More Like this

1. Kim, I., Frank, K., Wolf, J.R., & Pentland, B.T. (2021) “Predicting Process Structure after a Disruption: An Example from the Clinical Documentation Process”

   Kim, I.; Frank, K.; Wolf, J.R.; Pentland, B.T. (January 2021, BPM Routine Dynamics Workshop)

   null (Ed.)

   Using data from the audit trail of an electronic medical record system, we examine the effects of a disruption on the clinical documentation process. We use process mining to construct a network that describes the process and then we use a latent factor selection model to analyze changes to that network. Rather than attempting to discover a particular process model, our goal is to identify theory-based factors that explain change and stability in the overall pattern of actions. We conduct the analysis at two levels of granularity and we compare time periods with and without disruption. The paper contributes to current research on routine dynamics as network dy-namics by demonstrating the use of network science to predict the structure of an organizational routine. 

   more » « less
2. Improving Process Discovery Results by Filtering Out Outliers from Event Logs with Hidden Markov Models

   https://doi.org/10.1109/CBI52690.2021.00028  

   Zhang, Zhenyu; Hildebrant, Ryan; Asgarinejad, Fatemeh; Venkatasubramanian, Nalini; Ren, Shangping (September 2021, 2021 IEEE 23rd Conference on Business Informatics (CBI))

   Process Mining is a technique for extracting process models from event logs. Event logs contain abundant explicit information related to events, such as the timestamp and the actions that trigger the event. Much of the existing process mining research has focused on discovering the process models behind these event logs. However, Process Mining relies on the assumption that these event logs contain accurate representations of an ideal set of processes. These ideal sets of processes imply that the information contained within the log represents what is really happening in a given environment. However, many of these event logs might contain noisy, infrequent, missing, or false process information that is generally classified as outliers. Extending beyond process discovery, there are many research efforts towards cleaning the event logs to deal with these outliers. In this paper, we present an approach that uses hidden Markov models to filter out outliers from event logs prior to applying any process discovery algorithms. Our proposed filtering approach can detect outlier behavior, and consequently, help process discovery algorithms return models that better reflect the real processes within an organization. Furthermore, we show that this filtering method outperforms two commonly used filtering approaches, namely the Matrix Filter approach and the Anomaly Free Automation approach for both artificial event logs and real-life event logs. 

   more » « less
3. A comprehensive approach to analyzing community dynamics using rank abundance curves

   https://doi.org/10.1002/ecs2.2881  

   Avolio, Meghan L.; Carroll, Ian T.; Collins, Scott L.; Houseman, Gregory R.; Hallett, Lauren M.; Isbell, Forest; Koerner, Sally E.; Komatsu, Kimberly J.; Smith, Melinda D.; Wilcox, Kevin R. (October 2019, Ecosphere)

   Abstract Univariate and multivariate methods are commonly used to explore the spatial and temporal dynamics of ecological communities, but each has limitations, including oversimplification or abstraction of communities. Rank abundance curves (RACs) potentially integrate these existing methodologies by detailing species‐level community changes. Here, we had three goals: first, to simplify analysis of community dynamics by developing a coordinated set of R functions, and second, to demystify the relationships among univariate, multivariate, andRACs measures, and examine how each is influenced by the community parameters as well as data collection methods. We developed new functions for studying temporal changes and spatial differences in RACs in an update to the R package library(“codyn”), alongside other new functions to calculate univariate and multivariate measures of community dynamics. We also developed a new approach to studying changes in the shape ofRACcurves. The R package update presented here increases the accessibility of univariate and multivariate measures of community change over time and difference over space. Next, we use simulated and real data to assess theRACand multivariate measures that are output from our new functions, studying (1) if they are influenced by species richness and evenness, temporal turnover, and spatial variability and (2) how the measures are related to each other. Lastly, we explore the use of the measures with an example from a long‐term nutrient addition experiment. We find that theRACand multivariate measures are not sensitive to species richness and evenness and that all the measures detail unique aspects of temporal change or spatial differences. We also find that species reordering is the strongest correlate of a multivariate measure of compositional change and explains most community change observed in long‐term nutrient addition experiment. Overall, we show that species reordering is potentially an understudied determinant of community changes over time or differences between treatments. The functions developed here should enhance the use of RACs to further explore the dynamics of ecological communities. 

   more » « less
4. Effects of Concurrency in Complex Service Organizations: Evidence from Electronic Health Records

   https://doi.org/10.1007/978-3-031-25383-6_12  

   Pentland, B.T.; Kim, I.; Zhang, Q.; Ryan Wolf, J (January 2023, Business Process Management Workshops. BPM 2022. Lecture Notes in Business Information Processing)

   We use Kremser and Blagoev’s [1] role-routine ecology to theorize about the effects of concurrency in complex service organizations, such as outpatient medical clinics. In a typical clinic, teams of specialized individuals serve multiple clients at the same time. There can be concurrency within a patient visit (a technician may be preparing for a procedure while the doctor talks to the patient) and concurrency between patient visits (multiple patients being treated in the clinic). Using data from electronic health records, we estimate the effects of concurrency within and between patient visits on the duration of patient visits in a set of dermatology clinics. As expected, we find that concurrency within patient visits is associated with reduced duration, while concurrency between visits is associated with increased duration. We discuss the implication of these findings for process mining and discovery of process models in organizations where process instances are not independent. 

   more » « less
5. Studying Affect Dynamics and Chronometry Using Sensor-Free Detectors

   Botelho, A. F.; Baker, R. S.; Ocumpaugh, J.; Heffernan, N. T. (January 2018, Proceedings of the 11th International Conference on Educational Data Mining)

   Student affect has been found to correlate with short- and long-term learning outcomes, including college attendance as well as interest and involvement in Science, Technology, Engineering, and Mathematics (STEM) careers. However, there still remain significant questions about the processes by which affect shifts and develops during the learning process. Much of this research can be split into affect dynamics, the study of the temporal transitions between affective states, and affective chronometry, the study of how an affect state emerges and dissipates over time. Thus far, these affective processes have been primarily studied using field observations, sensors, or student self-report measures; however, these approaches can be coarse, and obtaining finer grained data produces challenges to data fidelity. Recent developments in sensor-free detectors of student affect, utilizing only the data from student interactions with a computer based learning platform, open an opportunity to study affect dynamics and chronometry at moment-to-moment levels of granularity. This work presents a novel approach, applying sensor-free detectors to study these two prominent problems in affective research. 

   more » « less