@inproceedings{StauferHauserEzeretal., author = {Staufer, Susanne and Hauser, Florian and Ezer, Timur and Grabinger, Lisa and Nadimpalli, Vamsi Krishna and R{\"o}hrl, Simon and Bugert, Flemming and Bittner, Dominik and Mottok, J{\"u}rgen}, title = {EVALUATION OF THE LEARNING MANAGEMENT SYSTEM PYTHIA}, series = {EDULEARN Proceedings: 16th International Conference on Education and New Learning Technologies, Palma, Spain. 1-3 July, 2024}, booktitle = {EDULEARN Proceedings: 16th International Conference on Education and New Learning Technologies, Palma, Spain. 1-3 July, 2024}, publisher = {IATED}, issn = {2340-1117}, doi = {10.21125/edulearn.2024.2189}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-73474}, pages = {9089 -- 9098}, abstract = {Learning management systems gain importance due to the Corona pandemic. To personalize such a learning management system, a Moodle adaption named Pythia was created. Pythia is implemented as a Moodle plugin solution. The possibility of generating learning paths with various algorithms and the change of the graphical user interface is significant. Learning paths are generated with two different algorithms. Nestor algorithm generates learning paths with a Bayesian network, while Tyche algorithm uses a Markov model. The graphical user interface is adapted to the generated learning paths where learning element symbols are customized and the individual learning paths are depicted. This paper presents the results of a qualitative survey among German students with the aim of evaluating the first version of the learning management system Pythia. 25 students took the subject "Software Engineering for Safe and Secure Systems" in the winter term of 2023/24. The 68 asked questions focus on usage behavior, graphical user interface, usability, preferences, volume, and quality of learning elements, and learning paths. Our Moodle course for the lecture contains seven sections, each section has minimum one subsection on which the learning path is calculated. Last three sections are designed by presenting minimum one learning element of each learning element category except the collaboration tool. In a previous work, ten learning element categories were designed and evaluated comprising manuscript, brief overview, learning goal, quiz, exercise, summary, three types of additional material, and collaboration tool. The first sections are left as they are initially designed by the lecturer (normal section). The survey considers the difference between normal sections and diverse sections. Furthermore, the generated learning paths are evaluated. 13 students undertook Nestor learning paths, whereas 12 undertook Tyche learning paths. The results suggest that for the majority of students, the learning element symbols are well-chosen. Moreover, all students wish the elements relevant to the exam to be labelled. 23 students are overwhelmed by the number of learning elements in the last sections of the Moodle course. In the future, some suggestions will be implemented such as the labelling of exam relevant learning elements. To overcome the overload of learning elements, a strategy could be developed like hiding learning elements not preferred by the learning path algorithm.}, language = {en} } @inproceedings{RoehrlStauferNadimpallietal., author = {R{\"o}hrl, Simon and Staufer, Susanne and Nadimpalli, Vamsi Krishna and Bugert, Flemming and Bugert, Flemming and Hauser, Florian and Grabinger, Lisa and Bittner, Dominik and Ezer, Timur and Mottok, J{\"u}rgen}, title = {PYTHIA - AI SUGGESTED INDIVIDUAL LEARNING PATHS FOR EVERY STUDENT}, series = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, booktitle = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, isbn = {978-84-09-59215-9}, issn = {2340-1079}, doi = {10.21125/inted.2024.0783}, pages = {2871 -- 2880}, abstract = {During the COVID-19 pandemic, the importance of digital course rooms, where teachers provide their learning materials, increased dramatically. While these platforms are crucial for providing teaching materials, they often fall short in addressing individual student needs. A system within an academic setting, capable of creating and presenting individual learning paths for each student, can solve these issues. These paths are composed of various learning elements - defined in our previous work as units of educational content with which a learner works. Currently, there is no suitable system that enables the integration of learning path generating algorithms into a digital course room. Therefore we present an application that enables this integration into the Moodle Learning Management System (LMS). More precisely, this paper presents a Moodle plugin together with its framework. It describes the mechanism for effectively collecting data from Moodle, which AI algorithms then use to generate personalized learning paths. Subsequently these paths are visualized with the help of the Moodle plugin. We started with a set of requirements and use cases for the interface connecting Moodle to the AI system, which were established with a group of experts. Based on the requirements, various relevant technologies were assessed, and the best ones were chosen for implementation. Following that, the paper develops a strategy for software structuring as well as an architecture, focusing on performance, modularity, and ease of deployment for widespread use. Furthermore, the architecture ensures a simple method for integrating the algorithms. Afterwards, the framework's concrete implementation is described. A technique for enriching learning elements with metadata is presented, and additionally a concept for presenting these learning elements within a hierarchy. Moreover, it is shown how questionnaire responses and learning analytics are utilized for data collection. We cover in detail techniques for extracting and storing data from the Moodle database, as well as methods for customizing Moodle course rooms and a standard API for incorporating AI algorithms. Finally, the paper discusses the application of the proposed framework in an actual course and how student feedback is collected, which could enhance the framework. It concludes with an assessment of the outcomes obtained and prospects for the framework's future advancements.}, language = {en} } @inproceedings{StauferHauserGrabingeretal., author = {Staufer, Susanne and Hauser, Florian and Grabinger, Lisa and Bittner, Dominik and Nadimpalli, Vamsi Krishna and Bugert, Flemming and Ezer, Timur and R{\"o}hrl, Simon and Mottok, J{\"u}rgen}, title = {Learning elements in LMS - a survey among students}, series = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, booktitle = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, publisher = {IATED}, isbn = {978-84-09-59215-9}, issn = {2340-1079}, doi = {10.21125/inted.2024.1087}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-71135}, pages = {4224 -- 4231}, abstract = {In higher education, improving learning and learning success are goals of general improvement. Lecturers teaches content and students acquire that content in an efficient way. To structure content, learning element categories are evaluated from the student's point of view in higher education area. The aim is to validate given definitions of ten learning element categories within a Learning Management System (LMS). This paper evaluates a categorization of learning elements for organizing learning content in online education within LMSs. Therefore, ten categories of learning elements and corresponding definitions were defined in a previous work as base for this paper. The learning elements to examine are manuscript, exercise, quiz, brief overview, learning goal, summary, collaboration tool, auditory additional material, textual additional material, and visual additional material. To validate the definitions and to get improvements to each learning element a survey is processed. Beside the demographic data questions, the survey consists of two questions to the acceptance of the definitions and asks for improvements. 148 students between the ages 19 and 35 participate in the survey in summer term 2023. The education level of the participants ranges from undergraduates to Ph.D. students. The results of this paper are that more than 80\% accept the given definitions. Some definitions of the learning elements are changed, but the changes are restricted to additions of maximal four words. This categorization of learning elements could lead to improvements in learning by giving the content more structure. With the structure students get the possibility to learn with preferred learning elements which could lead to more success in learning and to a decreasing dropout rate in universities. In the future, the learning elements allow to classify content within LMSs with the goal of generating individual learning paths. Furthermore, our project will integrate these learning elements, use them to generate learning paths, and could set a new standard in the way of personalized learning.}, language = {en} } @inproceedings{EzerGrabingerHauseretal., author = {Ezer, Timur and Grabinger, Lisa and Hauser, Florian and Staufer, Susanne and Mottok, J{\"u}rgen}, title = {EYE TRACKING AS TECHNOLOGY IN EDUCATION: FURTHER INVESTIGATION OF DATA QUALITY AND IMPROVEMENTS}, series = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, booktitle = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, publisher = {IATED Academy}, isbn = {978-84-09-59215-9}, issn = {2340-1079}, doi = {10.21125/inted.2024.0802}, pages = {2955 -- 2961}, abstract = {Eye tracking serves as a powerful tool across a variety of empirical research areas: From usability research over cognitive research to educational research and applications in classrooms. However, data noise in eye tracking data poses a challenge to researchers and educators, as it leads to gaze positions being measured imprecisely under unfavorable conditions. In our previous study, we systematically investigated factors that influence data quality and are easily controllable in a classroom or laboratory environment, such as illumination, sampling frequency, and head orientation. However, no recommendations regarding the light source and light orientation could be provided, as these influences could not be analyzed in sufficient detail. Yet, a further examination of these factors, eliminating human influences by using an artificial head, revealed significant differences between individual settings. Hence, in this empirical study of eye tracking as an educational technology, we delve deeper into examining the impact of both light source and light orientation on data quality. This is investigated with an artificial head together with the Tobii Pro Spectrum eye tracking device. To measure data quality, we use the metrics precision and standard deviation as indicators of data noise. The obtained results derive practical advice for educators and researchers, such as not to illuminate the subject from the rear, in order to gather useful data for research and future classroom applications. Thereby, this study serves as a complement to our previous research, answering open questions regarding best practices for researchers and educators when using eye trackers. It aims to provide valuable insights into producing data of the highest quality possible when using eye trackers, both in laboratory settings and in future classrooms applications.}, language = {en} } @inproceedings{EzerGrabingerHauseretal., author = {Ezer, Timur and Grabinger, Lisa and Hauser, Florian and Staufer, Susanne and Mottok, J{\"u}rgen}, title = {EYE TRACKING METRICS FOR DISTINGUISHING GLOBAL AND FOCAL GAZE PATTERNS: A SYSTEMATIC LITERATURE REVIEW}, series = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, booktitle = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, publisher = {IATED Academy}, isbn = {978-84-09-59215-9}, issn = {2340-1079}, doi = {10.21125/inted.2024.0814}, pages = {3005 -- 3014}, abstract = {Global and focal eye tracking gaze patterns are distinguished in a variety of domains, such as radiology research, empirical software engineering, behavioral psychology, and cartography research. A global gaze pattern is present if a participant in an eye tracking study gains an overview of the stimulus, whereas a focal pattern emerges during a more detailed evaluation of a specific part of the stimulus. With this higher-level measure of global and focal gaze patterns, cognitive effort, expertise levels, and even neurological characteristics - such as autism - can be quantified. Furthermore, with this measure, cognitive efforts of students in certain tasks can be investigated and clues for targeted support can be provided. These properties in particular make this eye tracking measure valuable for the education domain. In the field of eye tracking research, many researchers have established distinct criteria for differentiating global and focal gaze patterns. Moreover, it is rarely quantitatively measured which of these patterns is present, but rather determined based on qualitative considerations. However, some studies explicitly use quantitative metrics like saccade-length or fixation-duration to distinguish between global and focal gaze patterns. This is the focus of the present systematic literature review: It searches for and compiles the eye tracking metrics used in the literature to distinguish global from focal gaze patterns in a quantitative way. In doing so, this study fills the yawning gap in quantitative metrics for determining global and focal gaze patterns. In the long run, this can be used to answer questions about cognitive load and problems in processing tasks during eye tracking studies and to provide further insights into the cognitive processes of students. The present paper first discusses definitions of global and focal gaze patterns and presents examples in which this metric has already been applied. Then, the research questions together with search strings and search engines used for the systematic literature review are described. Finally, the results are summarized, presenting a compilation of and connections between quantitative metrics utilized in the literature to distinguish between global and focal viewing patterns.}, language = {en} } @inproceedings{StauferHauserGrabingeretal., author = {Staufer, Susanne and Hauser, Florian and Grabinger, Lisa and Bittner, Dominik and Nadimpalli, Vamsi Krishna and Mottok, J{\"u}rgen}, title = {Learning elements in online learning management systems}, series = {Proceedings of the 16th annual International Conference of Education, Research and Innovation (ICERI 2023), Seville, Spain, 11/13/2023 - 11/14/2023}, booktitle = {Proceedings of the 16th annual International Conference of Education, Research and Innovation (ICERI 2023), Seville, Spain, 11/13/2023 - 11/14/2023}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED}, doi = {10.21125/iceri.2023.0815}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-67725}, pages = {3121 -- 3130}, language = {en} } @inproceedings{StauferNadimpalliHauseretal., author = {Staufer, Susanne and Nadimpalli, Vamsi Krishna and Hauser, Florian and Bittner, Dominik and Grabinger, Lisa and Bugert, Flemming and Ezer, Timur and R{\"o}hrl, Simon and Mottok, J{\"u}rgen}, title = {Persistence of Questionnaire Data on Learning Styles, Learning Strategies AND Personality Traits}, series = {ICERI2024 Proceedings: 17th Annual International Conference of Education, Research and Innovation, 11th-13th November 2024, Seville, Spain}, booktitle = {ICERI2024 Proceedings: 17th Annual International Conference of Education, Research and Innovation, 11th-13th November 2024, Seville, Spain}, publisher = {IATED}, isbn = {978-84-09-63010-3}, issn = {2340-1095}, doi = {10.21125/iceri.2024.1524}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-78246}, pages = {6310 -- 6319}, abstract = {In the context of learning, questionnaires are carried out to gain information about learners. This learner profile has various benefits, like adapting the learning content, promoting of self-reflection, or increasing motivation in learning and therefore better learning results. Some researchers take the results of such questionnaires as ground truth or trust them blindly. Examining how persistent questionnaires are, asking the same questions twice, and comparing the results is a way to prove or falsify this statement. Our research procedure compares the results of three different questionnaires. The first questionnaire is the ILS (index of learning styles) according to the Felder-Silverman Learning Style Model (FSLSM) with about 44 questions. Learning strategies are examined with the LIST-K questionnaire consisting of 39 questions after Klingsieck. The third and last questionnaire is about personality traits. With only ten questions, Rammstedt et al. describe the BFI-10 questionnaire for personality traits. To compare results, the same study with the three named questionnaires is carried out twice with three to four months in between. The sample includes 35 students in their bachelor studies taking the course Software Engineering in a blended learning format. After collecting the questionnaire data from the students, a comparison is made in two different ways. First, we compare the data points based on the given answers. Second, a comparison is made based on the calculated questionnaire results because each of the three questionnaires has its own analysis method. For example, the analysis method of the ILS delivers as a result four integers representing the four characteristics of a learning style, while LIST-K makes a factor analysis. The results of the paper show that both the given answers and the calculated questionnaire results differ. Not one student has given the same answer inside each questionnaire with causes, like a change of learning style, learning strategy, and personality traits, or not filling in conscientiously.}, language = {en} } @article{RoehrlStauferBugertetal., author = {R{\"o}hrl, Simon and Staufer, Susanne and Bugert, Flemming and Nadimpalli, Vamsi Krishna and Hauser, Florian and Grabinger, Lisa and Bittner, Dominik and Ezer, Timur and Altweck, Ricardo and Graf, Sabine and Kriza, Thomas and Mottok, J{\"u}rgen}, title = {Ethical Considerations of AI in Education: A Case Study based on Pythia Learning Enhancement System}, series = {IEEE Access}, volume = {11}, journal = {IEEE Access}, publisher = {Institute of Electrical and Electronics Engineers (IEEE)}, issn = {2169-3536}, doi = {10.1109/ACCESS.2025.3583975}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-83403}, pages = {1 -- 25}, abstract = {As the use of AI in education increases, addressing ethical concerns is essential for its responsible use and implementation. This study explores the ethical considerations of AI in education by analyzing a concrete application, the Pythia Learning Enhancement System, which enables personalized and adaptive educational experiences. In this case study, the concrete implementation of Pythia, contrary to high-level abstract analyses, provides practical insights into these ethical issues. To achieve this goal, we first define Pythia and outline its implementation. Using the various topics covered by Pythia, the ethical issues associated with each of these topics are outlined, resulting in a list of ten distinct ethical issues. Because there is overlap in issues covered by similar applications, they can benefit from this identification. An ethical (human rights) and didactic (constructivism) theoretical background is then provided as a basis for further discussion. This is used to build the arguments used in the main study. This consists of analyzing each of the issues and providing recommendations to mitigate potential harm arising from these issues. This helps to provide guidelines for future applications. In the analysis, the ethical discussion is mainly based on the human rights of the stakeholders involved, emphasizing human dignity. After the analysis, each of the issues is contextualized with the recent AI Act of the European Union, providing a novel mapping of the issues of AI in education to the AI Act. Finally, the limitations of the study are stated and a plan for future research is presented.}, language = {en} } @inproceedings{BugertStauferBittneretal., author = {Bugert, Flemming and Staufer, Susanne and Bittner, Dominik and Nadimpalli, Vamsi Krishna and Ezer, Timur and Hauser, Florian and Grabinger, Lisa and Mottok, J{\"u}rgen}, title = {Ariadne's Thread for Unravelling Learning Paths: Identifying Learning Styles via Hidden Markov Models}, series = {2024 IEEE Global Engineering Education Conference (EDUCON), 08-11 May 2024, Kos Island, Greece}, booktitle = {2024 IEEE Global Engineering Education Conference (EDUCON), 08-11 May 2024, Kos Island, Greece}, publisher = {IEEE}, issn = {2165-9567}, doi = {10.1109/EDUCON60312.2024.10578825}, pages = {1 -- 7}, abstract = {Modern education through Learning Management Systems (LMSs) provides learners with personalized learning paths. This is achieved by first querying the learning style according to the theory of Felder and Silverman to recommend suitable learning content. However, a rigid learning style representation is lacking of adaptability to the learners' choices. Therefore, the present study evaluates the idea of providing adaption to the representation of learning styles by using Hidden Markov Models (HMMs). Thus, data is collected from participants out of the Higher Education Area. The Index of Learning Styles questionnaire is used to obtain the learning style based on the theory of Felder and Silverman. Also, a questionnaire that asks the respondents to create a preferred learning path with the sequence length of nine learning elements is provided. From the given data, we initially evaluate the probability relationships between learning styles and learning elements. Then, we use the Viterbi algorithm in HMMs to identify alterations in learning styles from the provided learning paths. The alignment is then quantified by introducing a metric called support value. The findings imply that our concept can be used to adapt the learning style based on the user's real choice of learning elements. Thus, the proposed model also offers a way to integrate a feedback loop within LMSs leading to an improvement of learning path recommendation algorithms.}, language = {en} } @inproceedings{BugertGrabingerBittneretal., author = {Bugert, Flemming and Grabinger, Lisa and Bittner, Dominik and Hauser, Florian and Nadimpalli, Vamsi Krishna and Staufer, Susanne and Mottok, J{\"u}rgen}, title = {Towards Learning Style Prediction based on Personality}, series = {Proceedings of the 5th European Conference on Software Engineering Education, Seeon/Germany, June 19 - 21, 2023}, booktitle = {Proceedings of the 5th European Conference on Software Engineering Education, Seeon/Germany, June 19 - 21, 2023}, publisher = {ACM}, isbn = {978-1-4503-9956-2}, doi = {10.1145/3593663.3593682}, pages = {48 -- 55}, abstract = {This paper assesses the relation between personality, demographics, and learning style. Hence, data is collected from 200 participants using 1) the BFI-10 to obtain the participant's expression of personality traits according to the five-factor model, 2) the ILS to determine the participant's learning style according to Felder and Silverman, and 3) a demographic questionnaire. From the obtained data, we train and evaluate a Bayesian network. Using Bayesian statistics, we show that age and gender slightly influence personality and that demographics as well as personality have at least a minor effect on learning styles. We also discuss the limitations and future work of the presented approach.}, language = {en} } @inproceedings{BittnerHauserNadimpallietal., author = {Bittner, Dominik and Hauser, Florian and Nadimpalli, Vamsi Krishna and Grabinger, Lisa and Staufer, Susanne and Mottok, J{\"u}rgen}, title = {Towards Eye Tracking based Learning Style Identification}, series = {Proceedings of the 5th European Conference on Software Engineering Education, Seeon/Germany, June 19 - 21, 2023}, booktitle = {Proceedings of the 5th European Conference on Software Engineering Education, Seeon/Germany, June 19 - 21, 2023}, publisher = {ACM}, isbn = {978-1-4503-9956-2}, doi = {10.1145/3593663.3593680}, pages = {138 -- 147}, abstract = {The dropout rate at universities has been very high for years. Thereby, the inexperience and lack of knowledge of students in dealing with individual learning paths in various courses of study plays a decisive role. Adaptive learning management systems are suitable countermeasures, in which learners' learning styles are classified using questionnaires or computationally intensive algorithms before a learning path is suggested accordingly. In this paper, a study design for student learning style classification using eye tracking is presented. Furthermore, qualitative and quantitative analyses clarify certain relationships between students' eye movements and learning styles. With the help of classification based on eye tracking, the filling out of questionnaires or the integration of computationally or cost-intensive algorithms can be made redundant in the future.}, language = {en} } @inproceedings{NadimpalliBugertBittneretal., author = {Nadimpalli, Vamsi Krishna and Bugert, Flemming and Bittner, Dominik and Hauser, Florian and Grabinger, Lisa and Staufer, Susanne and Mottok, J{\"u}rgen}, title = {Towards personalized learning paths in adaptive learning management systems: bayesian modelling of psychological theories}, series = {Proceedings of the 16th annual International Conference of Education, Research and Innovation (ICERI 2023), Seville, Spain, 11/13/2023 - 11/14/2023}, booktitle = {Proceedings of the 16th annual International Conference of Education, Research and Innovation (ICERI 2023), Seville, Spain, 11/13/2023 - 11/14/2023}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED}, doi = {10.21125/iceri.2023.1144}, pages = {4593 -- 4603}, abstract = {In educational research, non-personalized learning content increases learners' cognitive load, causing them to lower their performance and sometimes drop out of the course. Personalizing learning content with learners' unique characteristics, like learning styles, personality traits, and learning strategies, is being suggested to improve learners' success. Several theories exist for assessing learners' unique characteristics. By the end of 2020, 71 learning style theories have been formulated, and research has shown that combining multiple learning style theories to recommend learning paths yields better results. As of the end of 2022, there is no single research that demonstrates a relationship between the Index of Learning Styles (ILS) based Felder-Silverman learning style model (FSLSM) dimensions, Big Five (BFI-10) based personality traits, and the Learning strategies in studying (LIST-K) based learning strategies factors for personalizing learning content. In this paper, an innovative approach is proposed to estimate the relationship between these theories and map the corresponding learning elements to create personalized learning paths. Respective questionnaires were distributed to 297 higher education students for data collection. A three-step approach was formulated to estimate the relationship between the models. First, a literature search was conducted to find existing studies. Then, an expert interview was carried out with a group of one software engineering education research professor, three doctoral students, and two master's students. Finally, the correlations between the students' questionnaire responses were calculated. To achieve this, a Bayesian Network was built with expert knowledge from the three-step approach, and the weights were learned from collected data. The probability of individual FSLSM learning style dimensions was estimated for a new test sample. Based on the literature, the learning elements were mapped to the respective FSLSM learning style dimensions and were initiated as learning paths to the learners. The next steps are proposed to extend this framework and dynamically recommend learning paths in real time. In addition, the individual levels of learning style dimensions, personality traits, and learning strategies can be considered to improve the recommendations. Further, using probabilities for mapping learning elements to learning styles can increase the chance of initiating multiple learning paths for an individual learner.}, language = {en} } @inproceedings{EzerPloesslGrabingeretal., author = {Ezer, Timur and Pl{\"o}ßl, Moritz and Grabinger, Lisa and Bittner, Dominik and Staufer, Susanne and Nadimpalli, Vamsi Krishna and Bugert, Flemming and Hauser, Florian and Mottok, J{\"u}rgen}, title = {Deep learning for eye movement classification}, series = {ICERI 2024 Proceedings, 17th annual International Conference of Education, Research and Innovation,11-13 November, 2024, Seville, Spain}, booktitle = {ICERI 2024 Proceedings, 17th annual International Conference of Education, Research and Innovation,11-13 November, 2024, Seville, Spain}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED}, doi = {10.21125/iceri.2024.1028}, pages = {4056 -- 4065}, language = {en} } @inproceedings{HauserStauferGrabingeretal., author = {Hauser, Florian and Staufer, Susanne and Grabinger, Lisa and R{\"o}hrl, Simon and Mottok, J{\"u}rgen}, title = {On the analysis of student learning strategies: using the LIST-K questionnaire ro generate ai-based individualized learning paths}, series = {Proceedings of the 16th annual International Conference of Education, Research and Innovation (ICERI 2023), Seville, Spain, 11/13/2023 - 11/14/2023}, booktitle = {Proceedings of the 16th annual International Conference of Education, Research and Innovation (ICERI 2023), Seville, Spain, 11/13/2023 - 11/14/2023}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED}, doi = {10.21125/iceri.2023.1147}, pages = {4611 -- 4620}, abstract = {This paper presents the results of a data collection with the LIST-K questionnaire. This questionnaire measures students' learning strategies and shows which strategies are particularly dominant or rather weak. Learning strategies have long been a major area of research in educational science and psychology. In these disciplines, learning strategies are understood as intentional behaviors and cognitive skills that learners employ to effectively complete learning tasks, by selecting, acquiring, organizing, and integrating information into their existing knowledge for long-term retention. The LIST-K, developed by Klingsieck in 2018, was chosen for accessing learning strategies due to its thematic suitability, widespread use, and test economy. It covers a total of four main categories (i.e., cognitive strategies, metacognitive strategies, management of internal resources, and management of external resources), each of which are subdivided into further subscales. With a total of 39 items answered via a 5-step Likert scale, the LIST-K can cover the topic relatively comprehensively and at the same time be completed in a reasonable amount of time of approximately 10 minutes. The LIST-K was used as part of a combined data collection along with other questionnaires on their personal data, their preferences regarding certain learning elements, their learning style (i.e. the ILS), and personality (i.e. the BFI-10). A total of 207 students from different study programs participated via an online survey created using the survey tool "LimeSurvey". Participation in the study was voluntary, anonymously, and in compliance with the GDPR. Overall, the results of the LIST-K show that students are willing to work intensively on relevant topics intensively and to perform beyond the requirements of the course seeking additional learning material. At the same time, however, it is apparent that the organization of their own learning process could still be improved. For example, students start repeating content too late (mean=2.70; SD=0.92) and do not set goals for themselves and do not create a learning plan (mean=3.19; SD=0.90). They also learn without a schedule (mean=2.23; SD=0.97) and miss opportunities to learn together with other students (mean=3.17; SD=0.94). The findings of the data collection will be used to create an AI-based adaptive learning management system that will create individualized learning paths for students in their respective courses. From the results of the LIST-K, it appears that the adaptive learning management system should primarily support organizational aspects of student learning. Even small impulses (an individual schedule of when to learn what or a hierarchical structuring of the learning material) could help students to complete their courses more successfully and improve their learning.}, language = {en} } @inproceedings{BugertNadimpalliBittneretal., author = {Bugert, Flemming and Nadimpalli, Vamsi Krishna and Bittner, Dominik and Ezer, Timur and Grabinger, Lisa and Maier, Robert and R{\"o}hrl, Simon and Staufer, Susanne and Hauser, Florian and Mottok, J{\"u}rgen}, title = {ML based Evaluation Methodology for Learning Path Recommender Systems}, series = {Proceedings of the 6th European Conference on Software Engineering Education : ECSEE 2025, Seeon Germany, June 02-04, 2025}, booktitle = {Proceedings of the 6th European Conference on Software Engineering Education : ECSEE 2025, Seeon Germany, June 02-04, 2025}, editor = {Mottok, J{\"u}rgen and Hagel, Georg}, publisher = {ACM}, isbn = {9798400712821}, doi = {10.1145/3723010.3723022}, pages = {40 -- 48}, abstract = {In education, recommender systems can provide students with personalized learning materials based on their preferences. When comparing various recommendation algorithms, the main question is, which algorithm provides the most suitable recommendations for each student. Answering this question requires a quantitative evaluation methodology (i.e. a concrete metric) for ranking the results of (even non-deterministic) recommender systems. While there is already literature on this topic, the uniqueness of our approach lies in the application of machine learning: we deploy a likelihood based analysis via Hidden Markov Models named Aiakos. With this strategy, we aim to provide data-driven insights about accuracy and stability of recommendations towards a more reasonable selection of the appropriate recommender system. The training data for the Hidden Markov Models is collected from 80 students. Data from another 26 students is then used to discuss the behavior of our evaluation procedure considering a single recommendation as well as the results from 100 recommendations. Furthermore, the proposed concept allows to be applied to other domains as well.}, language = {en} } @inproceedings{NadimpalliBugertBittneretal., author = {Nadimpalli, Vamsi Krishna and Bugert, Flemming and Bittner, Dominik and Staufer, Susanne and R{\"o}hrl, Simon and Hauser, Florian and Ezer, Timur and Grabinger, Lisa and Maier, Robert and Mottok, J{\"u}rgen}, title = {Probabilistic Machine Learning for Simulating Complex Learner Profiles}, series = {2024 21st International Conference on Information Technology Based Higher Education and Training (ITHET), 06-08 November 2024, Paris, France}, booktitle = {2024 21st International Conference on Information Technology Based Higher Education and Training (ITHET), 06-08 November 2024, Paris, France}, publisher = {IEEE}, isbn = {979-8-3315-1663-5}, issn = {2473-2060}, doi = {10.1109/ITHET61869.2024.10837641}, pages = {1 -- 10}, abstract = {Data is the foundation of all machine learning applications. In education science, especially for the learner characteristics that drive personalized learning, it is difficult to collect and often uncertain. It is challenging to model, train, evaluate, and analyze the underlying algorithms when developing AI-based systems and having small sample sizes. To address these problems, we present a synthetic data generator utilising probabilistic models. This generator can effectively model and simulate complex learner profiles. To achieve this, we collected extensive data on learning styles, learning strategies, personalities, and preferred learning paths from 593 students over several semesters at a higher education level. Then, Bayesian networks, Hidden Markov Models, and Markov Chains are used to model the relationships between learner profiles. Using the Bayesian information criterion, and cross-validation with log-likelihood scores, we compare various models to select the best fitting one for synthesizing the data. The synthetic data is then evaluated using statistical validation techniques. In addition, we developed a simulation module with the option to simulate learner profiles based on manual user-defined inputs. The data and code used in this work are available as open source1110.5281/zenodo.13768136 [Titel anhand dieser DOI in Citavi-Projekt {\"u}bernehmen] contributing to open science and developers for customized simulated data. In the future, this data will refine the training, evaluation, analysis, and benchmarking of algorithms for personalized learning.}, language = {en} } @inproceedings{NadimpalliStauferEzeretal., author = {Nadimpalli, Vamsi Krishna and Staufer, Susanne and Ezer, Timur and Bugert, Flemming and Bittner, Dominik and Hauser, Florian and Grabinger, Lisa and R{\"o}hrl, Simon and Maier, Robert and Mottok, J{\"u}rgen}, title = {Predicting learner characteristics using machine learning}, series = {ICERI 2024 Proceedings, 17th annual International Conference of Education, Research and Innovation,11-13 November, 2024, Seville, Spain}, booktitle = {ICERI 2024 Proceedings, 17th annual International Conference of Education, Research and Innovation,11-13 November, 2024, Seville, Spain}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED}, doi = {10.21125/iceri.2024.1454}, pages = {6005 -- 6014}, abstract = {In education science research, data collection is challenging due to difficulty identifying students at the higher education level, privacy concerns, and varying levels of student engagement. Importantly, psychological questionnaires can be lengthy, leading to incomplete responses. We conducted repeated studies, and over time, the focus of the research adapted, introducing new materials and consequently leading to missing learner characteristics in some datasets. In this research, the issue of incomplete learner characteristics is addressed using data from three different studies: winter term 2022/2023 (n=297), summer term 2023 (n=274), and winter term 2023/2024 (n=25). These studies collected various learner characteristics, such as learning styles, personalities, learning strategies, and learning element preferences. However, learning element preferences and learning strategies were missing in the winter term of 2022, and the summer term of 2023 respectively. To analyze the data and predict these missing features, statistical analysis, and machine learning techniques were employed. Then, these models are rigorously evaluated using cross-validation and performance metrics like accuracy, precision, recall, and F1-score. Our findings provide insights into the relationships between learners' learning styles, personalities, learning strategies, and learning element preferences. This offers valuable implications for the design and implementation of educational interventions, like learning path recommendations. The results imply that machine learning models can predict missing learner characteristics, thus addressing the problem of incomplete data in educational research}, language = {en} } @inproceedings{StauferBugertHauseretal., author = {Staufer, Susanne and Bugert, Flemming and Hauser, Florian and Grabinger, Lisa and Ezer, Timur and Nadimpalli, Vamsi Krishna and Bittner, Dominik and R{\"o}hrl, Simon and Mottok, J{\"u}rgen}, title = {Tyche algorithm: Markov models for generating learning paths in learning management systems}, series = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, booktitle = {INTED2024 Proceedings: 18th International Technology, Education and Development Conference, Valencia, Spain. 4-6 March, 2024}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED Academy}, isbn = {978-84-09-59215-9}, issn = {2340-1079}, doi = {10.21125/inted.2024.1080}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-71148}, pages = {4195 -- 4205}, abstract = {In the intricate tapestry of the cosmos, where celestial threads weave stories of destiny, an enigmatic deity stands at the intersection of chance and fate — Tyche, the goddess of fortune. In science and probability, coincidence plays a distinctive role in Bayesian Networks (BNs) and Markov Models (MMs). This paper introduces the Tyche algorithm named after the goddess of fortune. The Tyche algorithm is a Markov model designed to generate learning paths in Learning Management Systems (LMSs). A learning path is a type of individualization that personalises the order of learning elements within an LMS course. Learning elements are fundamental components within an LMS course, depicting the learning content in diverse ways. In our case, the learning path tailoring is based on the learner's learning style according to Felder-Silverman Learning Style Model (FSLSM) - an indicator for the ideal pathway and learning element for the learner's optimum learning. The Tyche algorithm offers the advantage to provide students the most suitable learning path. Tyche is a MM structure with various matrices containing transition probabilities depending on the learning style. Nine categories of learning elements defined in a previous survey form the basis of the node structure of the MM. For a generic approach, a survey was designed to obtain the transition probabilities depending on the individual learning style. The survey with more than 100 German students participated is processed with the tool LimeSurvey. Students are asked about their learning style using the Index of Learning Styles (ILS) questionnaire according to Felder-Silverman and about the percentage probabilities of learning elements to get their individual sequence as learning path. The percentages are queried in two different ways. Firstly, the students were asked to provide a sequence of learning elements within each position of the sequence filled with probabilities for all learning elements. Secondly, a learning element is given and the students are asked to indicate which learning element they would work on next. The first way of asking for the probabilities is used to find the start node in the MM, whereas the second approach forms the probability matrices between the nodes within the MM. As result of the survey, the Tyche algorithm presents generic transition probabilities. It improves the learning process of individuals only by asking for their learning style: it generates individual learning paths through the learning elements within an LMS based on the MM explained above by solely getting the answers of the ILS questionnaire as input. In the future, other questionnaires such as BFI-10 for personality traits or LIST-K for learning strategies may offer a more comprehensive input. However, the next step is to evaluate Tyche with about 25 students in a software engineering lecture. This is planned for the year 2024.}, language = {en} } @inproceedings{StauferEzerRoehrletal., author = {Staufer, Susanne and Ezer, Timur and R{\"o}hrl, Simon and Grabinger, Lisa and Hauser, Florian and Nadimpalli, Vamsi Krishna and Antoni, Erika and Mottok, J{\"u}rgen and Schaffer, Josefa}, title = {TYCHE ALGORITHM 2.0: Learning Paths from Questionnaire Responses and Learning Analytics}, series = {ICERI2025 Proceedings}, booktitle = {ICERI2025 Proceedings}, publisher = {IATED}, isbn = {978-84-09-78706-7}, doi = {10.21125/iceri.2025.0831}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-87846}, pages = {2562 -- 2572}, abstract = {The digitalization of learning processes has increased the need for adaptive learning paths tailored individually to learners. A novel algorithm for learning path generation is presented in this paper, namely Tyche 2.0. It extends the original Tyche approach after Staufer et al. - a Markov model for generating learning paths - by integrating additional learner data beyond learning styles (Index of Learning Styles (ILS)), including learning strategies (LIST-K questionnaire), personality traits (BFI-10 questionnaire), and learning analytics captured through screen recordings. In order to be able to use the screen recordings, a heuristic evaluates them. Furthermore, this enhanced algorithm employs Markov models to dynamically generate personalized learning paths. These are based on both questionnaire responses and real-time engagement data, the weights of which undergo dynamic adjustment over time. We made a small evaluation of Tyche 2.0 without the learning analytics influence, which shows that there is room for further improvements. Future research will focus on evaluating whole Tyche 2.0 in another university setting to further improve personalization and user engagement.}, language = {en} } @inproceedings{HauserStauferRoehrletal., author = {Hauser, Florian and Staufer, Susanne and R{\"o}hrl, Simon and Nadimpalli, Vamsi Krishna and Ezer, Timur and Grabinger, Lisa and Mottok, J{\"u}rgen and Falter, Thomas}, title = {LEVERAGING FIVE QUESTIONNAIRES TO ANALYZE STUDENT LEARNING STRATEGIES AND GENERATE AI-POWERED INDIVIDUALIZED LEARNING PATHS}, series = {ICERI2025 Proceedings}, booktitle = {ICERI2025 Proceedings}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED}, doi = {10.21125/iceri.2025.0658}, pages = {1775 -- 1784}, abstract = {Background: The COVID-19 pandemic has significantly accelerated the shift toward online and blended learning in higher education, placing renewed emphasis on the individualization of learning content to meet diverse student needs. Even high-quality learning materials may fail to engage learners if they do not align with students' personal preferences and learning styles. Identifying these learner preferences, therefore, emerges as a critical challenge. Objectives: This paper presents ongoing work within a larger research project aimed at employing artificial intelligence to recommend optimal learning path for students in specific courses. Beyond mere optimization, the goal is to ensure the best possible fit between learning materials and individual learners. Sample \& Methods: A total of 27 students from technical degree programs took part in this survey. All participation was voluntary, and data were handled in full compliance with GDPR regulations. Although our broader project integrates fine-grained learning analytics from Moodle, the present abstract focuses exclusively on the self-report questionnaire results. Participants completed five instruments: 1. Index of Learning Styles (ILS) 2. LIST-K (Learning and Study Strategies Inventory - Short version) 3. BFI-10 (Big Five Inventory - 10 items) 4. Custom Preferences Instrument, capturing preferences for specific learning elements (e.g. instructional videos, lecture notes, summaries) and basic demographic data 5. Motivational Value Systems Questionnaire (MVSQ), piloted last semester to assess value orientations and motivational drivers Results: Preliminary analyses of the questionnaire data reveal: - Learning Styles (ILS): The majority lean toward the visual learning type (M = 5.740, SD = 3.430). - Learning Strategies (LIST-K): High scores on metacognitive strategies (M = 3.000; SD = 0.520) and collaboration with peers (M = 3.190; SD = 0.540). - Preferred Learning Elements: Summaries, overviews, and self-checks are most favored. - Value Orientations (MVSQ): Students are primarily driven by the pursuit of personal achievement (M = 4.400; SD = 11.140). Conclusion \& Significance: By integrating these five standardized questionnaires, we gain valuable insights into student learning preferences—insights that complement our Moodle analytics in the broader project. Observed trends suggest that learning materials should be concise and designed to facilitate peer interaction and knowledge deepening. These findings will guide the refinement of our AI-driven recommendation engine, enhancing its ability to deliver personalized learning paths that boost both engagement and effectiveness.}, language = {en} } @inproceedings{SchafferEzerRoehrletal., author = {Schaffer, Josefa and Ezer, Timur and R{\"o}hrl, Simon and Hauser, Florian and Staufer, Susanne and Nadimpalli, Vamsi Krishna and Grabinger, Lisa and Antoni, Erika and Mottok, J{\"u}rgen}, title = {EYE TRACKING GLASSES IN EDUCATIONAL SETTINGS: GUIDELINES ON DATA QUALITY}, series = {ICERI2025 Proceedings}, booktitle = {ICERI2025 Proceedings}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED}, doi = {10.21125/iceri.2025.1419}, pages = {5027 -- 5038}, abstract = {Eye movement modeling examples, so-called EMME videos, are a valuable tool in education, helping learners better understand instructional content. Like conventional educational videos, EMME videos combine text, images, and voiceovers. However, they also display the instructor's or teacher's gaze, guiding learners attention to key elements. Although various approaches exist for creating EMME videos, there is currently no standardised guideline for ensuring gaze data quality. Eye tracking technology is essential to capture gaze behaviour, and in educational settings without a fixed computer monitor - such as when using blackboards or conducting live experiments - the usage of mobile eye tracking glasses is beneficial. An accuracy study is conducted using mobile eye tracking glasses to provide empirical guidance for the development of high-quality educational EMME videos and ensure that the instructor's or teacher's gaze is captured with high precision. The study uses the Tobii Pro Glasses 3 and involves a static and a dynamic setup with 34 participants. To gain insight into the effects of visual impairments on accuracy, we also include participants who wear contact lenses. In the static setup, participants are seated at a desk with a headrest and focus on a poster with nine fixation points. In the dynamic setup, participants are walking in a controlled half-circle around the poster while maintaining focus on its centre. Each setup is performed multiple times under varying lighting levels (300 lux, 700 lux) and distances between participant and poster (80 cm, 120 cm, 180 cm). This enables the simulation of diverse educational environments, including the possibility of a teacher's or instructor's movement. The study results will be evaluated regarding lighting conditions, the distance between the person wearing eye tracking glasses and the object, and possible influences of contact lenses. Based on these findings, favourable conditions for creating EMME videos in educational settings are collected, especially when working without a fixed computer monitor. The results address the outlined research gap by providing instructors and teachers with guidelines enabling them to produce high-quality educational EMME videos.}, language = {en} } @inproceedings{NadimpalliMaierStauferetal., author = {Nadimpalli, Vamsi Krishna and Maier, Robert and Staufer, Susanne and R{\"o}hrl, Simon and Ezer, Timur and Grabinger, Lisa and Hauser, Florian and Mottok, J{\"u}rgen}, title = {EXPERT SURVEYS TO REAL TIME ADAPTATION OF LEARNING PATHS}, series = {ICERI2025 Proceedings}, booktitle = {ICERI2025 Proceedings}, editor = {G{\´o}mez Chova, Luis and Gonz{\´a}lez Mart{\´i}nez, Chelo and Lees, Joanna}, publisher = {IATED}, doi = {10.21125/iceri.2025.1571}, pages = {5677 -- 5687}, abstract = {Learning management systems rely on adaptive algorithms that use learner preferences to personalize the instructional content in form of learning paths. However, these preferences are uncertain in nature, and change over time. The present solutions are either static or purely data-driven missing the dynamic adaption to changes in the preferences and infusion of pedagogical nuances respectively. This paper introduces an extended variant of Nestor, our Bayesian network engine that models personality traits, learning styles, and learning strategies. This extension overlays a lightweight rule-based mechanism whose "secret recipe'' lies in the infusion of expert-derived weights adapting learning paths dynamically whenever a learner selects new material in Moodle. To parameterise these rules, we conducted a structured survey with 12 hand-picked professors and researchers in educational science. Each expert responded to 4 demographic items and 12 item that are distributed across algorithm-overview, scenario-based, and example-based categories, thereby supplying the nuanced weightings that result the personalised recommendations. This hybrid system (Nestor plus the expert-infused rule layer) operated during the winter term of 2025. 18 students completed an end-of-term questionnaire. Although their learning gains were not recorded, the majority of respondents reported positive or neutral experiences with the dynamically adapted learning paths. The {Future work} will compare three engines: (i) the present dynamic, expert-infused rule layer on top of the static Bayesian network, (ii) purely data-driven machine-learning models that neglect expert weighting, and (iii) the original static-adaptation Bayesian network without rules. Analyses of log files, intermediate satisfaction surveys, and pre/post term surveys will clarify whether this on-the-fly adaptation and pedagogical nuance lead to measurable learning benefits.}, language = {en} } @inproceedings{NadimpalliHauserBittneretal., author = {Nadimpalli, Vamsi Krishna and Hauser, Florian and Bittner, Dominik and Grabinger, Lisa and Staufer, Susanne and Mottok, J{\"u}rgen}, title = {Systematic Literature Review for the Use of AI Based Techniques in Adaptive Learning Management Systems}, series = {Proceedings of the 5th European Conference on Software Engineering Education, Seeon/Germany, June 19 - 21, 2023}, booktitle = {Proceedings of the 5th European Conference on Software Engineering Education, Seeon/Germany, June 19 - 21, 2023}, editor = {Mottok, J{\"u}rgen}, publisher = {Association for Computing Machinery}, address = {New York}, isbn = {978-1-4503-9956-2}, doi = {10.1145/3593663.3593681}, pages = {83 -- 92}, abstract = {Nowadays, learning management systems are widely employed in all educational institutions to instruct students as a result of the increasing in online usage. Today's learning management systems provide learning paths without personalizing them to the characteristics of the learner. Therefore, research these days is concentrated on employing AI-based strategies to personalize the systems. However, there are many different AI algorithms, making it challenging to determine which ones are most suited for taking into account the many different features of learner data and learning contents. This paper conducts a systematic literature review in order to discuss the AI-based methods that are frequently used to identify learner characteristics, organize the learning contents, recommend learning paths, and highlight their advantages and disadvantages.}, language = {en} } @misc{Staufer, author = {Staufer, Susanne}, title = {Persistence of Learning Style, Learning Strategy, and Personality Traits [Data set]}, doi = {10.5281/zenodo.12743866}, abstract = {This dataset contains the result of the survey to learning styles, learning strategies, and personality traits. The survey was executed in winter term 2023/24 and summer term 2023 in a German university (OTH Regensburg) during the course "Software Engineering". Examined questionnaires are ILS (learning styles), LIST-K (learning strategies), and BFI-10 (personality traits). The same three questionnaires were asked at two different survey periods three to four months apart while each survey period lasts one to two weeks. Pretest data were examined at the start of the term, while posttest data at the end.}, language = {en} }