in Open-Source Projects\"","publication_year":2025,"publication_date":"2025-04-14","ids":{"openalex":"https://openalex.org/W6968468993","doi":"https://doi.org/10.5281/zenodo.16789964"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.16789964","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.16789964","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Dataset"},"type":"dataset","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.16789964","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Robredo, Mikel","orcid":"https://orcid.org/0009-0001-9870-1504"},"institutions":[{"id":"https://openalex.org/I98381234","display_name":"University of Oulu","ror":"https://ror.org/03yj89h83","country_code":"FI","type":"education","lineage":["https://openalex.org/I98381234"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Robredo, Mikel","raw_affiliation_strings":["University of Oulu"],"raw_orcid":"https://orcid.org/0009-0001-9870-1504","affiliations":[{"raw_affiliation_string":"University of Oulu","institution_ids":["https://openalex.org/I98381234"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Esposito, Matteo","orcid":"https://orcid.org/0000-0002-8451-3668"},"institutions":[{"id":"https://openalex.org/I116067653","display_name":"University of Rome Tor Vergata","ror":"https://ror.org/02p77k626","country_code":"IT","type":"education","lineage":["https://openalex.org/I116067653"]},{"id":"https://openalex.org/I98381234","display_name":"University of Oulu","ror":"https://ror.org/03yj89h83","country_code":"FI","type":"education","lineage":["https://openalex.org/I98381234"]}],"countries":["FI","IT"],"is_corresponding":false,"raw_author_name":"Esposito, Matteo","raw_affiliation_strings":["University of Oulu","University of Rome Tor Vergata"],"raw_orcid":"https://orcid.org/0000-0002-8451-3668","affiliations":[{"raw_affiliation_string":"University of Oulu","institution_ids":["https://openalex.org/I98381234"]},{"raw_affiliation_string":"University of Rome Tor Vergata","institution_ids":["https://openalex.org/I116067653"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Palomba, Fabio","orcid":"https://orcid.org/0000-0001-9337-5116"},"institutions":[{"id":"https://openalex.org/I131729948","display_name":"University of Salerno","ror":"https://ror.org/0192m2k53","country_code":"IT","type":"education","lineage":["https://openalex.org/I131729948"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Palomba, Fabio","raw_affiliation_strings":["University of Salerno"],"raw_orcid":"https://orcid.org/0000-0001-9337-5116","affiliations":[{"raw_affiliation_string":"University of Salerno","institution_ids":["https://openalex.org/I131729948"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Pe\u00f1aloza, Rafael","orcid":"https://orcid.org/0000-0002-2693-5790"},"institutions":[{"id":"https://openalex.org/I66752286","display_name":"University of Milano-Bicocca","ror":"https://ror.org/01ynf4891","country_code":"IT","type":"education","lineage":["https://openalex.org/I66752286"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Pe\u00f1aloza, Rafael","raw_affiliation_strings":["University of Milano-Bicocca"],"raw_orcid":"https://orcid.org/0000-0002-2693-5790","affiliations":[{"raw_affiliation_string":"University of Milano-Bicocca","institution_ids":["https://openalex.org/I66752286"]}]},{"author_position":"last","author":{"id":null,"display_name":"Lenarduzzi, Valentina","orcid":"https://orcid.org/0000-0003-0511-5133"},"institutions":[{"id":"https://openalex.org/I98381234","display_name":"University of Oulu","ror":"https://ror.org/03yj89h83","country_code":"FI","type":"education","lineage":["https://openalex.org/I98381234"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Lenarduzzi, Valentina","raw_affiliation_strings":["University of Oulu"],"raw_orcid":"https://orcid.org/0000-0003-0511-5133","affiliations":[{"raw_affiliation_string":"University of Oulu","institution_ids":["https://openalex.org/I98381234"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T11675","display_name":"Open Source Software Innovations","score":0.8162000179290771,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11675","display_name":"Open Source Software Innovations","score":0.8162000179290771,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10260","display_name":"Software Engineering Research","score":0.02969999983906746,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11986","display_name":"Scientific Computing and Data Management","score":0.0210999995470047,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.7257999777793884},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.6134999990463257},{"id":"https://openalex.org/keywords/code-refactoring","display_name":"Code refactoring","score":0.6098999977111816},{"id":"https://openalex.org/keywords/feature-model","display_name":"Feature model","score":0.5184000134468079},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.46309998631477356},{"id":"https://openalex.org/keywords/python","display_name":"Python (programming language)","score":0.4618000090122223},{"id":"https://openalex.org/keywords/process-mining","display_name":"Process mining","score":0.34619998931884766}],"concepts":[{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.7257999777793884},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.663100004196167},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.6134999990463257},{"id":"https://openalex.org/C152752567","wikidata":"https://www.wikidata.org/wiki/Q116877","display_name":"Code refactoring","level":3,"score":0.6098999977111816},{"id":"https://openalex.org/C101814296","wikidata":"https://www.wikidata.org/wiki/Q5439685","display_name":"Feature model","level":3,"score":0.5184000134468079},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5065000057220459},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.46309998631477356},{"id":"https://openalex.org/C519991488","wikidata":"https://www.wikidata.org/wiki/Q28865","display_name":"Python (programming language)","level":2,"score":0.4618000090122223},{"id":"https://openalex.org/C124670913","wikidata":"https://www.wikidata.org/wiki/Q2608526","display_name":"Process mining","level":5,"score":0.34619998931884766},{"id":"https://openalex.org/C12590798","wikidata":"https://www.wikidata.org/wiki/Q3933199","display_name":"Replication (statistics)","level":2,"score":0.3434999883174896},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3095000088214874},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2962000072002411},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.2928999960422516},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.2921999990940094},{"id":"https://openalex.org/C61423126","wikidata":"https://www.wikidata.org/wiki/Q187432","display_name":"Scripting language","level":2,"score":0.2761000096797943},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.2752000093460083},{"id":"https://openalex.org/C19351080","wikidata":"https://www.wikidata.org/wiki/Q1395034","display_name":"New product development","level":2,"score":0.2653000056743622},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.26499998569488525},{"id":"https://openalex.org/C207850805","wikidata":"https://www.wikidata.org/wiki/Q269608","display_name":"Reverse engineering","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2624000012874603}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.5281/zenodo.16789964","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.16789964","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Dataset"}],"best_oa_location":{"id":"doi:10.5281/zenodo.16789964","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.16789964","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Dataset"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"This":[0,28],"is":[1,687],"a":[2,648,711,721,955,1019],"replication":[3,642],"package":[4,643],"and":[5,157,179,203,287,290,307,310,344,364,371,397,420,468,570,625,800,809,899,918,1161,1308,1325],"online":[6],"appendix":[7],"for":[8,37,143,147,165,169,187,192,211,215,232,236,253,258,354,437,446,520,529,740,825,848,944,980,994,1025,1034,1058,1075,1117,1134,1189,1206,1242,1259,1296],"the":[9,31,40,46,55,64,68,78,89,94,105,116,123,133,137,152,174,198,220,224,241,245,264,268,278,283,288,298,303,308,317,327,331,357,405,438,443,469,492,497,504,510,539,551,576,582,589,609,619,637,652,657,660,664,674,677,695,718,734,737,743,752,771,787,791,795,806,819,826,842,849,863,873,883,905,908,911,915,919,958,964,969,981,995,1009,1014,1022,1028,1059,1068,1092,1102,1118,1127,1151,1165,1168,1190,1199,1223,1243,1252,1276,1285,1290,1294,1298,1302,1318,1328,1333],"TOSEM":[10],"Journal":[11],"paper":[12],"\"What":[13],"Were":[14],"You":[15],"Thinking?":[16],"An":[17],"LLM-Driven":[18],"Large-Scale":[19],"Study":[20,57,59,319,321],"of":[21,39,326,352,356,404,459,474,496,566,581,618,632,659,663,673,714,742,910,924,1027,1313],"Refactoring":[22,432,516,525,544,1341],"Motivations":[23,379,476],"in":[24,54,316,330,375,442,555,588,726,790,794,896,922,1036,1043],"Open-Source":[25],"Projects\".":[26],"Contents":[27],"repository":[29,653],"contains":[30],"following:":[32],"INSTALL:":[33],"Detailed":[34],"installation":[35],"instructions":[36],"each":[38,355,952,1026],"used":[41,665,789,793],"tools":[42],"as":[43,45,703,705],"well":[44,704],"required":[47],"Python":[48],"dependencies.":[49],"FIGURES:":[50],"Graphical":[51],"contents":[52],"shared":[53],"manuscript.":[56],"design:":[58,60,320],"Vertical":[61],"diagram":[62,76,87,103],"with":[63,77,88,104,132,155,177,201,223,244,267,277,282,297,302,409,486,558,651,751,770,786,805,818,828,841,851,862,872,882,904,954,984,998,1021,1040,1052,1067,1081,1111,1126,1140,1183,1198,1212,1236,1251,1265,1306,1349],"steps":[65],"undergone":[66,92],"during":[67,93],"study":[69,332,557,590],"design.":[70],"Data":[71,98,1006],"collection":[72,80,96],"process":[73,81,84,91,100,108,144,148,212,216,372,507,702,801,810,1162],"diagram:":[74,85,101],"Workflow":[75,86,102],"data":[79,95,106,603,824,835,847,932,1054,1113,1185,1238,1348],"undergone.":[82,109],"Sampling":[83],"sampling":[90,1029,1343],"stage.":[97,798,1062,1121,1193,1246,1330],"analysis":[99,107,394],"CoT":[110,124,781,796,1060,1119,1191,1244],"learning":[111,125,797,1061,1120,1192,1245],"example:":[112],"Example":[113],"prompt":[114,118],"from":[115,136,381,491,503,699,860,914,933,1356],"performed":[117,138,284,304],"engineering":[119],"stage":[120],"based":[121,150,172,196,218,239,262,384,389,454,614,1175,1228],"on":[122,151,173,197,219,240,263,385,390,538,543,615,1091,1150,1176,1222,1229,1275],"technique.":[126],"Results:":[127,423],"RMC":[128],"frequencies":[129,573],"histogram:":[130],"Histogram":[131],"RMCs":[134,490],"resulting":[135,502,698],"Open-Coding.":[139],"RF":[140,153,162,175,184,199],"feature":[141,159,163,181,185,205,209,226,230,247,251,270],"importance":[142,160,164,182,186,206,210,227,231,248,252,271],"metrics:":[145,167,190,213,234,256,369,802,868,877],"Results":[146,168,191,214,235,257,1174,1227],"metrics":[149,171,195,217,238,261,373,871,881,1163],"algorithm":[154,176,200,222,243,266],"MDA":[156,178,202],"MDG":[158,180,204],"metrics.":[161,183,207,811],"product":[166,170,233,237,808,1160],"all":[188,193,254,259,550,601,968,1284],"SW":[189,194,255,260],"XGB":[208,221,229,242,250,265],"IG":[225,246,269],"metric.":[228,249,272],"Spearman's":[273,279],"correlation":[274,280,294,300,399],"matrix:":[275,295],"Heatmap":[276,296],"matrix":[281,301],"uncorrected":[285,305],"correlations,":[286,306],"Bonferroni":[289,309],"BH-corrected":[291,311],"correlations.":[292,312],"Kendall's":[293,299],"TABLES:":[313],"Tables":[314],"appearing":[315],"article.":[318],"Context:":[322],"Summary":[323,458,473],"descriptive":[324],"statistics":[325],"projects":[328,586,913,1326],"considered":[329,406,554,587],"context.":[333,591],"RMT":[334,347,747,753,763,772,856,861],"detectable":[335,340,416],"refactoring":[336,353,358,387,407,439,498,512,552,563,611,760,779,830,853,865,930,1035,1314,1321,1323,1346],"types:":[337],"Refactorings":[338,414],"Types":[339,415,517,526],"by":[341,361,417],"RefactoringMiner":[342,362,418,1307],"3.0":[343,363,419],"newer":[345,365,421],"versions.":[346,422],"mined":[348,757,776],"output:":[349,748,857],"Detected":[350],"density":[351],"types":[359,408,440,553],"detected":[360,461,829,852,864],"version.":[366],"Computed":[367,869],"Software":[368,585,606],"Product":[370,799],"adopted":[374,620,935],"this":[376,556,641,900],"study.":[377],"Motivations:":[378],"derived":[380],"developers\u2019":[382,1103,1169],"answers":[383],"their":[386,559,571,633],"activity":[388],"related":[391,936],"work.":[392,937],"Correlation":[393],"interpretability:":[395],"Spearman\u2019s":[396],"Kendall\u2019s":[398],"coefficients'":[400],"interpretation.":[401],"refactoring-motivations-related:":[402],"Definitions":[403],"ground":[410],"truth":[411],"motivations.":[412,513],"refactorings:":[413],"RQ1:":[424,1031],"LLMvsHumCT:":[425],"LLM":[426,453,465,475,484,1063,1069,1082,1122,1128,1141,1177,1194,1200,1213,1247,1253,1266],"Motivation":[427,451],"Alignment":[428,485],"Agreement":[429],"Contingency":[430],"Table.":[431],"Type":[433],"representativeness:":[434],"Representativeness":[435],"level":[436],"sampled":[441,610,1056,1115,1187,1240],"manual":[444],"validation":[445,1070,1083,1090,1129,1142,1149,1201,1214,1221,1254,1267,1274],"RQ$_1$.":[447],"AgreementHvsLLM:":[448],"Testing":[449],"Human":[450,482],"against":[452],"motivation.":[455],"Disagreement":[456],"categories:":[457],"manually":[460],"disagreement":[462],"categories":[463,500,569],"between":[464,893,907],"extracted":[466,511,604],"motivations":[467,1033,1100,1166],"literature.":[470],"RQ2:":[471,1096],"ExtendedMotivations:":[472],"that":[477,508,629,1289],"Extend":[478],"Human-Reported":[479],"Motivations.":[480],"HumanValidatedAlignment:":[481],"Validated":[483],"Literature.":[487],"RQ3:":[488,1155],"Resulting":[489],"Open-Coding":[493],"stage:":[494],"Description":[495],"motivation":[499,568,931],"(RMC)":[501],"Open":[505],"Coding":[506],"group":[509],"CCR":[514,521],"RMC:":[515,524],"Frequencies":[518,527],"(#)":[519,528],"RMC.":[522,531],"CSRR":[523,530],"Related":[532,534,536],"Works:":[533],"Work:":[535],"work":[537,898,917,921],"existing":[540],"previous":[541],"research":[542],"Recommendation":[545],"Appendix:":[546],"Abbreviations:":[547],"Table":[548],"including":[549],"abbreviations.":[560],"Unique":[561],"LLM-based":[562],"motivations:":[564,1178,1231],"List":[565,580],"unique":[567],"respective":[572],"grouped":[574],"into":[575,720],"defined":[577],"RMCs.":[578],"PROJECTS:":[579],"initial":[583],"124":[584],"(Links":[592],"to":[593,655,683,1012,1171,1336],"GitHub":[594,949,956],"repositories":[595,638,666,680],"provided).":[596],"DATA:":[597],"A":[598,1281],"folder":[599,675,739,1020,1282],"containing":[600,676,1008,1283],"raw":[602,696],"Compressed":[605],"Repositories:":[607],"Since":[608],"observations":[612],"were":[613],"different":[616],"versions":[617],"repositories,":[621],"we":[622,646],"cloned":[623,678],"them":[624,627,890,1305],"analyzed":[626,912,1010],"at":[628],"specific":[630,661],"stager":[631],"development":[634],"process.":[635],"Providing":[636],"made":[639],"publishing":[640],"impossible,":[644],"therefore":[645],"provide":[647],"CSV":[649,902],"table":[650,686,903],"information":[654],"make":[656],"clonning":[658],"version":[662,821,844,1310,1350],"available.":[667],"(IMPORTANT)":[668],"NOTE:":[669],"The":[670,685,729],"real":[671],"size":[672,688,713,895,1335],"software":[679,1044],"summed":[681],"up":[682,725],"232GB.":[684],"58KB":[689],"now.":[690],"Mined":[691],"output":[692,754,764,773,859],"items:":[693,783],"Similarly,":[694],"outputs":[697,709],"RMiner's":[700],"mining":[701,708,1329],"other":[706],"additional":[707,1099],"reached":[710],"total":[712],"178GB.":[715],"We":[716],"compressed":[717,738],"content":[719,735,788],"7z":[722],"file":[723,750,769,785,804,817,840],"ending":[724],"5GB":[727],"approximately.":[728],"following":[730],"itemized":[731],"list":[732,965],"presents":[733],"within":[736],"one":[741],"processed":[744],"projects:":[745],"Filtered":[746,762],"JSON":[749,768,803,816,839,1051,1066,1110,1125,1182,1197,1235,1250],"after":[755,774,1327],"cleaning":[756,775],"commits":[758,777,827,850],"without":[759,778],"activity.":[761,780,831,854,866],"(UTF-8):":[765,836],"UTF-8":[766,837],"encoded":[767,838],"input":[782,1049,1053,1108,1112,1180,1184,1233,1237],"Pickle":[784],"prompts":[792],"computed":[807,879],"Version":[812,832],"Control":[813,833],"History":[814,834],"data:":[815,1050,1109,1181,1234],"ordered":[820,843],"control":[822,845,1311,1351],"history":[823,846,1312],"Raw":[855,858,1017],"Class":[867],"class":[870],"CK":[874,884,888],"tool.":[875],"Method":[876],"Additional":[878],"method":[880],"tool":[885],"(Not":[886],"used,":[887],"calculated":[889],"automatically).":[891],"Delta":[892],"project":[894,1045],"prior":[897,916],"work:":[901],"difference":[906],"status":[909],"current":[920],"terms":[923],"commits.":[925,1315],"Silva's":[926,934,1230],"Thematic":[927],"Data:":[928],"Ground-truth":[929],"Email":[938],"password:":[939],"Put":[940],"your":[941],"email":[942],"password":[943],"remote":[945],"emailing":[946],"(https://developers.google.com/workspace/gmail/api/guides/sending)":[947],".":[948],"tokens:":[950],"Fill":[951],"line":[953],"token,":[957],"script's":[959],"logic":[960],"will":[961],"parse":[962],"through":[963],"automatically":[966],"using":[967],"request":[970],"API":[971],"rate.":[972],"System":[973,987],"Message":[974,988],"Reasoning:":[975],"Initial":[976,990],"reasoning":[977,991],"system":[978,992],"message":[979,993],"first":[982,996],"interaction":[983,997],"Marco-o1":[985],"model.":[986],"Validation:":[989],"Mistral":[999],"NeMo,":[1000],"DeepSeek":[1001],"R1,":[1002],"Phi-4":[1003],"models.":[1004],"RESULTS:":[1005],"files":[1007,1024],"results":[1011,1018,1023],"answer":[1013],"Research":[1015],"Questions.":[1016],"strategies.":[1030],"Do":[1032],"past":[1037],"studies":[1038],"align":[1039],"those":[1041],"found":[1042],"change":[1046],"histories?":[1047],"Prompt":[1048,1107,1179,1232],"per":[1055,1114,1186,1239,1320],"observation":[1057,1116,1188,1241],"Validation":[1064,1123,1195,1248],"results:":[1065,1088,1124,1147,1196,1219,1249,1272],"results.":[1071,1084,1130,1143,1202,1215,1255,1268],"95%":[1072,1078,1093,1131,1137,1152,1203,1209,1224,1256,1262,1277],"CI":[1073,1079,1094,1132,1138,1153,1204,1210,1225,1257,1263,1278],"sample":[1074,1080,1087,1133,1139,1146,1205,1211,1218,1258,1264,1271,1334,1347,1358],"validation:":[1076,1135,1207,1260],"Drawn":[1077,1136,1208,1261],"Manually":[1085,1144,1216,1269],"validated":[1086,1145,1217,1270],"Manual":[1089,1148,1220,1273],"sample.":[1095,1154,1226,1279],"Are":[1097],"there":[1098],"driving":[1101,1167],"willingness":[1104,1170],"toperform":[1105],"refactoring?":[1106,1173],"To":[1156],"what":[1157],"extent":[1158],"can":[1159],"capture":[1164],"perform":[1172],"SCRIPTS:":[1280],"scripts.":[1286,1299],"Please,":[1287],"note":[1288],"numerical":[1291],"prefix":[1292],"denotes":[1293],"order":[1295],"launching":[1297],"00_clonning_repo_raw.py:":[1300],"Clones":[1301],"projects,":[1303],"mines":[1304],"collects":[1309],"01_stream_population_calculation.py:":[1316],"Calculates":[1317,1332],"population":[1319],"type,":[1322],"commit":[1324],"02_create_config.py:":[1331],"reach":[1337],"statistical":[1338],"significance.":[1339],"03_stream_sampling.py:":[1340],"case":[1342],"04_stream_matching.py:":[1344],"Matching":[1345],"data.":[1352],"05_stream_final_sample.py:":[1353],"Duplicate":[1354],"removal":[1355],"obtained":[1357],"06_ID_addi":[1359]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2025-10-10T00:00:00"}