pith. machine review for the scientific record. sign in

arxiv: 2605.01134 · v1 · submitted 2026-05-01 · 💻 cs.AI

Recognition: unknown

To Use AI as Dice of Possibilities with Timing Computation

Jia Li , Vipin Kumar , Rui Zhang
Authors on Pith no claims yet

Pith reviewed 2026-05-09 18:57 UTC · model grok-4.3

classification 💻 cs.AI
keywords verb-based paradigmtiming computationpossibilitylongitudinal EHRpatient trajectoriescounterfactual deductiondata-driven discoverybreast cancer data
0
0 comments X

The pith

A verb-based paradigm with timing computation lets AI discover patient trajectories and deduce counterfactual timings purely from data.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper argues that the prevailing noun-centered way of building AI models has blocked any proper handling of time as an open dimension filled with possibilities. It proposes instead a verb-based framework that defines timing computation and possibility in precise terms so AI can act as a tool for capturing the grammar of thought about the future. When tested on electronic health records from thousands of breast cancer patients, the approach automatically surfaces clinically meaningful trajectories and counterfactual timing deductions. These outcomes emerge without any supplied medical knowledge or manual tuning, which would matter because it shows AI can extract actionable patterns from raw longitudinal records in domains where expert input is costly or incomplete.

Core claim

The paper claims that replacing noun-based modeling with a verb-based paradigm, together with explicit definitions of timing computation and possibility, enables AI to treat the future as an open temporal space. When applied to longitudinal EHR data from 3,276 breast cancer patients, this framework produces automatic discovery of clinically significant patient trajectories and counterfactual timing deductions, all in a purely data-driven manner that requires no prior domain knowledge.

What carries the argument

The verb-based paradigm equipped with timing computation and a formal definition of possibility, which shifts AI from static object descriptions to dynamic action sequences that carry temporal structure.

If this is right

  • Longitudinal health records can yield meaningful patient pathways without expert annotation.
  • Counterfactual timing deductions become feasible directly from observed sequences.
  • The same data-driven process could apply to any other longitudinal dataset for trajectory analysis.
  • AI systems could model future possibilities more explicitly by treating verbs and their timings as primary elements.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The approach might reduce dependence on labeled training data in other sequential domains such as finance or logistics.
  • Extending the timing definitions to non-medical time series could test whether the paradigm generalizes beyond healthcare.
  • If the verb-based structure holds, it would imply that many current AI limitations stem from representational choices rather than data volume.

Load-bearing premise

That the verb-based paradigm and timing computation produce clinically significant trajectories and counterfactuals from raw data without any hidden domain knowledge or post-hoc adjustments.

What would settle it

Running the same method on the 3,276 breast cancer EHR records and finding that the discovered trajectories lack clinical significance or that the process implicitly relies on domain knowledge would falsify the central claim.

Figures

Figures reproduced from arXiv: 2605.01134 by Jia Li, Rui Zhang, Vipin Kumar.

Figure 1
Figure 1. Figure 1: Two representative trajectory clusters automatically discovered by the model. Red values denote view at source ↗
Figure 2
Figure 2. Figure 2: KDE plots of model-derived relative and counterfactual timing distributions for three events view at source ↗
read the original abstract

The dominant noun-based modeling paradigm has fundamentally constrained AI development, precluding any adequate representation of the future as an open temporal dimension. This paper introduces a verb-based paradigm, together with precise definitions of \emph{timing computation} and \emph{possibility}, that enables AI to function as an effective instrument for realizing the grammar of our thought. Applied to longitudinal EHR data from 3,276 breast cancer patients, the framework empirically demonstrates: (1) automatic discovery of clinically significant patient trajectories, and (2) counterfactual timing deduction. Both results are purely data-driven, require no prior domain knowledge, and, to our knowledge, represent the first such demonstrations in the machine learning literature.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper introduces a verb-based paradigm for AI modeling, along with definitions of timing computation and possibility, to represent the future as an open temporal dimension. Applied to longitudinal EHR data from 3,276 breast cancer patients, it claims to empirically demonstrate (1) automatic discovery of clinically significant patient trajectories and (2) counterfactual timing deduction, with both results being purely data-driven, requiring no prior domain knowledge, and representing the first such demonstrations in the ML literature.

Significance. If the methods and formal definitions can be provided and validated, the work could introduce a novel paradigm shift in temporal reasoning for AI, with implications for dynamic modeling in healthcare analytics. The purely data-driven claim for trajectory discovery and counterfactuals, if substantiated without embedded knowledge, would be noteworthy for its potential to reduce reliance on curated features in EHR analysis.

major comments (3)
  1. [Abstract] Abstract: The abstract asserts empirical results on 3,276 patients demonstrating automatic discovery of clinically significant trajectories and counterfactual timing deduction, but supplies no methods, validation details, error bars, baselines, or statistical tests. This prevents verification of the claims against the data.
  2. [Framework description] Framework description (verb-based paradigm, timing computation, and possibility): The central results are described as purely data-driven, yet they rest on newly introduced definitions whose precise form, equations, or algorithms are not shown. Without these, it is impossible to assess whether the definitions of timing computation and possibility embed the desired outcomes by construction, as is common in custom temporal formalisms.
  3. [Empirical demonstration] Empirical demonstration section: No algorithm for trajectory extraction from raw EHR events, no procedure for counterfactual deduction, and no mapping from events to 'possibility' or 'timing' are provided. Labeling trajectories as 'clinically significant' without explicit clinical features or post-hoc interpretation typically requires domain knowledge; the absence of this concrete mapping undermines the no-prior-domain-knowledge claim.
minor comments (2)
  1. The manuscript would benefit from pseudocode, formal mathematical definitions, or a reproducibility appendix for the key components to allow independent verification.
  2. Consider adding references and comparisons to existing literature on temporal data mining, patient trajectory modeling in EHR, and counterfactual reasoning in ML to contextualize the novelty claim.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments and the recognition of the potential significance of the verb-based paradigm. We agree that the current manuscript version requires expanded detail on the formal definitions, algorithms, and empirical procedures to enable verification. We will revise accordingly and address each point below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The abstract asserts empirical results on 3,276 patients demonstrating automatic discovery of clinically significant trajectories and counterfactual timing deduction, but supplies no methods, validation details, error bars, baselines, or statistical tests. This prevents verification of the claims against the data.

    Authors: We acknowledge that the abstract, due to length limits, omits methodological specifics and statistical information. In the revised manuscript we will augment the abstract with a brief outline of the timing computation approach, the validation strategy, and references to the statistical tests, baselines, and any error measures reported in the main text. revision: yes

  2. Referee: [Framework description] Framework description (verb-based paradigm, timing computation, and possibility): The central results are described as purely data-driven, yet they rest on newly introduced definitions whose precise form, equations, or algorithms are not shown. Without these, it is impossible to assess whether the definitions of timing computation and possibility embed the desired outcomes by construction, as is common in custom temporal formalisms.

    Authors: The initial submission presented the definitions at a conceptual level. We will revise the framework section to supply the explicit mathematical formulations and algorithmic descriptions of timing computation and possibility. These additions will make clear that the definitions provide a general mechanism applied to data rather than presupposing specific trajectories or counterfactuals. revision: yes

  3. Referee: [Empirical demonstration] Empirical demonstration section: No algorithm for trajectory extraction from raw EHR events, no procedure for counterfactual deduction, and no mapping from events to 'possibility' or 'timing' are provided. Labeling trajectories as 'clinically significant' without explicit clinical features or post-hoc interpretation typically requires domain knowledge; the absence of this concrete mapping undermines the no-prior-domain-knowledge claim.

    Authors: We will add a dedicated subsection detailing the algorithm that extracts trajectories by applying timing computation directly to raw event sequences. The counterfactual deduction procedure and the event-to-possibility/timing mapping will be specified step by step. Clinical significance will be justified through purely data-driven criteria (e.g., recurrence patterns and outcome correlations observable in the 3,276-patient cohort) without the use of external clinical features or prior knowledge during discovery; any post-hoc interpretation will be clearly separated from the automated process. revision: yes

Circularity Check

0 steps flagged

No circularity identified; derivation chain not exhibited in accessible text.

full rationale

The paper introduces a verb-based paradigm along with definitions of timing computation and possibility, then reports empirical results on EHR data as purely data-driven and free of prior domain knowledge. No specific equations, algorithms, or derivation steps are quoted in the abstract or referenced full-text placeholder that would allow inspection for self-definition, fitted-input prediction, or reduction of outputs to inputs by construction. The central claims therefore remain self-contained assertions without any load-bearing step that reduces to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 3 invented entities

Only the abstract is available, so the ledger is reconstructed from stated claims. The approach rests on two new definitions and a paradigm shift whose details and grounding are not provided.

axioms (1)
  • domain assumption A verb-based paradigm with precise definitions of timing computation and possibility enables AI to represent the future as an open temporal dimension.
    Presented in the abstract as the foundational shift that overcomes noun-based limitations.
invented entities (3)
  • verb-based paradigm no independent evidence
    purpose: To replace noun-based modeling and allow representation of time and possibility.
    Introduced as the core new framework.
  • timing computation no independent evidence
    purpose: Precise mechanism for handling when possibilities occur.
    Defined in the paper as part of the new paradigm.
  • possibility no independent evidence
    purpose: To model open future branches in AI reasoning.
    Defined alongside timing computation.

pith-pipeline@v0.9.0 · 5406 in / 1492 out tokens · 47874 ms · 2026-05-09T18:57:01.732907+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

142 extracted references · 21 canonical work pages · 2 internal anchors

  1. [1]

    2026 , eprint=

    Capture Timing-Attention of Events in Clinical Time Series , author=. 2026 , eprint=

    2026

  2. [2]

    Communications of the ACM , volume=

    The seven tools of causal inference, with reflections on machine learning , author=. Communications of the ACM , volume=. 2019 , publisher=

    2019

  3. [3]

    Scientific reports , volume=

    A comparison of machine learning methods for survival analysis of high-dimensional clinical data for dementia prediction , author=. Scientific reports , volume=. 2020 , publisher=

    2020

  4. [4]

    Proceedings of the 13th ACM international conference on bioinformatics, computational biology and health informatics , pages=

    Survtrace: Transformers for survival analysis with competing events , author=. Proceedings of the 13th ACM international conference on bioinformatics, computational biology and health informatics , pages=

  5. [5]

    Proceedings of the AAAI conference on artificial intelligence , volume=

    Deephit: A deep learning approach to survival analysis with competing risks , author=. Proceedings of the AAAI conference on artificial intelligence , volume=

  6. [6]

    BMC medical research methodology , volume=

    DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network , author=. BMC medical research methodology , volume=. 2018 , publisher=

    2018

  7. [7]

    International conference on machine learning , pages=

    Causal transformer for estimating counterfactual outcomes , author=. International conference on machine learning , pages=. 2022 , organization=

    2022

  8. [8]

    Frontiers in Artificial Intelligence , volume=

    InferBERT: a transformer-based causal inference framework for enhancing pharmacovigilance , author=. Frontiers in Artificial Intelligence , volume=. 2021 , publisher=

    2021

  9. [9]

    arXiv preprint arXiv:2210.15417 , year=

    Dynamic survival transformers for causal inference with electronic health records , author=. arXiv preprint arXiv:2210.15417 , year=

    work page arXiv

  10. [10]

    Time2Vec: Learning a Vector Representation of Time

    Time2vec: Learning a vector representation of time , author=. arXiv preprint arXiv:1907.05321 , year=

    work page Pith review arXiv 1907

  11. [11]

    Statistics in medicine , volume=

    Evaluation of the bias in using the time to the first event when the inter-event intervals have a Weibull distribution , author=. Statistics in medicine , volume=. 1999 , publisher=

    1999

  12. [12]

    ACM Computing Surveys (CSUR) , volume=

    Machine learning for survival analysis: A survey , author=. ACM Computing Surveys (CSUR) , volume=. 2019 , publisher=

    2019

  13. [13]

    International journal of mathematics and mathematical sciences , volume=

    Handling censoring and censored data in survival analysis: a standalone systematic literature review , author=. International journal of mathematics and mathematical sciences , volume=. 2021 , publisher=

    2021

  14. [14]

    European Journal for Philosophy of Science , volume=

    Individualisation and individualised science across disciplinary perspectives , author=. European Journal for Philosophy of Science , volume=. 2024 , publisher=

    2024

  15. [15]

    BMC medical research methodology , volume=

    Application of machine learning in predicting survival outcomes involving real-world data: a scoping review , author=. BMC medical research methodology , volume=. 2023 , publisher=

    2023

  16. [16]

    Statistical Analysis and Data Mining: The ASA Data Science Journal , volume=

    Survival analysis with electronic health record data: Experiments with chronic kidney disease , author=. Statistical Analysis and Data Mining: The ASA Data Science Journal , volume=. 2014 , publisher=

    2014

  17. [17]

    Operations Research for Health Care , volume=

    Comparing Markov and non-Markov alternatives for cost-effectiveness analysis: Insights from a cervical cancer case , author=. Operations Research for Health Care , volume=. 2019 , publisher=

    2019

  18. [18]

    medRxiv , pages=

    A Multi-state Non-Markov Framework to Estimate Course of Chronic Disease Progression , author=. medRxiv , pages=. 2024 , publisher=

    2024

  19. [19]

    arXiv preprint arXiv:2305.12640 , year=

    Limited resource allocation in a non-Markovian world: the case of maternal and child healthcare , author=. arXiv preprint arXiv:2305.12640 , year=

    work page arXiv

  20. [20]

    , author=

    Reasoning with Conditional Time-Intervals. , author=. FLAIRS , pages=

  21. [21]

    Sensors , volume=

    LSTM and GRU neural networks as models of dynamical processes used in predictive control: A comparison of models developed for two chemical reactors , author=. Sensors , volume=. 2021 , publisher=

    2021

  22. [22]

    2020 International workshop on electronic communication and artificial intelligence (IWECAI) , pages=

    Lstm and gru neural network performance comparison study: Taking yelp review dataset as an example , author=. 2020 International workshop on electronic communication and artificial intelligence (IWECAI) , pages=. 2020 , organization=

    2020

  23. [23]

    Journal of biomedical informatics , volume=

    Deep learning for temporal data representation in electronic health records: A systematic review of challenges and methodologies , author=. Journal of biomedical informatics , volume=. 2022 , publisher=

    2022

  24. [24]

    Journal of biomedical informatics , volume=

    Learning from heterogeneous temporal data in electronic health records , author=. Journal of biomedical informatics , volume=. 2017 , publisher=

    2017

  25. [25]

    IEEE journal of biomedical and health informatics , volume=

    Deep EHR: a survey of recent advances in deep learning techniques for electronic health record (EHR) analysis , author=. IEEE journal of biomedical and health informatics , volume=. 2017 , publisher=

    2017

  26. [26]

    Healthcare informatics research , volume=

    Detailed clinical models: representing knowledge, data and semantics in healthcare information technology , author=. Healthcare informatics research , volume=. 2014 , publisher=

    2014

  27. [27]

    Computer methods and programs in biomedicine , volume=

    Temporal data representation, normalization, extraction, and reasoning: A review from clinical domain , author=. Computer methods and programs in biomedicine , volume=. 2016 , publisher=

    2016

  28. [28]

    Health Data Science , volume=

    Moving beyond medical statistics: A systematic review on missing data handling in electronic health records , author=. Health Data Science , volume=. 2024 , publisher=

    2024

  29. [29]

    , title =

    Meiss, James D. , title =. 2007 , publisher =

    2007

  30. [30]

    2020 , publisher =

    Dawkins, Paul , title =. 2020 , publisher =

    2020

  31. [31]

    Entropy , volume=

    Emergence and causality in complex systems: a survey of causal emergence and related quantitative studies , author=. Entropy , volume=. 2024 , publisher=

    2024

  32. [32]

    2022 , publisher=

    Complexity science: the study of emergence , author=. 2022 , publisher=

    2022

  33. [33]

    Journal of biomedical informatics , volume=

    A practical perspective on the concordance index for the evaluation and selection of prognostic time-to-event models , author=. Journal of biomedical informatics , volume=. 2020 , publisher=

    2020

  34. [34]

    Korean Journal of Radiology , volume=

    Review of statistical methods for evaluating the performance of survival or other time-to-event prediction models (from conventional to deep learning approaches) , author=. Korean Journal of Radiology , volume=

  35. [35]

    Biostatistics , volume=

    The c-index is not proper for the evaluation of-year predicted risks , author=. Biostatistics , volume=. 2019 , publisher=

    2019

  36. [36]

    British journal of cancer , volume=

    Survival analysis part I: basic concepts and first analyses , author=. British journal of cancer , volume=. 2003 , publisher=

    2003

  37. [37]

    Circulation , volume=

    Cardiovascular disease and breast cancer: where these entities intersect: a scientific statement from the American Heart Association , author=. Circulation , volume=. 2018 , publisher=

    2018

  38. [38]

    Pitt, David , title =. The. 2022 , edition =

    2022

  39. [39]

    Scaling Learning Algorithms Towards

    Bengio, Yoshua and LeCun, Yann , booktitle =. Scaling Learning Algorithms Towards

  40. [40]

    Causality: Objectives and Assessment , pages=

    Distinguishing causes from effects using nonlinear acyclic causal models , author=. Causality: Objectives and Assessment , pages=. 2010 , organization=

    2010

  41. [41]

    NPJ digital medicine , volume=

    Deep representation learning of electronic health records to unlock patient stratification at scale , author=. NPJ digital medicine , volume=. 2020 , publisher=

    2020

  42. [42]

    SIAM Journal on Control and Optimization , volume=

    The remarkable effectiveness of time-dependent damping terms for second order evolution equations , author=. SIAM Journal on Control and Optimization , volume=. 2016 , publisher=

    2016

  43. [43]

    Ordinary Differential Equations/Motion with a Damping Force , year =

  44. [44]

    , title =

    Nave, Carl R. , title =. 2017 , publisher =

    2017

  45. [45]

    Canrecurrentneuralnetworkswarptime? arXiv:1804.11188,

    Can recurrent neural networks warp time? , author=. arXiv preprint arXiv:1804.11188 , year=

    work page arXiv

  46. [46]

    2019 , publisher=

    Statistical analysis with missing data , author=. 2019 , publisher=

    2019

  47. [47]

    Mathematical and Scientific Machine Learning , pages=

    Gating creates slow modes and controls phase-space complexity in grus and lstms , author=. Mathematical and Scientific Machine Learning , pages=. 2020 , organization=

    2020

  48. [48]

    Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling

    Empirical evaluation of gated recurrent neural networks on sequence modeling , author=. arXiv preprint arXiv:1412.3555 , year=

    work page internal anchor Pith review arXiv

  49. [49]

    Scientific reports , volume=

    Recurrent neural networks for multivariate time series with missing values , author=. Scientific reports , volume=. 2018 , publisher=

    2018

  50. [50]

    Iscience , volume=

    Risk prediction of heart diseases in patients with breast cancer: A deep learning approach with longitudinal electronic health records data , author=. Iscience , volume=. 2024 , publisher=

    2024

  51. [51]

    Philosophies , volume=

    Causal emergence: When distortions in a map obscure the territory , author=. Philosophies , volume=. 2022 , publisher=

    2022

  52. [52]

    Proceedings of the National Academy of Sciences , volume=

    Quantifying causal emergence shows that macro can beat micro , author=. Proceedings of the National Academy of Sciences , volume=. 2013 , publisher=

    2013

  53. [53]

    Entropy , volume=

    When the map is better than the territory , author=. Entropy , volume=. 2017 , publisher=

    2017

  54. [54]

    BMC neuroscience , volume=

    Measuring information integration , author=. BMC neuroscience , volume=. 2003 , publisher=

    2003

  55. [55]

    Causality, feedback and directed information , author=. Proc. Int. Symp. Inf. Theory Applic.(ISITA-90) , pages=

  56. [56]

    Physical review letters , volume=

    Measuring information transfer , author=. Physical review letters , volume=. 2000 , publisher=

    2000

  57. [57]

    nature , volume=

    Learning representations by back-propagating errors , author=. nature , volume=. 1986 , publisher=

    1986

  58. [58]

    , author=

    012: A Mathematical Examination of the Methods of Determining the Accuracy of an Observation by the Mean Error, and by the Mean Square Error. , author=

  59. [59]

    XXXII: Laplace, Fisher, and the discovery of the concept of sufficiency , author=

    Studies in the history of probability and statistics. XXXII: Laplace, Fisher, and the discovery of the concept of sufficiency , author=. Biometrika , volume=. 1973 , publisher=

    1973

  60. [60]

    Nature , pages=

    Human-like systematic generalization through a meta-learning neural network , author=. Nature , pages=. 2023 , publisher=

    2023

  61. [61]

    IEEE transactions on pattern analysis and machine intelligence , volume=

    Meta-learning in neural networks: A survey , author=. IEEE transactions on pattern analysis and machine intelligence , volume=. 2021 , publisher=

    2021

  62. [62]

    2023 , eprint=

    Language Models Represent Space and Time , author=. 2023 , eprint=

    2023

  63. [63]

    The next decade in ai: four steps towards robust artificial intelligence

    The next decade in AI: four steps towards robust artificial intelligence , author=. arXiv preprint arXiv:2002.06177 , year=

    work page arXiv 2002

  64. [64]

    2023 , month = may, journal =

    Are emergent abilities of Large Language Models a mirage? , author=. arXiv preprint arXiv:2304.15004 , year=

    work page arXiv

  65. [65]

    ACM Turing award lectures , pages=

    Computer science as empirical inquiry: Symbols and search , author=. ACM Turing award lectures , pages=

  66. [66]

    Philosophical Transactions of the Royal Society A , volume=

    Symbols and grounding in large language models , author=. Philosophical Transactions of the Royal Society A , volume=. 2023 , publisher=

    2023

  67. [67]

    Journal of Mathematical Psychology , volume=

    A tutorial on Fisher information , author=. Journal of Mathematical Psychology , volume=. 2017 , publisher=

    2017

  68. [68]

    Statistical science , volume=

    Confounding and collapsibility in causal inference , author=. Statistical science , volume=. 1999 , publisher=

    1999

  69. [69]

    Annual Review of Statistics and Its Application , volume=

    Granger causality: A review and recent advances , author=. Annual Review of Statistics and Its Application , volume=. 2022 , publisher=

    2022

  70. [70]

    and Osindero, Simon and Teh, Yee Whye , journal =

    Hinton, Geoffrey E. and Osindero, Simon and Teh, Yee Whye , journal =. A Fast Learning Algorithm for Deep Belief Nets , volume =

  71. [71]

    OUP Catalogue , year=

    Modelling non-linear economic relationships , author=. OUP Catalogue , year=

  72. [72]

    Journal of motor behavior , volume=

    Reducing knowledge of results about relative versus absolute timing: Differential effects on learning , author=. Journal of motor behavior , volume=. 1994 , publisher=

    1994

  73. [73]

    Journal of motor behavior , volume=

    Effects of an auditory model on the learning of relative and absolute timing , author=. Journal of motor behavior , volume=. 2001 , publisher=

    2001

  74. [74]

    Cognitive Semiotics , volume=

    Understanding timelines: Conceptual metaphor and conceptual integration , author=. Cognitive Semiotics , volume=. 2009 , publisher=

    2009

  75. [75]

    Artificial Intelligence , volume=

    A survey of inverse reinforcement learning: Challenges, methods and progress , author=. Artificial Intelligence , volume=. 2021 , publisher=

    2021

  76. [76]

    2016 , publisher=

    Deep learning , author=. 2016 , publisher=

    2016

  77. [77]

    2018 , publisher=

    The book of why: the new science of cause and effect , author=. 2018 , publisher=

    2018

  78. [78]

    IEEE Access , volume=

    A survey on trajectory data mining: Techniques and applications , author=. IEEE Access , volume=. 2016 , publisher=

    2016

  79. [79]

    Australian journal of ecology , volume=

    Partitioning the variation among spatial, temporal and environmental components in a multivariate data set , author=. Australian journal of ecology , volume=. 1998 , publisher=

    1998

  80. [80]

    Knowledge and information systems , volume=

    A survey of methods for time series change point detection , author=. Knowledge and information systems , volume=. 2017 , publisher=

    2017

Showing first 80 references.