Szymon Bobek

Szymon Bobek
Jagiellonian University | UJ · Faculty of Physics, Astronomy and Applied Computer Science

PhD

About

66
Publications
23,425
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
537
Citations
Introduction
I am a member of the GEIST research team at the AGH University of Science and Technology. I participated in several international research projects where I was involved in tasks concerning knowledge engineering and rule based inference. My current reaserch interests are: machine learning, ambient intelligence, context awareness and knowledge engineering.
Additional affiliations
December 2016 - present
AGH University of Science and Technology in Kraków
Position
  • Professor (Assistant)
Description
  • Research in areas of knowledge engineering, machine learning, ambient intelligence and context awareness.
September 2009 - November 2016
AGH University of Science and Technology in Kraków
Position
  • Research Assistant
Description
  • Research in areas of knowledge engineering, machine learning, ambient intelligence and context awareness.
AGH University of Science and Technology in Kraków
Position
  • Professor (Assistant)
Education
October 2013 - July 2016

Publications

Publications (66)
Article
Uncertainty handling is one of the most important aspects of modelling of context-aware systems. It has direct impact on the adaptability, understood as an ability of the system to adjust to changing environmental conditions or hardware configuration (missing data), changing user habits (ambiguous concepts), or imperfect information (low quality se...
Article
Building systems that acquire, process and reason with context data is a major challenge. Model updates and modifications are required for the mobile context-aware systems. Additionally, the nature of the sensor-based systems implies that the data required for the reasoning is not always available nor it is certain. Finally, the amount of context d...
Conference Paper
Full-text available
In this paper we introduce a new method to incorporate the user as an additional information source for the purpose of indoor localization. Therefore, the user is interrogated about certain characteristics in his/her environment. The questions are generated by a knowledge-based system built on ontologies. We provide a new statistical model to evalu...
Article
Full-text available
With the rapid evolution of mobile devices, the concept of context aware applications has gained a remarkable popularity in recent years. Smartphones and tablets are equipped with a variety of sensors including accelerometers, gyroscopes, and GPS, pressure gauges, light and GPS sensors. Additionally, the devices become computationally powerful whic...
Conference Paper
Full-text available
Context-aware systems gained huge popularity in recent years due to rapid evolution of personal mobile devices. Equipped with variety of sensors, such devices are sources of a lot of valuable information that allows the system to act in an intelligent way. However, the certainty and presence of this information may depend on many factors like measu...
Article
Full-text available
Generic emotion prediction models based on physiological data developed in the field of affective computing apparently are not robust enough. To improve their effectiveness, one needs to personalize them to specific individuals and incorporate broader contextual information. To address the lack of relevant datasets, we propose the 2nd Study in Bio-...
Article
Full-text available
Development of predictive maintenance (PdM) solutions is one of the key aspects of Industry 4.0. In recent years, more attention has been paid to data-driven techniques, which use machine learning to monitor the health of an industrial asset. The major issue in the implementation of PdM models is a lack of good quality labelled data. In the paper w...
Article
Full-text available
We present a refinement of the Immersive Parallel Coordinates Plots (IPCP) system for Virtual Reality (VR). The evolved system provides data-science analytics built around a well-known method for visualization of multidimensional datasets in VR. The data-science analytics enhancements consist of importance analysis and a number of clustering algori...
Preprint
Pattern discovery in multidimensional data sets has been a subject of research since decades. There exists a wide spectrum of clustering algorithms that can be used for that purpose. However, their practical applications share in common the post-clustering phase, which concerns expert-based interpretation and analysis of the obtained results. We ar...
Conference Paper
Explainable Artificial Intelligence (XAI) aims at introducing transparency and intelligibility into decision-making process of AI systems. Most of the work in this area is focused on supervised machine learning tasks such as classification and regression. Unsupervised algorithms such as clustering can also be explained with existing approaches. Thi...
Conference Paper
Full-text available
The paper provides insights into two main threads of analysis of the BIRAFFE2 dataset concerning the associations between personality and physiological signals and concerning the game logs' generation and processing. Alongside the presentation of results, we propose the generation of event-marked maps as an important step in the exploratory analysi...
Article
Full-text available
Mining ubiquitous sensing data is important but also challenging, due to many factors, such as heterogeneous large-scale data that is often at various levels of abstraction. This also relates particularly to the important aspects of the explainability and interpretability of the applied models and their results, and thus ultimately to the outcome o...
Chapter
Full-text available
With advances of artificial intelligence (AI), there is a growing need for provisioning of transparency and accountability to AI systems. These properties can be achieved with eXplainable AI (XAI) methods, extensively developed over the last few years with relation for machine learning (ML) models. However, the practical usage of XAI is limited now...
Chapter
Full-text available
Learning from uncertain or incomplete data is one of the major challenges in building artificial intelligence systems. However, the research in this area is more focused on the impact of uncertainty on the algorithms performance or robustness, rather than on human understanding of the model and the explainability of the system. In this paper we pre...
Chapter
Full-text available
Explainable Artificial Intelligence (XAI) methods form a large portfolio of different frameworks and algorithms. Although the main goal of all of explanation methods is to provide an insight into the decision process of AI system, their underlying mechanisms may differ. This may result in very different explanations for the same tasks. In this work...
Chapter
Cluster discovery from highly-dimensional data is a challenging task, that has been studied for years in the fields of data mining and machine learning. Most of them focus on automation of the process, resulting in the clusters that once discovered have to be carefully analyzed to assign semantics for numerical labels. However, it is often the case...
Article
Full-text available
In this article, we propose using personality assessment as a way to adapt affective intelligent systems. This psychologically-grounded mechanism will divide users into groups that differ in their reactions to affective stimuli for which the behaviour of the system can be adjusted. In order to verify the hypotheses, we conducted an experiment on 20...
Conference Paper
Full-text available
We propose an experimental framework for Affective Computing based of video games. We developed a set of specially designed mini-games, based of carefully selected game mechanics, to evoke emotions of participants of a larger experiment. We believe, that games provide a controllable yet overall ecological environment for studying emotions. We discu...
Chapter
In the paper we describe the industrial process of hot rolling of steel. In cooperation with ArcelorMittal Poland we consider a specific fully automated production line. While it is equipped with a number of industrial sensors, the acquired data has only been analyzed on a basic statistical level, mainly for reporting. In the paper we outline oppor...
Chapter
Development of models for emotion detection is often based on the use of machine learning. However, it poses practical challenges, due to the limited understanding of modeling of emotions, as well as the problems regarding measurements of bodily signals. In this paper we report on our recent work on improving such models, by the use of explainable...
Preprint
Full-text available
The paper describes BIRAFFE2 data set, which is a result of an affective computing experiment conducted between 2019 and 2020, that aimed to develop computer models for classification and recognition of emotion. Such work is important to develop new methods of natural Human-AI interaction. As we believe that models of emotion should be personalized...
Data
This is our 2nd Study in Bio-Reactions and Faces for Emotion-based Personalization for AI Systems (BIRAFFE2). It is a dataset consisting of electrocardiogram (ECG), galvanic skin response (GSR), changes in facial expression signals and hand movements (represented by gamepad's accelerometer and gyroscope) recorded during affect elicitation by means...
Conference Paper
Full-text available
In this paper we introduce the BIRAFFE data set which is the result of the experiment in affective computing we conducted in early 2019. The experiment is part of the work aimed at the development of computer models for emotion classification and recognition. We strongly believe that such models should be personalized by design as emotional respons...
Data
We present BIRAFFE, a dataset consisting of electrocardiogram (ECG), galvanic skin reaction (GSR) and changes in facial expression signals recorded during affect elicitation by means of audio-visual stimuli (from IADS and IAPS databases) and our two proof-of-concept affective games ("Affective SpaceShooter 2" and "Fred Me Out 2"). All the signals w...
Article
Full-text available
In this paper, we consider the use of wearable sensors for providing affect-based adaptation in Ambient Intelligence (AmI) systems. We begin with discussion of selected issues regarding the applications of affective computing techniques. We describe our experiments for affect change detection with a range of wearable devices, such as wristbands and...
Chapter
Affective computing gained a lot of attention from researchers and business over the last decade. However, most of the attempts for building systems that try to predict, or provoke affective state of users were done for specific and narrow domains. This complicates reusing such systems in other, even similar domains. In this paper we present such a...
Article
Full-text available
In our work, we focus on detection of affective states, their proper identification and interpretation with use of wearable and mobile devices. We propose a data acquisition layer based on wearable devices able to gather physiological data, and we integrate it with mobile context-aware framework. Furthermore, we formulate a method for personalizati...
Article
Full-text available
Building mobile context‐aware systems is inherently complex and non‐trivial task. It consists of several phases starting from acquisition of context, through modeling to execution of contextual models. Today, such systems are mostly implemented on mobile platforms, that introduce specific requirements, such as intelligibility, robustness, privacy,...
Chapter
Knowledge discovery from data streams in recent years become one of the most important research area in a domain of data science. This is mainly due to the rapid development of mobile devices, and Internet of things solutions which allow for obtaining petabytes of data within minutes. All of the modern approaches either use representation that is f...
Chapter
Over the last decades, number of embedded and portable computer systems for monitoring of activities of miners and underground environmental conditions that have been developed has increased. However, their potential in terms of computing power and analytic capabilities is still underestimated. In this paper we elaborate on the recent examples of t...
Conference Paper
Full-text available
The paper outlines a mobile sensor platform aimed at processing physiological data from wearable sensors. We discuss the requirements related to the use of low-cost portable devices in this scenario. Experimental analysis of four such devices, namely Microsoft Band 2, Empatica E4, eHealth Sensor Platform and BITalino (r)evolution is provided. Criti...
Conference Paper
Full-text available
In this overview paper we focus on our recent progress in the work on the mobile platform for AfC. We provide the main assumptions about the platform, as well as describe affective data acquisition and interpretation. We discuss our most recent experiments and provide an outlook of our future works.
Conference Paper
Full-text available
Design of Business Intelligence systems capable of effectively handling a domain knowledge is a well known, but currently not solved challenge for both Software and Knowledge Engineers. There exist several approaches to extract and model the Business Knowledge, most notably Business Processes and Business Rules. However, each of them has its own we...
Conference Paper
Rule-based systems constitute the state of the art solutions in the area of artificial intelligence. They provide fast, human readable and self explanatory mechanism for encoding knowledge. Due to large popularity of rules, dozens of inference engines were developed over last few decades. They differ in the reasoning efficiency depending on many fa...
Conference Paper
Knowledge engineering aims at providing methods for efficient knowledge encoding to allow for automatic reasoning. Most of the research in this field is devoted to the design of expressive modeling languages or effective reasoning mechanisms. We argue that powerful knowledge representation and inference mechanism is not enough to assure high qualit...
Conference Paper
Mobile context-aware systems are becoming more and more popular due to the rapid evolution of personal mobile devices. The variety of sensors that are available on such devices allow building intelligent applications that adapt automatically to user preferences and needs. Together with a growth of such self-adaptable systems, number of tools for co...
Conference Paper
Full-text available
Abstract. Despite a large amount of research devoted to improving meta-learning techniques, providing and using background knowledge for this task remains a challenge. In this paper we propose a mechanism for automatic recommendation of suitable machine learning algorithms and their parameters. We used OpenML database and use rule-based configu...
Conference Paper
Location is one of the most valuable and extensively used information in mobile context-aware systems. Its understanding may vary from geolocation that uses GPS infrastructure to locate objects on Earth, up to microlocation, which aims at locating users and objects inside closed areas. Although geolocation can be considered as a mature field, there...
Article
Mobile context-aware systems gained huge popularity in recent years due to the rapid evolution of personal mobile devices. Nowadays smartphones are equipped with a variety of sensors that allow for on-line monitoring of user context and reasoning upon it. Contextual information in such systems is very dynamic. It changes rapidly and these changes m...
Article
Location is one of the most commonly used contextual information in mobile context-aware systems. It can be considered on many different levels of granularity, varying from geolocation that is based on GPS systems, up to microlocation that uses Bluetooth Low Energy devices and WiFi access points for locating users inside buildings. Most common use...
Conference Paper
Full-text available
Context-aware systems make use of contextual information to adapt their functionality to current environment state, or user needs and habits. One of the major problems concerning them is the fact, that there is no war-ranty that the contextual information will be available, nor certain at the time when the reasoning should be performed. This may be...
Conference Paper
Research in the area of context-awareness has recently been revolutionized by the rapid development of mobile devices like smart phones and tablets, which became omnipresent in daily human life. Such devices are valuable sources of information about their user location, physical and social activity, profiles and habits, etc. However, the informatio...
Article
Full-text available
Engaging users in threat reporting is important in order to improve threat monitoring in urban environments. Today, mobile applications are mostly used to provide basic reporting interfaces. With a rapid evolution of mobile devices, the idea of context awareness has gained a remarkable popularity in recent years. Modern smartphones and tablets are...
Article
Formalized process models help to handle, design and store processes in a form understandable for the designers and users. As model repositories often contain similar or related models, they should be used when modelling new processes in a form of automated recommendations. It is important, as designers prefer to receive and use suggestions during...
Article
Mobile devices are valuable sources of information about their user location, physical and social activity, profiles and habits. Such an information can be used to build context-aware applications, that are able to adapt their functionality to user needs and preferences. A lot of research have been done in this field of science, providing multiple...
Conference Paper
With the rapid development of mobile technologies like, e.g., RFID tags, smartphones, and tablets, ambient intelligence applications have gained a huge popularity in recent years. However, most of the existing approaches aim at developing ambient environments that are rather static, and do not take the aspect of social interaction between the inhab...
Conference Paper
Full-text available
With a rapid evolution of mobile devices, the idea of context awareness has gained a remarkable popularity in recent years. Modern smartphones and tablets are equipped with a variety of sensors including accelerometers, gyroscopes, pressure gauges, light and GPS sensors. Additionally, the devices become computationally powerful which allows real-ti...
Conference Paper
Full-text available
Context-aware mobile systems have gained a remarkable popularity in recent years. Mobile devices are equipped with a variety of sensors and become computationally powerful, which allows for real-time fusion and processing of data gathered by them. However, most of existing frameworks for context-aware systems, are usually dedicated to static, centr...
Conference Paper
Formalized process models help to handle, design and store processes in a form understandable for the designers and users. Modeling of business processes is a complex task, which can be supported by recommendations. It is important, as designers prefer to receive and use suggestions during the modeling process. Recommendations make modeling faster...
Conference Paper
Full-text available
Modeling business processes is an important issue in Business Process Management. As model repositories often contain similar or related models, they should be used when modeling new processes. The goal of this paper is to provide an overview of recommendation possibilities for business process models. We introduce a categorization and give example...
Conference Paper
Full-text available
Formalized knowledge representation methods allow to build useful and semantically enriched knowledge bases which can be shared and reasoned upon. Unfortunately, knowledge acquisition for such formalized systems is often a time-consuming and tedious task. The process requires a domain expert to provide terminological knowledge, a knowledge engineer...
Conference Paper
Full-text available
Semantic wikis constitute an increasingly popular class of systems for collaborative knowledge engineering. We developed Loki, a semantic wiki that uses a logic-based knowledge representation. It is compatible with semantic annotations mechanism as well as Semantic Web languages. We integrated the system with a rule engine called HeaRT that support...
Chapter
Semantic wikis provide a flexible solution for collaborative knowledge engineering. The reasoning capabilities of most of the semantic wiki systems are limited to lightweight reasoning, e.g. with the use of RDF. Number of systems try to integrate strong reasoning with rules. Since most of the regular rule engines are hard to integrate, custom solut...
Conference Paper
Full-text available
Quality and reliability issues are important in development and exploration of rule-based systems. In the paper a formalized knowledge representation for rules called XTT2 is considered. It is a rule language based on an expressive attribute logic called ALSV(FD). A custom runtime and verification framework for XTT2 called HalVA is proposed. It all...
Chapter
Full-text available
Wprowadzenie Regułowe systemy ekspertowe stanowią uniwersalne narzędzie reprezentacji i przetwarzania wiedzy. Znajdują one wiele różnorodnych zastosowań praktycznych. Klasyczne obszary zastosowań obejmują m. in. sterowanie, diagnostykę i wspomaganie decyzji. W ostatnim dziesięcioleciu systemy takie znajdują także zastosowanie we wspomaganiu zarządz...
Conference Paper
Full-text available
In the paper an extended knowledge representation for rules is considered. It is called Extended Tabular Trees (XTT2) and it provides a network of decision units grouping rules working in the same context. The units are linked into an inference network, where a number of inference options are considered. The original contribution of the paper is th...

Questions

Questions (8)
Question
I am looking for methods that allow for automated discovery of rules that can have temporal conditions.
The very simple example would be:
IF user activity WAS running, and now the user activity IS idle THEN user turns of tracking application
The more advanced example could be
IF user activity during last 30 minutes was running, and during last 5 minutes was idle, THEN user turns of tracking application
Do you think that some solutions from process mining techniques could be useful? Or some techniques from discovery of Markov models? I am specifically interested in rules.
Question
I am working on an open source, non-commercial editor for visualization and edition of structured rule-based knowledge bases.
I'm looking for volunteers who would like to go through simple evaluation task for the tool and fill the usability evaluation form.
Also, if you worked on similar tools, let me know -- I will be glad to include your work in the evaluation paper.
Question
I am interested in datasets for knowledge discovery from data streams.
It would be valuable to obtain a real datasets, but artificial are fine for me as well. I am aware of: http://www.liaad.up.pt/kdus/products/datasets-for-concept-drift
Question
As anyone aware if there exists some kind of a repository (or some single examples) of deep neural networks models that can be reused for transfer learning? I am aware of one: 'http://s3.amazonaws.com/GraphLab-Datasets/deeplearning/imagenet_model_iter45, However it only can be used as far as I'm concerned with GraphLab Create, not with open source scikit-learn.
Question
Lets assume I have three phenomena A, B C that i can measure. Furthermore, I know that P(A) = 0.7, P(B) = 0.2, P(C) = 0.1.  Now, I get the measurement which says that the observed phenomena is C with probability 0.8.The question is what is the probability that the phenomena is A, B or C knowing these facts. 

Network

Cited By

Projects

Projects (5)
Archived project
Nowadays great attention is paid to the Industry 4.0 concept, whose central idea is the exploitation of large amounts of data generated by different kinds of sensors, to enact highly automatized, robust processes and to develop high quality process monitoring systems that support intelligent semi-autonomous decision making. At the same time, big data analytics as core competency and a process-oriented management approach are very often indicated as pillars of any modern company. Towards this, the main objective of PACMEL is to develop a process-aware analytics framework for analyzing data from sensors and devices, enabling its use for process modeling and analysis, with the aim of improving the business processes according to the BPM cycle. The framework can be applied to the data system of smart factories to support the business process management activities in the scope of process modeling and analysis. On the one hand, it will allow the creation of conceptual models of particular industrial processes being executed in the factory, taking into account the various abstraction levels of the collected data. This will be achieved by combining knowledge extraction techniques with semantic technologies such as ontology-based data access and integration. On the other hand, it will support model mapping methods and visualization techniques that allow one to relate the interpreted sensor data to the process models for process analysis. We will use a real dataset, related to a very complex and specific process, from an industrial domain (mining). The complexity of the considered process is a consequence both of its intrinsic characteristics, and of the conditions under which it is realized. Working with this challenging example will bring valuable insights and results that can be applied across various industrial domains including aeronautics and manufacturing.