Formulating event-based critical observations in diagnostic problems

Christopher, Cody James; Grastien, Alban

doi:10.1109/cdc.2015.7402916

Cited by 5 publications

(7 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This extension does not ultimately affect the complexity of the algorithm, as it is equivalent to the worst-case exhaustive search under the greedy approach. In [11] it was demonstrated that the equivalent of Algorithm 3 had complexity O(n 2 m 2 ) in the number of calls to the model based diagnoser ∆. We provably retain the same bound but in calls to ∆ L (·) at worse for each trace, and can be dramatically improved with heuristics that allow us to recall and then avoid the types of abstraction that caused pruning earlier in the search.…”

Section: Inferring the Critical Observations Setmentioning

confidence: 78%

“…This then provides all the conditions necessary to define a terminating search algorithm (such as that given in [11], along with child generation, which as been adapted for this setting) presented in Figure 3.…”

Section: Inferring the Critical Observations Setmentioning

confidence: 99%

“…We propose updating the theory of critical observations introduced in [11] to infer fault signatures as critical observations. Before the notion of critical observations can be introduced, we first define the idea of a sub-observation.…”

Section: Theory Of Sub-observationsmentioning

confidence: 99%

“…In [11], critical observations are computed with respect to the model of the system S. Given a trace o, a critical observation is a sub-observation θ that is precise enough to infer the original diagnosis of o given S and that is also maximally abstracted. That is, the sub-observation θ cannot further be abstracted without compromising the known diagnosis of the original observation o.…”

Section: Critical Observations and Log-based Diagnosismentioning

confidence: 99%

“…to design a diagnoser of the system without the need for a model that would require expertise, simply based on data logs produced by the system; 2. to abstract away irrelevant information to provide a diagnoser that performs more efficiently by simply recognising fault signatures represented by critical observations [11] and emit the associated diagnosis;…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Inference of fault signatures of discrete-event systems from event logs

Christopher¹,

Pencolé²,

Grastien³

Kalpa Publications in Computing

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a method to diagnose faults in a discrete event system that only relies on past observed logs and not on any behavioural model of the system. Given a set of tagged logs produced by the system, the first objective is to extract from them a set of fault signatures. These fault signatures are represented with a set of critical observations that are the support of the diagnosis method. We first propose a method to compute the fault signatures from an initial log journal and follow with detail on how the signatures can then be updated when new logs are available.

show abstract

Section: Inferring the Critical Observations Setmentioning

confidence: 78%

Section: Inferring the Critical Observations Setmentioning

confidence: 99%

Section: Theory Of Sub-observationsmentioning

confidence: 99%

Section: Critical Observations and Log-based Diagnosismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Inference of fault signatures of discrete-event systems from event logs

Christopher¹,

Pencolé²,

Grastien³

Kalpa Publications in Computing

Self Cite

View full text Add to dashboard Cite

show abstract

Model‐based diagnosis with uncertain observations

Cazes

Kalech

2021

Int J Intell Syst

View full text Add to dashboard Cite

Classical model-based diagnosis uses a model of the system to infer diagnoses-explanations-of a given abnormal observation. In this study we address the case where there is uncertainty over a given observation and explore it. This can happen, for example, when the observation outputs are collected by sensors with some noise, that are known to return incorrect value with some probability. We formally define this problem for abductive and consistency-based forms domains. Furthermore, we propose two complete and sound algorithms for finding and ranking all diagnoses and analyze their complexity. We dive deeper and improve the first algorithm's efficiency even more by exploiting past results.Finally, we propose a third algorithm that returns the most likely diagnosis without finding all possible diagnoses, assuming the uncertainty over the observations and the components behavior modes are independents. Experimental evaluation shows that the last algorithm can be very efficient in cases where the diagnosis cardinality, as well as the uncertainty likelihood over the observation, are expected to be small. If, however, all possible diagnoses are desired, then the choice between the first two algorithms depends on whether the domain's diagnosis form is abductive or consistent.

show abstract

Critical observations in model-based diagnosis

Christopher,

Grastien

2024

Artificial Intelligence

View full text Add to dashboard Cite

Formulating event-based critical observations in diagnostic problems

Cited by 5 publications

References 15 publications

Inference of fault signatures of discrete-event systems from event logs

Inference of fault signatures of discrete-event systems from event logs

Model‐based diagnosis with uncertain observations

Critical observations in model-based diagnosis

Contact Info

Product

Resources

About