PhD Seminar: Process mining for trading systems, intentional mining for educational games
Мероприятие завершено
Докладчик: Никитин Константин Иванович, аспирант первого года обучения департамента программной инженерии факультета компьютерных наук.
Доклад: Analysis of Educational Games Using Intentional Mining
Gamification is a common practice nowadays that, however, makes systems more unpredictable and difficult to analyze. The quality measurement of serious game design is usually based on expertise. To formalize this nontrivial task, ludologists have created a number of frameworks that require manual expertise and so are very human-depended. Existing methods of player modeling are not precise enough and do not take into account features of educational projects. To automate the analysis, we suggest using educational process mining techniques based on game logs. According to the literature, goal-oriented process mining is more suitable for non-structured gamified systems. This area is quite new, but can improve interpretability of unstructured models for stakeholders, which is very important in educational games. We suggest that intentional mining could be used to mine Map models that can be semi-automatically translated into ATMSG framework for formal game analysis. In the present talk, reasons of using these techniques and formalism, as well as a description of the suggested method, will be presented.
Докладчик: Карраскель Гамес Хулио Сесар, аспирант первого года обучения департамента программной инженерии факультета компьютерных наук.
Доклад: Analysis of Trading Systems Using Process Mining and Petri Nets
The analysis of trading systems is a task of utmost importance on financial markets. Logs of trading systems are widely used to analyze their internal behavior. In such context, we present in this talk our research on the modeling and validation of trading systems using Process mining and Petri nets.
Fueled by event logs, process mining allows for the analysis of business and software processes. In our research, we focus on trading processes in order books, i.e, where orders to buy or sell securities from participants are submitted and crossed. We will present an approach to extract event logs of these processes from network messages, in particular, Financial Information Exchange (FIX) protocol messages. Thus, the obtained event logs can be leveraged by process mining techniques. For instance, we aim to relate these logs describing the process real behavior against formal models which specify the desired behavior. Hence, we conclude our talk providing a small glimpse on how the expected behavior of trading sessions can be described using formal models, specifically using (high-level) Petri nets - one of the leading formalisms for modeling and analyzing distributed systems. These formal models can be used not only for system diagnosis, but also for simulating complex trading scenarios.
Место: Факультет компьютерных наук, Покровский б, д. 11, ауд. D510.
Время: 11 октября, 16:40 - 18:10
Доклады состоятся в рамках научно-исследовательского семинара аспирантской школы по компьютерным наукам.
Доклад: Analysis of Educational Games Using Intentional Mining
Gamification is a common practice nowadays that, however, makes systems more unpredictable and difficult to analyze. The quality measurement of serious game design is usually based on expertise. To formalize this nontrivial task, ludologists have created a number of frameworks that require manual expertise and so are very human-depended. Existing methods of player modeling are not precise enough and do not take into account features of educational projects. To automate the analysis, we suggest using educational process mining techniques based on game logs. According to the literature, goal-oriented process mining is more suitable for non-structured gamified systems. This area is quite new, but can improve interpretability of unstructured models for stakeholders, which is very important in educational games. We suggest that intentional mining could be used to mine Map models that can be semi-automatically translated into ATMSG framework for formal game analysis. In the present talk, reasons of using these techniques and formalism, as well as a description of the suggested method, will be presented.
Докладчик: Карраскель Гамес Хулио Сесар, аспирант первого года обучения департамента программной инженерии факультета компьютерных наук.
Доклад: Analysis of Trading Systems Using Process Mining and Petri Nets
The analysis of trading systems is a task of utmost importance on financial markets. Logs of trading systems are widely used to analyze their internal behavior. In such context, we present in this talk our research on the modeling and validation of trading systems using Process mining and Petri nets.
Fueled by event logs, process mining allows for the analysis of business and software processes. In our research, we focus on trading processes in order books, i.e, where orders to buy or sell securities from participants are submitted and crossed. We will present an approach to extract event logs of these processes from network messages, in particular, Financial Information Exchange (FIX) protocol messages. Thus, the obtained event logs can be leveraged by process mining techniques. For instance, we aim to relate these logs describing the process real behavior against formal models which specify the desired behavior. Hence, we conclude our talk providing a small glimpse on how the expected behavior of trading sessions can be described using formal models, specifically using (high-level) Petri nets - one of the leading formalisms for modeling and analyzing distributed systems. These formal models can be used not only for system diagnosis, but also for simulating complex trading scenarios.
Место: Факультет компьютерных наук, Покровский б, д. 11, ауд. D510.
Время: 11 октября, 16:40 - 18:10
Доклады состоятся в рамках научно-исследовательского семинара аспирантской школы по компьютерным наукам.
Дата
11 октября, 2019 г. 16:40
Адрес
Покровский бульвар, д. 11, корп. D
В статье упомянуты