Houshang Darabi

Julian Theis, Ilia Mokhtarian, Houshang Darabi

This paper presents an approach to model an unknown Ladder Logic based Programmable Logic Controller (PLC) program consisting of Boolean logic and counters using Process Mining techniques. First, we tap the inputs and outputs of a PLC to create a data flow log. Second, we propose a method to translate the obtained data flow log to an event log suitable for Process Mining. In a third step, we propose a hybrid Petri net (PN) and neural network approach to approximate the logic of the actual underlying PLC program. We demonstrate the applicability of our proposed approach on a case study with three simulated scenarios.

Samuel Harford, Fazle Karim, Houshang Darabi

Deep learning models utilizing convolution layers have achieved state-of-the-art performance on univariate time series classification tasks. In this work, we propose improving CNN based time series classifiers by utilizing Octave Convolutions (OctConv) to outperform themselves. These network architectures include Fully Convolutional Networks (FCN), Residual Neural Networks (ResNets), LSTM-Fully Convolutional Networks (LSTM-FCN), and Attention LSTM-Fully Convolutional Networks (ALSTM-FCN). The proposed layers significantly improve each of these models with minimally increased network parameters. In this paper, we experimentally show that by substituting convolutions with OctConv, we significantly improve accuracy for time series classification tasks for most of the benchmark datasets. In addition, the updated ALSTM-OctFCN performs statistically the same as the top two time series classifers, TS-CHIEF and HIVE-COTE (both ensemble models). To further explore the impact of the OctConv layers, we perform ablation tests of the augmented model compared to their base model.

Maryam Pishgar, Martha Razo, Julian Theis et al.

Paralytic Ileus (PI) patients are at high risk of death when admitted to the Intensive care unit (ICU), with mortality as high as 40\%. There is minimal research concerning PI patient mortality prediction. There is a need for more accurate prediction modeling for ICU patients diagnosed with PI. This paper demonstrates performance improvements in predicting the mortality of ICU patients diagnosed with PI after 24 hours of being admitted. The proposed framework, PMPI(Process Mining Model to predict mortality of PI patients), is a modification of the work used for prediction of in-hospital mortality for ICU patients with diabetes. PMPI demonstrates similar if not better performance with an Area under the ROC Curve (AUC) score of 0.82 compared to the best results of the existing literature. PMPI uses patient medical history, the time related to the events, and demographic information for prediction. The PMPI prediction framework has the potential to help medical teams in making better decisions for treatment and care for ICU patients with PI to increase their life expectancy.

Julian Theis, Houshang Darabi

Decay Replay Mining is a deep learning method that utilizes process model notations to predict the next event. However, this method does not intertwine the neural network with the structure of the process model to its full extent. This paper proposes an approach to further interlock the process model of Decay Replay Mining with its neural network for next event prediction. The approach uses a masking layer which is initialized based on the reachability graph of the process model. Additionally, modifications to the neural network architecture are proposed to increase the predictive performance. Experimental results demonstrate the value of the approach and underscore the importance of discovering precise and generalized process models.

Julian Theis, Ilia Mokhtarian, Houshang Darabi

Process mining algorithms discover a process model from an event log. The resulting process model is supposed to describe all possible event sequences of the underlying system. Generalization is a process model quality dimension of interest. A generalization metric should quantify the extent to which a process model represents the observed event sequences contained in the event log and the unobserved event sequences of the system. Most of the available metrics in the literature cannot properly quantify the generalization of a process model. A recently published method [1] called Adversarial System Variant Approximation leverages Generative Adversarial Networks to approximate the underlying event sequence distribution of a system from an event log. While this method demonstrated performance gains over existing methods in measuring the generalization of process models, its experimental evaluations have been performed under ideal conditions. This paper experimentally investigates the performance of Adversarial System Variant Approximation under non-ideal conditions such as biased and limited event logs. Moreover, experiments are performed to investigate the originally proposed sampling hyperparameter value of the method on its performance to measure the generalization. The results confirm the need to raise awareness about the working conditions of the Adversarial System Variant Approximation method. The outcomes of this paper also serve to initiate future research directions. [1] Theis, Julian, and Houshang Darabi. "Adversarial System Variant Approximation to Quantify Process Model Generalization." IEEE Access 8 (2020): 194410-194427.

Samuel Harford, Fazle Karim, Houshang Darabi

Classification models for the multivariate time series have gained significant importance in the research community, but not much research has been done on generating adversarial samples for these models. Such samples of adversaries could become a security concern. In this paper, we propose transforming the existing adversarial transformation network (ATN) on a distilled model to attack various multivariate time series classification models. The proposed attack on the classification model utilizes a distilled model as a surrogate that mimics the behavior of the attacked classical multivariate time series classification models. The proposed methodology is tested onto 1-Nearest Neighbor Dynamic Time Warping (1-NN DTW) and a Fully Convolutional Network (FCN), all of which are trained on 18 University of East Anglia (UEA) and University of California Riverside (UCR) datasets. We show both models were susceptible to attacks on all 18 datasets. To the best of our knowledge, adversarial attacks have only been conducted in the domain of univariate time series and have not been conducted on multivariate time series. such an attack on time series classification models has never been done before. Additionally, we recommend future researchers that develop time series classification models to incorporating adversarial data samples into their training data sets to improve resilience on adversarial samples and to consider model robustness as an evaluative metric.

Julian Theis, Houshang Darabi

In process mining, process models are extracted from event logs using process discovery algorithms and are commonly assessed using multiple quality dimensions. While the metrics that measure the relationship of an extracted process model to its event log are well-studied, quantifying the level by which a process model can describe the unobserved behavior of its underlying system falls short in the literature. In this paper, a novel deep learning-based methodology called Adversarial System Variant Approximation (AVATAR) is proposed to overcome this issue. Sequence Generative Adversarial Networks are trained on the variants contained in an event log with the intention to approximate the underlying variant distribution of the system behavior. Unobserved realistic variants are sampled either directly from the Sequence Generative Adversarial Network or by leveraging the Metropolis-Hastings algorithm. The degree by which a process model relates to its underlying unknown system behavior is then quantified based on the realistic observed and estimated unobserved variants using established process model quality metrics. Significant performance improvements in revealing realistic unobserved variants are demonstrated in a controlled experiment on 15 ground truth systems. Additionally, the proposed methodology is experimentally tested and evaluated to quantify the generalization of 60 discovered process models with respect to their systems.

Ashkan Sharabiani, Adam Bress, William Galanter et al.

Determining the optimal initial dose for warfarin is a critically important task. Several factors have an impact on the therapeutic dose for individual patients, such as patients' physical attributes (Age, Height, etc.), medication profile, co-morbidities, and metabolic genotypes (CYP2C9 and VKORC1). These wide range factors influencing therapeutic dose, create a complex environment for clinicians to determine the optimal initial dose. Using a sample of 4,237 patients, we have proposed a companion classification model to one of the most popular dosing algorithms (International Warfarin Pharmacogenetics Consortium (IWPC) clinical model), which identifies the appropriate cohort of patients for applying this model. The proposed model functions as a clinical decision support system which assists clinicians in dosing. We have developed a classification model using Support Vector Machines, with a polynomial kernel function to determine if applying the dose prediction model is appropriate for a given patient. The IWPC clinical model will only be used if the patient is classified as "Safe for model". By using the proposed methodology, the dosing mode's prediction accuracy increases by 15 percent in terms of Root Mean Squared Error and 17 percent in terms of Mean Absolute Error in dose estimates of patients classified as "Safe for model".

Julian Theis, Houshang Darabi

In complex processes, various events can happen in different sequences. The prediction of the next event given an a-priori process state is of importance in such processes. Recent methods have proposed deep learning techniques such as recurrent neural networks, developed on raw event logs, to predict the next event from a process state. However, such deep learning models by themselves lack a clear representation of the process states. At the same time, recent methods have neglected the time feature of event instances. In this paper, we take advantage of Petri nets as a powerful tool in modeling complex process behaviors considering time as an elemental variable. We propose an approach which starts from a Petri net process model constructed by a process mining algorithm. We enhance the Petri net model with time decay functions to create continuous process state samples. Finally, we use these samples in combination with discrete token movement counters and Petri net markings to train a deep learning model that predicts the next event. We demonstrate significant performance improvements and outperform the state-of-the-art methods on nine real-world benchmark event logs.

Fazle Karim, Somshubra Majumdar, Houshang Darabi

Long Short Term Memory Fully Convolutional Neural Networks (LSTM-FCN) and Attention LSTM-FCN (ALSTM-FCN) have shown to achieve state-of-the-art performance on the task of classifying time series signals on the old University of California-Riverside (UCR) time series repository. However, there has been no study on why LSTM-FCN and ALSTM-FCN perform well. In this paper, we perform a series of ablation tests (3627 experiments) on LSTM-FCN and ALSTM-FCN to provide a better understanding of the model and each of its sub-module. Results from the ablation tests on ALSTM-FCN and LSTM-FCN show that the LSTM and the FCN blocks perform better when applied in a conjoined manner. Two z-normalizing techniques, z-normalizing each sample independently and z-normalizing the whole dataset, are compared using a Wilcoxson signed-rank test to show a statistical difference in performance. In addition, we provide an understanding of the impact dimension shuffle has on LSTM-FCN by comparing its performance with LSTM-FCN when no dimension shuffle is applied. Finally, we demonstrate the performance of the LSTM-FCN when the LSTM block is replaced by a GRU, basic RNN, and Dense Block.

Fazle Karim, Somshubra Majumdar, Houshang Darabi

Time series classification models have been garnering significant importance in the research community. However, not much research has been done on generating adversarial samples for these models. These adversarial samples can become a security concern. In this paper, we propose utilizing an adversarial transformation network (ATN) on a distilled model to attack various time series classification models. The proposed attack on the classification model utilizes a distilled model as a surrogate that mimics the behavior of the attacked classical time series classification models. Our proposed methodology is applied onto 1-Nearest Neighbor Dynamic Time Warping (1-NN ) DTW, a Fully Connected Network and a Fully Convolutional Network (FCN), all of which are trained on 42 University of California Riverside (UCR) datasets. In this paper, we show both models were susceptible to attacks on all 42 datasets. To the best of our knowledge, such an attack on time series classification models has never been done before. Finally, we recommend future researchers that develop time series classification models to incorporating adversarial data samples into their training data sets to improve resilience on adversarial samples and to consider model robustness as an evaluative metric.

Julian Theis, Houshang Darabi

Process Mining is a famous technique which is frequently applied to Software Development Processes, while being neglected in Human-Computer Interaction (HCI) recommendation applications. Organizations usually train employees to interact with required IT systems. Often, employees, or users in general, develop their own strategies for solving repetitive tasks and processes. However, organizations find it hard to detect whether employees interact efficiently with IT systems or not. Hence, we have developed a method which detects inefficient behavior assuming that at least one optimal HCI strategy is known. This method provides recommendations to gradually adapt users' behavior towards the optimal way of interaction considering satisfaction of users. Based on users' behavior logs tracked by a Java application suitable for multi-application and multi-instance environments, we demonstrate the applicability for a specific task in a common Windows environment utilizing realistic simulated behaviors of users.

Maryam Pishgar, Fazle Karim, Somshubra Majumdar et al.

Vocal disorders have affected several patients all over the world. Due to the inherent difficulty of diagnosing vocal disorders without sophisticated equipment and trained personnel, a number of patients remain undiagnosed. To alleviate the monetary cost of diagnosis, there has been a recent growth in the use of data analysis to accurately detect and diagnose individuals for a fraction of the cost. We propose a cheap, efficient and accurate model to diagnose whether a patient suffers from one of three vocal disorders on the FEMH 2018 challenge.

Fazle Karim, Somshubra Majumdar, Houshang Darabi et al.

Over the past decade, multivariate time series classification has received great attention. We propose transforming the existing univariate time series classification models, the Long Short Term Memory Fully Convolutional Network (LSTM-FCN) and Attention LSTM-FCN (ALSTM-FCN), into a multivariate time series classification model by augmenting the fully convolutional block with a squeeze-and-excitation block to further improve accuracy. Our proposed models outperform most state-of-the-art models while requiring minimum preprocessing. The proposed models work efficiently on various complex multivariate time series classification tasks such as activity recognition or action recognition. Furthermore, the proposed models are highly efficient at test time and small enough to deploy on memory constrained systems.

Fazle Karim, Somshubra Majumdar, Houshang Darabi et al.

Fully convolutional neural networks (FCN) have been shown to achieve state-of-the-art performance on the task of classifying time series sequences. We propose the augmentation of fully convolutional networks with long short term memory recurrent neural network (LSTM RNN) sub-modules for time series classification. Our proposed models significantly enhance the performance of fully convolutional networks with a nominal increase in model size and require minimal preprocessing of the dataset. The proposed Long Short Term Memory Fully Convolutional Network (LSTM-FCN) achieves state-of-the-art performance compared to others. We also explore the usage of attention mechanism to improve time series classification with the Attention Long Short Term Memory Fully Convolutional Network (ALSTM-FCN). Utilization of the attention mechanism allows one to visualize the decision process of the LSTM cell. Furthermore, we propose fine-tuning as a method to enhance the performance of trained models. An overall analysis of the performance of our model is provided and compared to other techniques.