{
 "about": "Every reference of the Phase 4 manuscript with the source of its bibliographic record, the retrieval time and what was read. VERIFIED: the record was retrieved and the content the manuscript relies on was read (abstract or more, or the record itself for a method or software citation). PARTIAL: record retrieved, content not read; such a reference carries a title-level statement only. Journal names are abbreviated by ISO 4 word abbreviations (LTWA); where Crossref holds a short title it was used, the others were abbreviated by hand and are listed for a check against the LTWA at the final gate. After the independent review (item R13), no PARTIAL reference remains in the list.",
 "abbreviations_from_crossref": [
  "Sci Data",
  "Int J Adv Manuf Technol",
  "Nat Mach Intell",
  "IEEE Trans Audio Electroacoust",
  "Comput Sci Eng"
 ],
 "journal_names_edited": "Crossref short titles are printed without full stops, as in the journal's reference example",
 "not_in_list": {
  "U3": {
   "what": "HBAF: Bayesian milling monitoring, code repository with an unpublished Russian-language manuscript",
   "url": "https://github.com/Guldek1987/bayesian-milling-monitoring-hbaf",
   "accessed": "2026-09-22",
   "handling": "mentioned in the text with its URL and kept out of the reference list (Adam's answer to Q4)"
  }
 },
 "references": [
  {
   "key": "kim2026msm",
   "bibtype": "article",
   "author": [
    "Kim, Eunseob",
    "Sim, Yuseop",
    "Li, Adrian Shuai",
    "Mostafiz, Mir Imtiaz",
    "Van Meter, Zachary",
    "Jun, Martin Byung-Guk",
    "Bertino, Elisa",
    "Shakouri, Ali"
   ],
   "title": "Multi-sensor and MTConnect dataset of metal cutting anomaly in milling from laboratory and industry settings",
   "journal_full": "Scientific Data",
   "journal": "Sci Data",
   "year": "2026",
   "volume": "13",
   "number": "1",
   "pages": "945",
   "doi": "10.1038/s41597-026-07255-7",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "full article text in the built-in browser (Phase 2, 2026-09-22; placement sentences re-read 2026-09-23)",
   "tag": "VERIFIED",
   "used_for": "the dataset, its machines, sensors and placements, the label definitions and the licence"
  },
  {
   "key": "kim2025kaggle",
   "bibtype": "misc",
   "author": [
    "Kim, Eunseob",
    "Sim, Yuseop",
    "Li, Adrian Shuai",
    "Mostafiz, Mir Imtiaz",
    "Van Meter, Zachary",
    "Jun, Martin Byung-Guk",
    "Bertino, Elisa",
    "Shakouri, Ali"
   ],
   "title": "Metal milling anomaly dataset",
   "publisher": "Kaggle",
   "year": "2025",
   "doi": "10.34740/kaggle/ds/8392825",
   "howpublished": "Kaggle",
   "note_bib": "Dataset, version 2",
   "source": "DataCite REST API (api.datacite.org/dois/DOI), read through the built-in browser; version 2 from the Kaggle dataset record read in Phase 2",
   "retrieved_utc": "2026-09-23T01:42:34Z",
   "content_read": "dataset files (Phase 2)",
   "tag": "VERIFIED",
   "used_for": "the dataset itself, cited with the DataCite minimum (authors, title, repository, identifier)"
  },
  {
   "key": "parmar2026",
   "bibtype": "article",
   "author": [
    "Parmar, Harshkumar K.",
    "Raman, Shivakumar"
   ],
   "title": "AI/ML-based interoperable anomaly detection in advanced manufacturing using MTconnect-derived multi-machine data",
   "journal_full": "The International Journal of Advanced Manufacturing Technology",
   "journal": "Int J Adv Manuf Technol",
   "year": "2026",
   "volume": "145",
   "number": "7-8",
   "pages": "5361-5379",
   "doi": "10.1007/s00170-026-18611-7",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "full text of the CC BY Research Square preprint (Phase 2); published article page for the article type (Phase 3); published article full text (Springer Nature Link, institutional access, read through the built-in browser 2026-09-23, between 02:48Z and 03:00Z): experiment-level statistical features from the MTConnect files, labels from the documented run remarks of the dataset summary, one feature vector per run, held-out testing, stratified and leave-one-out cross-validation over runs; the description in the Introduction matches it",
   "tag": "VERIFIED",
   "used_for": "U1: run-level MTConnect models with labels from the run remarks, cross-validation over runs",
   "review_item": "R13"
  },
  {
   "key": "karkadakattil2026",
   "bibtype": "article",
   "author": [
    "Karkadakattil, Aswin"
   ],
   "title": "A physics-guided unified damage index framework for explainable milling anomaly detection using interval-wise MTConnect multi-sensor fusion",
   "journal_full": "Journal of Intelligent Manufacturing and Special Equipment",
   "journal": "J Intell Manuf Spec Equip",
   "year": "2026",
   "pages": "1-33",
   "doi": "10.1108/jimse-05-2026-0020",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "full text, open access (Phase 2)",
   "tag": "VERIFIED",
   "used_for": "U2: interval-wise MTConnect features, labels 1 and 2 merged, stratified splits over samples"
  },
  {
   "key": "kim2024chatter",
   "bibtype": "article",
   "author": [
    "Kim, Eunseob",
    "Bui, Thu",
    "Yuan, Junyi",
    "Chandra Mouli, S.",
    "Ribeiro, Bruno",
    "Yeh, Raymond A.",
    "Fassnacht, Michael P.",
    "Jun, Martin B. G."
   ],
   "title": "Online real-time machining chatter sound detection using convolutional neural network by adopting expert knowledge",
   "journal_full": "Manufacturing Letters",
   "journal": "Manuf Lett",
   "year": "2024",
   "volume": "41",
   "pages": "1386-1397",
   "doi": "10.1016/j.mfglet.2024.09.165",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "abstract (OpenAlex, 2026-09-23T01:41:52Z)",
   "tag": "VERIFIED",
   "used_for": "earlier work of two of the dataset authors: chatter events labelled from machining sound with expert knowledge, then a convolutional network"
  },
  {
   "key": "navarro2023",
   "bibtype": "article",
   "author": [
    "Navarro-Devia, John Henry",
    "Chen, Yun",
    "Dao, Dzung Viet",
    "Li, Huaizhong"
   ],
   "title": "Chatter detection in milling processes: a review on signal processing and condition classification",
   "journal_full": "The International Journal of Advanced Manufacturing Technology",
   "journal": "Int J Adv Manuf Technol",
   "year": "2023",
   "volume": "125",
   "number": "9-10",
   "pages": "3943-3980",
   "doi": "10.1007/s00170-023-10969-2",
   "source": "Crossref bibliographic query through the built-in browser",
   "retrieved_utc": "2026-09-23T01:42:34Z",
   "content_read": "abstract (OpenAlex, 2026-09-23T01:42:52Z)",
   "tag": "VERIFIED",
   "title_edit": "the em dash after 'processes' in the registered title is printed as a colon (house rule against em dashes)",
   "used_for": "chatter harms surface quality and tool life and limits the material removal rate, and online detection of its onset has been studied over several decades (both stated in the abstract, re-read 2026-09-23, between 02:48Z and 03:00Z); sensing, signal processing, chatter indicators and data-driven classification reviewed",
   "review_item": "R13"
  },
  {
   "key": "delio1992",
   "bibtype": "article",
   "author": [
    "Delio, T.",
    "Tlusty, J.",
    "Smith, S."
   ],
   "title": "Use of audio signals for chatter detection and control",
   "journal_full": "Journal of Engineering for Industry",
   "journal": "J Eng Ind",
   "year": "1992",
   "volume": "114",
   "number": "2",
   "pages": "146-157",
   "doi": "10.1115/1.2899767",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:58Z",
   "content_read": "abstract (Crossref)",
   "tag": "VERIFIED",
   "used_for": "a microphone compared with dynamometers, displacement probes and accelerometers for sensing unstable milling, giving a competitive and in many instances superior chatter signal (abstract, re-read 2026-09-23, between 02:48Z and 03:00Z)",
   "review_item": "R13"
  },
  {
   "key": "kapoor2023",
   "bibtype": "article",
   "author": [
    "Kapoor, Sayash",
    "Narayanan, Arvind"
   ],
   "title": "Leakage and the reproducibility crisis in machine-learning-based science",
   "journal_full": "Patterns",
   "journal": "Patterns",
   "year": "2023",
   "volume": "4",
   "number": "9",
   "pages": "100804",
   "doi": "10.1016/j.patter.2023.100804",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "abstract (OpenAlex)",
   "tag": "VERIFIED",
   "used_for": "data leakage in machine-learning-based science and its over-optimistic results"
  },
  {
   "key": "roberts2017",
   "bibtype": "article",
   "author": [
    "Roberts, David R.",
    "Bahn, Volker",
    "Ciuti, Simone",
    "Boyce, Mark S.",
    "Elith, Jane",
    "Guillera-Arroita, Gurutzeta",
    "Hauenstein, Severin",
    "Lahoz-Monfort, Jos\\'{e} J.",
    "Schr\\\"{o}der, Boris",
    "Thuiller, Wilfried",
    "Warton, David I.",
    "Wintle, Brendan A.",
    "Hartig, Florian",
    "Dormann, Carsten F."
   ],
   "title": "Cross-validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure",
   "journal_full": "Ecography",
   "journal": "Ecography",
   "year": "2017",
   "volume": "40",
   "number": "8",
   "pages": "913-929",
   "doi": "10.1111/ecog.02881",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:58Z",
   "content_read": "abstract (Crossref)",
   "tag": "VERIFIED",
   "used_for": "random cross-validation on structured data underestimates predictive error; blocked splits"
  },
  {
   "key": "geirhos2020",
   "bibtype": "article",
   "author": [
    "Geirhos, Robert",
    "Jacobsen, J\\\"{o}rn-Henrik",
    "Michaelis, Claudio",
    "Zemel, Richard",
    "Brendel, Wieland",
    "Bethge, Matthias",
    "Wichmann, Felix A."
   ],
   "title": "Shortcut learning in deep neural networks",
   "journal_full": "Nature Machine Intelligence",
   "journal": "Nat Mach Intell",
   "year": "2020",
   "volume": "2",
   "number": "11",
   "pages": "665-673",
   "doi": "10.1038/s42256-020-00257-z",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:58Z",
   "content_read": "abstract of the arXiv version 2004.07780 (arxiv.org, 2026-09-23T01:43:15Z), which names the Nature Machine Intelligence publication",
   "tag": "VERIFIED",
   "used_for": "shortcuts: decision rules that do well on benchmarks and fail to transfer"
  },
  {
   "key": "smith2015",
   "bibtype": "article",
   "author": [
    "Smith, Wade A.",
    "Randall, Robert B."
   ],
   "title": "Rolling element bearing diagnostics using the Case Western Reserve University data: a benchmark study",
   "journal_full": "Mechanical Systems and Signal Processing",
   "journal": "Mech Syst Signal Process",
   "year": "2015",
   "volume": "64-65",
   "pages": "100-131",
   "doi": "10.1016/j.ymssp.2015.04.021",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "abstract, from the UNSWorks record of the authors' institution (http://hdl.handle.net/1959.4/unsworks_37602), read through the built-in browser 2026-09-23, between 02:48Z and 03:00Z; the publisher page showed a captcha and was not used",
   "tag": "VERIFIED",
   "used_for": "the CWRU bearing data as a standard reference for testing diagnostic algorithms, and a benchmark study that examined the whole data set and recommended how it should be used (all in the abstract)",
   "review_item": "R13"
  },
  {
   "key": "matania2024",
   "bibtype": "article",
   "author": [
    "Matania, Omri",
    "Cohen, Roee",
    "Bechhoefer, Eric",
    "Bortman, Jacob"
   ],
   "title": "Test-training leakage in evaluation of machine learning algorithms for condition-based maintenance",
   "journal_full": "PHM Society European Conference",
   "journal": "PHM Soc Eur Conf",
   "year": "2024",
   "volume": "8",
   "number": "1",
   "pages": "13",
   "doi": "10.36001/phme.2024.v8i1.4125",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser; the publisher's article page (papers.phmsociety.org, article 4125) gives the same citation, 8(1), 13",
   "retrieved_utc": "2026-09-23, between 02:48Z and 03:00Z",
   "content_read": "abstract (Crossref record and the publisher's article page; licence CC BY 3.0)",
   "tag": "VERIFIED",
   "used_for": "incorrect separation of training and test examples described as a fairly common problem in evaluations of machine learning for bearing fault classification, leading to optimistic conclusions (abstract)",
   "journal_abbrev_note": "abbreviated by hand (Society, European, Conference); to be checked against the LTWA with the others",
   "review_item": "R13"
  },
  {
   "key": "vonhahn2022",
   "bibtype": "article",
   "author": [
    "von Hahn, Tim",
    "Mechefske, Chris K."
   ],
   "title": "Machine learning in CNC machining: best practices",
   "journal_full": "Machines",
   "journal": "Machines",
   "year": "2022",
   "volume": "10",
   "number": "12",
   "pages": "1233",
   "doi": "10.3390/machines10121233",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:58Z",
   "content_read": "abstract (Crossref); full text in Phase 2",
   "tag": "VERIFIED",
   "used_for": "best practices for machine learning in CNC machining, including awareness of data leakage"
  },
  {
   "key": "hanley1982",
   "bibtype": "article",
   "author": [
    "Hanley, J. A.",
    "McNeil, B. J."
   ],
   "title": "The meaning and use of the area under a receiver operating characteristic (ROC) curve",
   "journal_full": "Radiology",
   "journal": "Radiology",
   "year": "1982",
   "volume": "143",
   "number": "1",
   "pages": "29-36",
   "doi": "10.1148/radiology.143.1.7063747",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "abstract (OpenAlex)",
   "tag": "VERIFIED",
   "used_for": "the AUC as the probability that a random positive ranks above a random negative (Wilcoxon statistic)"
  },
  {
   "key": "cohen1960",
   "bibtype": "article",
   "author": [
    "Cohen, Jacob"
   ],
   "title": "A coefficient of agreement for nominal scales",
   "journal_full": "Educational and Psychological Measurement",
   "journal": "Educ Psychol Meas",
   "year": "1960",
   "volume": "20",
   "number": "1",
   "pages": "37-46",
   "doi": "10.1177/001316446002000104",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "title and bibliographic record; cited for the definition of kappa",
   "tag": "VERIFIED",
   "used_for": "Cohen's kappa"
  },
  {
   "key": "holm1979",
   "bibtype": "article",
   "author": [
    "Holm, Sture"
   ],
   "title": "A simple sequentially rejective multiple test procedure",
   "journal_full": "Scandinavian Journal of Statistics",
   "journal": "Scand J Stat",
   "year": "1979",
   "volume": "6",
   "pages": "65-70",
   "source": "OpenAlex record W2121044470 through the built-in browser (Crossref holds no record for the JSTOR identifier OpenAlex lists); the JSTOR stable page (read through the built-in browser 2026-09-23, between 02:48Z and 03:00Z) confirms volume 6, issue 2, pages 65 to 70. The DOI that OpenAlex lists for the JSTOR identifier (10.2307/4615733) is not registered: the doi.org handle API answers responseCode 100 (not found)",
   "retrieved_utc": "2026-09-23T01:42:52Z",
   "content_read": "bibliographic record; cited for the procedure",
   "tag": "VERIFIED",
   "used_for": "the Holm step-down procedure",
   "number": "2",
   "url": "https://www.jstor.org/stable/4615733",
   "review_item": "R13"
  },
  {
   "key": "welch1967",
   "bibtype": "article",
   "author": [
    "Welch, P."
   ],
   "title": "The use of fast Fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms",
   "journal_full": "IEEE Transactions on Audio and Electroacoustics",
   "journal": "IEEE Trans Audio Electroacoust",
   "year": "1967",
   "volume": "15",
   "number": "2",
   "pages": "70-73",
   "doi": "10.1109/TAU.1967.1161901",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "bibliographic record; cited for the method",
   "tag": "VERIFIED",
   "used_for": "Welch power spectra"
  },
  {
   "key": "field2007",
   "bibtype": "article",
   "author": [
    "Field, C. A.",
    "Welsh, A. H."
   ],
   "title": "Bootstrapping clustered data",
   "journal_full": "Journal of the Royal Statistical Society Series B: Statistical Methodology",
   "journal": "J R Stat Soc Ser B Stat Methodol",
   "year": "2007",
   "volume": "69",
   "number": "3",
   "pages": "369-390",
   "doi": "10.1111/j.1467-9868.2007.00593.x",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "abstract (Crossref)",
   "tag": "VERIFIED",
   "used_for": "resampling whole clusters (the cluster bootstrap) for clustered data"
  },
  {
   "key": "breiman2001",
   "bibtype": "article",
   "author": [
    "Breiman, Leo"
   ],
   "title": "Random forests",
   "journal_full": "Machine Learning",
   "journal": "Mach Learn",
   "year": "2001",
   "volume": "45",
   "number": "1",
   "pages": "5-32",
   "doi": "10.1023/A:1010933404324",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "bibliographic record; cited for the method",
   "tag": "VERIFIED",
   "used_for": "random forests"
  },
  {
   "key": "pedregosa2011",
   "bibtype": "article",
   "author": [
    "Pedregosa, Fabian",
    "Varoquaux, Ga\\\"{e}l",
    "Gramfort, Alexandre",
    "Michel, Vincent",
    "Thirion, Bertrand",
    "Grisel, Olivier",
    "Blondel, Mathieu",
    "Prettenhofer, Peter",
    "Weiss, Ron",
    "Dubourg, Vincent",
    "Vanderplas, Jake",
    "Passos, Alexandre",
    "Cournapeau, David",
    "Brucher, Matthieu",
    "Perrot, Matthieu",
    "Duchesnay, \\'{E}douard"
   ],
   "title": "Scikit-learn: machine learning in Python",
   "journal_full": "Journal of Machine Learning Research",
   "journal": "J Mach Learn Res",
   "year": "2011",
   "volume": "12",
   "number": "85",
   "pages": "2825-2830",
   "url": "https://jmlr.org/papers/v12/pedregosa11a.html",
   "source": "JMLR article page (jmlr.org/papers/v12/pedregosa11a.html) in the built-in browser",
   "retrieved_utc": "2026-09-23, between 01:43:15Z and 01:43:34Z (page read with get_page_text, which returns no time stamp)",
   "content_read": "abstract (JMLR page)",
   "tag": "VERIFIED",
   "used_for": "scikit-learn"
  },
  {
   "key": "harris2020",
   "bibtype": "article",
   "author": [
    "Harris, Charles R.",
    "Millman, K. Jarrod",
    "van der Walt, St\\'{e}fan J.",
    "Gommers, Ralf",
    "Virtanen, Pauli",
    "Cournapeau, David",
    "Wieser, Eric",
    "Taylor, Julian",
    "Berg, Sebastian",
    "Smith, Nathaniel J.",
    "Kern, Robert",
    "Picus, Matti",
    "Hoyer, Stephan",
    "van Kerkwijk, Marten H.",
    "Brett, Matthew",
    "Haldane, Allan",
    "del R\\'{i}o, Jaime Fern\\'{a}ndez",
    "Wiebe, Mark",
    "Peterson, Pearu",
    "G\\'{e}rard-Marchant, Pierre",
    "Sheppard, Kevin",
    "Reddy, Tyler",
    "Weckesser, Warren",
    "Abbasi, Hameer",
    "Gohlke, Christoph",
    "Oliphant, Travis E."
   ],
   "title": "Array programming with NumPy",
   "journal_full": "Nature",
   "journal": "Nature",
   "year": "2020",
   "volume": "585",
   "number": "7825",
   "pages": "357-362",
   "doi": "10.1038/s41586-020-2649-2",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "bibliographic record; cited for the software",
   "tag": "VERIFIED",
   "used_for": "NumPy"
  },
  {
   "key": "mckinney2010",
   "bibtype": "inproceedings",
   "author": [
    "McKinney, Wes"
   ],
   "title": "Data structures for statistical computing in Python",
   "booktitle": "Proceedings of the Python in Science Conference",
   "year": "2010",
   "pages": "56-61",
   "doi": "10.25080/Majora-92bf1922-00a",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "bibliographic record; cited for the software",
   "tag": "VERIFIED",
   "used_for": "pandas"
  },
  {
   "key": "hunter2007",
   "bibtype": "article",
   "author": [
    "Hunter, John D."
   ],
   "title": "Matplotlib: a 2D graphics environment",
   "journal_full": "Computing in Science & Engineering",
   "journal": "Comput Sci Eng",
   "year": "2007",
   "volume": "9",
   "number": "3",
   "pages": "90-95",
   "doi": "10.1109/MCSE.2007.55",
   "source": "Crossref REST API (api.crossref.org/works/DOI), read through the built-in browser",
   "retrieved_utc": "2026-09-23T01:33:41Z",
   "content_read": "bibliographic record; cited for the software",
   "tag": "VERIFIED",
   "used_for": "Matplotlib, the graphics program of the figures (R13)"
  }
 ],
 "dropped": {
  "hendriks2022": {
   "record": {
    "author": [
     "Hendriks, Jacob",
     "Dumond, Patrick",
     "Knox, D. A."
    ],
    "title": "Towards better benchmarking using the CWRU bearing fault dataset",
    "journal_full": "Mechanical Systems and Signal Processing",
    "year": "2022",
    "volume": "169",
    "pages": "108732",
    "doi": "10.1016/j.ymssp.2021.108732"
   },
   "reason": "only the title and venue could be read: the publisher page showed a captcha, NASA ADS a human-verification page, ResearchGate answered 429, and neither OpenAlex nor Semantic Scholar holds the abstract (2026-09-23, between 02:48Z and 03:00Z). A reference read at title level is no longer cited (review item R13); its place in the Introduction is taken by matania2024, whose abstract was read."
  }
 }
}