A Unified Approach for Measuring Precision and Generalization Based on Anti-alignments

@inproceedings{Dongen2016AUA,
  title={A Unified Approach for Measuring Precision and Generalization Based on Anti-alignments},
  author={Boudewijn F. van Dongen and Josep Carmona and Thomas Chatain},
  booktitle={BPM},
  year={2016}
}
The holy grail in process mining is an algorithm that, given an event log, produces fitting, precise, properly generalizing and simple process models. While there is consensus on the existence of solid metrics for fitness and simplicity, current metrics for precision and generalization have important flaws, which hamper their applicability in a general setting. In this paper, a novel approach to measure precision and generalization is presented, which relies on the notion of anti-alignments. An… 
Alignment-based Quality Metrics in Conformance Checking (Summary)
TLDR
These are metrics for fitness, precision and generalization, based on alignments and the newer concept named anti-alignments, which are presented in detail in this paper.
Anti-Alignments - Measuring The Precision of Process Models and Event Logs
TLDR
The notion of anti-alignment as a concept to help unveiling runs in the model that may deviate significantly from the observed behavior is presented and a new metric for precision is proposed.
Alignment-based Quality Metrics in Conformance Checking
TLDR
These are the first metrics for fitness, precision and generalization, based on alignments and the newer concept named anti-alignments, that are presented in the conformance checking domain.
Measuring Fitness and Precision of Automatically Discovered Process Models: A Principled and Scalable Approach
TLDR
A family of fitness and precision measures based on the idea of comparing the k-th order Markovian abstraction of a process model against that of an event log is presented, showing that this family of measures fulfils the aforementioned properties for suitably chosen values of k.
A Spectrum of Entropy-Based Precision and Recall Measurements Between Partially Matching Designed and Observed Processes
TLDR
A spectrum of conformance measurements results from a generalization of the recently introduced entropy-based approaches for measuring precision and recall between observed process executions and designed process models to provide analysts with flexible control over the sensitivity for identifying commonalities and discrepancies in the compared processes and performance of the techniques.
Abstract-and-Compare: A Family of Scalable Precision Measures for Automated Process Discovery
TLDR
A family of precision measures based on the idea of comparing the k-th order Markovian abstraction of a process model against that of an event log is presented and it is demonstrated that this family of measures fulfils the aforementioned axioms for a suitably chosen value of k.
Abstract and Compare: A Framework for Defining Precision Measures for Automated Process Discovery
TLDR
This paper presents a versatile framework for defining precision measures based on behavior abstractions and shows empirically that different instances of this framework allow us to strike different tradeoffs between scalability and sensitivity.
Behavioural Quotients for Precision and Recall in Process Mining
TLDR
This paper introduces a framework for defining language quotients, instantiate the framework with cardinality-based and entropy-based measures to handle finite and infinite behaviours, and proves important properties of the quotients.
An Experimental Evaluation of the Generalizing Capabilities of Process Discovery Techniques and Black-Box Sequence Models
TLDR
The results indicate that LSTM neural networks more accurately describe previously unseen traces than existing process discovery methods and black-box sequence models, on the basis of next activity prediction.
Conformance checking of partially matching processes: An entropy-based approach
TLDR
The recently introduced precision and recall conformance measures between an event log and process model that are based on the concept of entropy from information theory to account for partially matching processes are extended.
...
1
2
3
4
...

References

SHOWING 1-10 OF 13 REFERENCES
Anti-alignments in Conformance Checking - The Dark Side of Process Models
TLDR
The notion of anti-alignment is presented as a concept to help unveiling traces in the model that may deviate significantly from the observed behavior, and is shown how to express the problem of finding anti-alignments as the satisfiability of a Boolean formula, and provide a tool which can deal with large models efficiently.
Determining Process Model Precision and Generalization with Weighted Artificial Negative Events
TLDR
A novel conformance checking method to measure how well a process model performs in terms of precision and generalization with respect to the actual executions of a process as recorded in an event log is introduced.
Measuring precision of modeled behavior
TLDR
A method to measure precision of process models, given their event logs by first aligning the logs to the models, is proposed, which is not sensitive to non-fitting executions and more accurate values can be obtained for non- fitting logs.
Quality Dimensions in Process Discovery: The Importance of Fitness, Precision, Generalization and Simplicity
TLDR
This paper presents the ETM algorithm which allows the user to seamlessly steer the discovery process based on preferences with respect to the four quality dimensions, and shows that all dimensions are important for process discovery.
Aligning observed and modeled behavior
Aligning Observed and Modeled Behavior The availability of process models and event logs is rapidly increasing as more and more business processes are supported by IT. On the one hand, most
Replaying history on process models for conformance checking and performance analysis
TLDR
The importance of maintaining a proper alignment between event log and process model is elaborated on and their application to conformance checking and performance analysis is elaborated.
Conformance Checking and Diagnosis in Process Mining
  • J. Munoz-Gama
  • Computer Science
    Lecture Notes in Business Information Processing
  • 2016
TLDR
This thesis proposes the use of decomposed techniques in order to aid in checking and diagnosing fitness, and includes a novel technique based on detecting escaping arcs, i.e., points where the modeled behavior deviates from the one reflected in log.
Causal Nets: A Modeling Language Tailored towards Process Discovery
TLDR
This work provides declarative semantics more suitable for process mining, and relates causal nets to Petri nets to clarify these semantics and to illustrate the non-local nature of this new representation.
Conformance checking of processes based on monitoring real behavior
TLDR
An incremental approach to check the conformance of a process model and an event log is proposed and a Conformance Checker has been implemented within the ProM framework and it has been evaluated using artificial and real-life event logs.
Process mining : conformance and extension
TLDR
This dissertation broadens the field of process mining to include the aspect of conformance and extension, and develops several Petri-net based approaches to measure conformance in these dimensions and describes five case studies in which these conformance checking techniques were successfully applied to real and artificial examples.
...
1
2
...