Our Ph.D. student, Michael Arias, offered the presentation “Data Science and Process Mining” in the XXIV Congreso Nacional de Estudiantes de Ingenería de Sistemas y Computación (CONEISC 2016) in Pucallpa, Peru. During his visit Michael talk about the ideas behind the link between data science and process mining, available process mining tools, and also share some details about his research in the resource allocation topic.
The Process Mining UC research group congratulates Michael Arias, Jorge Munoz-Gama and Marcos Sepúlveda, for their accepted paper in the 1st workshop on Resource Management in Business Processes 2016 (ReMa’16), organized in conjunction with 14th International Conference on Business Process Management (BPM 2016), which will be held in the city of Rio de Janeiro between September 18th and 22nd of 2016.
The name of the accepted article is “A Multi-criteria Approach for Team Recommendation”.
General framework for Team Resource Recommendation
This work presents a novel approach that recommends the most suitable teams to execute a request defined at run-time. The approach considers four elements: (i) a resource request characterization, (ii) historical information on the process execution and expertise information, (iii) different metrics which calculate the suitability of the work teams taking into account both individual performance as well as collective performance of the resources, and (iv) a recommender system based on the Best Position Algorithm (BPA2) to obtain a ranking for the recommended work teams.
The Process Mining UC research group congratulates Michael Arias, Eric Rojas, Jonathan Lee, Jorge Munoz-Gama and Marcos Sepúlveda, for their accepted Demo paper in the 14th International Conference on Business Process Management (BPM2016), which will be held in the city of Rio de Janeiro between September 18th and 22nd of 2016.
The work presents the tool ResRec, a novel Multi-factor Criteria tool that can be used to recommend and allocate resources dynamically. The tool is a decision maker-oriented approach that provides the feature of solving individual requests (On-demand), or requests made in blocks (Batch) through a recommender system developed in ProM.
The authors must present their research and a poster at the at the BPM 2016 Demo Track.
Tool available here.
The group TECHNOLOGIES FOR DIGITAL LEARNING LAB congratulates Ronald Perez and Jorge Maldonado, who presented a publication in the 42nd Latin American Informatics Conference, CLEI 2016, which will be held in the city of Valparaiso between October 10th and October 14th of 2016.
The name of the accepted article is “How to design tools for supporting self-regulated learning in MOOCs? Lessons learned from a literature review from 2008 and 2016”.
This paper presents a systematic literature review that examines and analyzes the articles from 2008 until 2016 that have addressed the development of tools to support Self-Regulated Learning (SRL) in online and MOOC environments. The findings denote that: (1) there is a lack of tools to support SRL in MOOC environments; (2) the evaluation of the existing tools are not aligned whit the objectives of the research; (3) current research presents proposal of tools but very few achieve the stage of implementation; and (4) current existing tools tend to support many SRL strategies at the same time. Finally, it ends with a set of lessons learned for guiding the implementation of tools to support SRL strategies in MOOCs environments.
The authors Ronald Pérez (Department of Computer Science, Pontificia Universidad Católica de Chile, Chile and Universidad de Costa Rica, Headquarters Pacific, Costa Rica), Mar Pérez-Sanagustín (Department of Computer Science, Pontificia Universidad Católica de Chile, Chile) and Jorge Maldonado (Department of Computer Science, Pontificia Universidad Católica de Chile, Chile and Department of Computer Science, Universidad de Cuenca, Ecuador) must present their research at the conference in order to receive the feedback necessary within this event.
Pedro Rossel, professor from the Universidad Católica de la Santísima Concepción, visited DCC UC – HumaLab in July 2016 to continue work on applying software engineering concepts to collaborative work.
Recent joint work includes: Creating a family of collaborative applications for emergency management in the firefighting sub-domain. Rossel, Herskovic, Ormeño. Information Systems Frontiers 18(1), 2016.
Flux Capacitor is the blog of Fluxicon, the company behind Disco, one of the most widely used comercial tools for Process Mining. Since 2009, Flux Capacitor has published dozens of articles of high usefulness for both practitioners and researchers in Process Mining: practical tips, discussions, tutorials, case studies, interviews, … But, what is the most useful article for a process mining data scientist?
In order to answer this question, the ProcessMiningUC group, the research group in Process Mining of the Pontificia Universidad Católica de Chile (UC), has elaborated a ranking based on the opinion of 15 business process management and process mining experts. Each expert evaluated (from 1 to 15 points) the 15 most useful articles for a process scientist according to their criteria, among all the articles published in 7 years.
According to the ranking the most useful article is Why Process Mining is Ideal for Data Scientists. The article, with 136 points and a considerable distance with the second position, presents a brief description of the field and provides pointers to additional material. The second article, Change in Perspective with Process Mining, presents the capacities of process mining to obtain a process perspective on your data, while the third, How is Process Mining different from, highlights the differences between process mining and other areas such as Data Mining, BPM, Lean Six Sigma, Business Intelligence, or Simulation.
The complete ranking is the following:
The article “Does taking a MOOC as a complement for remedial courses have an effect on my learning outcomes? A pilot study on calculus” written by Mar Pérez-Sanagustín, Josefina Hernández, Claudio Gelmi, Isabel Hilliger and Fernanda Rodríguez will be published in EC-TEL 2016. This paper presents the results of a pilot study about students’ adoption and learning outcomes of 4 MOOCs proposed as a complementary resource for traditional remedial courses on calculus.
The data analysis shows that up to 16% of the students were active in the MOOCs under study, mostly during the days before taking the diagnostic exam that preceded the traditional face-to-face remedial courses. We observe that active students had more chances of passing the diagnostic exam and skipping the required remedial courses. However, we found no significant differences on the remedial course exam scores between the students that were active in the MOOCs and those that were not. These findings suggest that MOOCs are a good solution to strengthen-ing skills and reviewing concepts, but that more guidance is needed when used as a complement to traditional f2f courses.
The article is currently in press.
With the aim to provide feedback on three process mining projects related to healthcare and higher education, postdoctoral research assistant in Computer Science at the BPM Research Cluster of the University of Innsbruck, Andrea Burattin, visited UC Department of Computer Science at Catholic University of Chile (DCC UC).
Andrea Burattin was invited by DCC UC academics, Marcos Sepúlveda, Valeria Herskovic and Jorge Munoz-Gama, in the context of the project, “Analysis of multidisciplinary collaboration in primary healthcare using process mining”, funded by the Chilean National Fund for Scientific and Technological Development, Fondecyt.
“The idea is to see whether we can use process mining techniques to provide useful insights on these different domains. We saw some several techniques than can already be applied to these logs and we are expecting good outcomes that would be a contribution, not for Process Mining as a discipline, but for domains as higher education and healthcare”, said Andrea Burattin about the projects he will work on: “Process Mining in Higher Education” and “Answering Frequently Posed Questions in Emergency Room through Process Mining”.
During his visit, Andrea Burattin, who was awarded in 2014 by IEEE Task Force on Process Mining with the Best Process Mining Dissertation for the doctoral thesis, “Applicability of Process Mining Techniques in Business Environments”, offered the presentation “Process Mining Techniques in Business Environments” to explain the ideas behind his research area in Computer Science.
“Applied science open problems are different from the theoretical ones. In fact, they only partially overlap each other. With this work I tried to demonstrate that if you want a technique to be successful you need this technique to be also applicable in the real world”, said postdoctoral researcher, Andrea Burattin.
After his talk at UC Department of Computer Science at Catholic University of Chile (DCC UC), Andrea Burattin attended presentations by doctoral researchers and answered questions from first and second year students. “Their doubts were less technical, but reflected a lot of interest. The impression I had from them was positive and it was nice to give this talk”, assured Andrea Burattin.
In addition to his doctoral thesis developed into the Springer book “Process Mining Techniques in Business Environments”, Andrea Burattin has published 27 articles about Process Mining, Business Process Randomization and Simulation; and Declare Models (analysis and discovery) techniques and methodologies. He is a member of IEEE Task Force on Process Mining since 2009, and of IEEE since 2010.
Read more in here.
The article “Process Mining in Healthcare: A Literature Review” by Eric Rojas, Jorge Munoz-Gama, Marcos Sepúlveda and Daniel Capurro has been published in the Journal of Biomedical Informatics. The paper includes an analysis of the case studies according to eleven main aspects, including: process and data types; frequently posed questions; process mining techniques, perspectives and tools; methodologies; implementation and analysis strategies; geographical analysis; and medical fields. It also includes emerging topics and future trends in this field.
This review can provide a useful overview of the current work being undertaken in this field; help researchers to choose process mining algorithms, techniques, tools, methodologies, and approaches for their own applications; identify accomplishments and limitations in the case studies; and highlight the use of process mining to improve healthcare processes.