University of Ljubljana

01 January 2022 - 30 June 2024

PM - Project Management

Regarding project management, we planned the activities throughout the whole-time span of the project including the involvement of our students in the project. Planned results are in large part guided by the courses we teach at our university. Short summary of these activities includes:

Objectives were:
 Planning all project activities and results including meeting participation
 Planning project results: paper reviews, algorithms, and benchmarks
 Planning student involvement
 Planning researcher and lecturer involvement
 Planning dissemination events
 Design and implementation of 5 algorithms together with benchmarks

Period July - December 2023
 Involve students to test preliminary solutions
 Involve teachers and researchers to motivate students and discuss solutions
 Design OptLearn algorithm prototype
 Design a full OptLearn algorithm and propose a development cycle
 Understand and test automatic learning indicator estimation algorithms

Period January - July 2023
 Further development of OptLearn algorithm: concept map generation, question recommender
 Involve students to test preliminary solutions
 Provide tips on OptLearn algorithm usage and its extensions
 Propose applet structure
 Dissemination activities

Period July – December 2023
 Improvement of OptLearn Algorithm in terms of concept map generation
 Improvement of OptLearn Algorithm in terms of explicit maximization of student learning outcomes
 Applet: user requirements were acquired
 Applet: design and implementation of Concept map editor
 Applet: testing on a small set of teachers
 Dissemination: dissemination of the project
 Targeted dissemination event

Period January – June 2024
 Development of automatic concept map generation
 Evaluation of machine learning model to estimate students learning outcomes
 Evaluation and comparison different versions of OptLearn algorithm
 Dissemination: several presentations of iMath project to different target groups
 Dissemination: organization of the second Multiplier event

Basic activities carried out:
 Composing a list of researchers and teachers outside of our group we have and we plan to involve. Since their student involvement is scheduled according to the student calendar, these activities must be planned before semesters;
 Design and implementation of 5 algorithms and 5 benchmarks together with the typical student interaction with the system. These algorithms were selected according to expected student involvement in this and following student years.
 We integrated proposed algorithms and benchmarks into courses taught at our faculty;
 We designed and partly implement a system for automatic student indicator measurement based on Python Django framework. This is in part outside of our iMath project task description but results will contribute to the involvement of our students into the iMath project;
 Design of a dissemination event in conjunction with IEEE Education chapter active at our faculty. This event is planned for autumn 2022.

Period July - December 2022:
 Studying state of the art of automatic student learning indicator procedures and algorithms
 Design and description of OptLearn prototype
o Creation of a simple data example for the algorithm
o Simple knowledge graph description and generation
o Protype OptLearn algorithm description implementable without learning indicators
 Description of OptLearn algorithm as an extension of the above given prototype
o Selection of recommended items when learning indicators are known
o Outline of the testinf
 Involving students:
o Directly by live presentation of the project
o Through their teachers at other institutions

Period January - July 2023:
 Design, implementation, and testing of OptLearn algorithm: several attempts on automatic Concept map generation, improvements on concept map simplifications
 Automatic extraction of learning indicators: using student data from OptLearn algorithm testings, we tested several approaches of automatic learning indicators estimation
 OptLearn algorithm was released, documented, and reported


Period July – December 2023
 Study literature on Concept map generation
 Testing and verification different approaches of concept map generation
 Redesign and implementation of OptLearn, recommender system module to allow explicit maximisation of selected learning indicators
 User requirements for Concept map editor (applet) were composed and tested
 Concept map editor was designed and implemented
 Evaluation of Concept map editor on a small set of 6 teachers were performed and taken into account

Period January – June 2024
 A novel algorithm for automatic concept map generation was designed, implemented and evaluated on a group of students
 A machine learning model for estimation of student learning outcomes was selected, configured and tested on an iMath data
 UoL iMath applet (Concept map editor) was improved by adding new features and visualisation elements
 A methodology for concept map evaluation was selected, implemented and used on a group of students


Dissemination activities include:
 Formal and informal presentation of the iMath project to teaching colleagues from several institutions and to our project partners (active projects and projects in the phase of proposals). It includes mails with summaries, recommendations, and links to project portal, several events
 Presentation of iMath at SI-MATH-IN meeting
 Presentation of iMath project and out role to the national “Svetovalni center” SCOMS
 Presentation of iMath project to our Department of ICT

Activities regarding evaluation:
Discussion of benefits of iMath portal with our students in order to understand the preferred ways of using iMath portal regarding tasks and in order to design a student indicator measurement effectively.




Period July - December 2022:
 Preparation of project presentation to students with the aim of involving them as test users and providing necessary feedbacks
 Project was presented to several student groups
 Project was presented to several teachers and researches (to some just repeated to additionally motivate students);

Period January - July 2023:
 Dissemination activities in terms of presenting the iMath project to different communities
 Students of optimization course learned the project in terms of adapted user interaction and experimentation

Period July – December 2023
 A project presentation and discussion were organized at the conference ERK 23.
 Project goals and contributions were presented and a fruitful discussion emerged in terms how students and teachers could benefit from project results

Period January – June 2024
 An updated project presentation was prepared and used in several presentations to various target groups
 A project presentation and discussion (Multiplier event) were organized at the university of Maribor.

Results:
 Plan of project activities including dissemination
 List of researchers, lecturers and students that was and will be invited to involved;
 10 papers reviewed
 Design and pilot implementation of student interaction with optimisation algorithms
 Design and implementation of 5 algorithms and 5 benchmarks
 Description of 3 student indicators
 Dissemination in terms presenting the project to teachers and researchers at our and other institutions. Presentation of iMath to our project partners.
 Designed a dissemination event in conjunction with IEEE Education section

Period July - December 2022:
 Procedures of automatic student learning indicators are better understood and outlined as an algorithm.
 OptLearn algorithm prototype was designed and presented, including simple knowledge graph generation.
 Full OptLearn algorithm was designed and presented.

Period January - July 2023:
 A release of OptLearn algorithm with documentation
 Test results of OptLearn algorithm
 Design of the applet for Concept map editing
 Engagement of one student to work on iMath data for a period of 6 months

Period July – December 2023
 Improved version of OptLLearn algorithm was implemented and tested
 Applet Concept map editor was designed, implemented and tested
 Dissemination event were organized

Period January – June 2024
 Automatic concept map algorithm
 Methodology for concept map quality evaluation is implemented on a web platform
 Preliminary results of automatic concept map generation quality
 Preliminary results of OptLearn comparisons available
 Dissemination events and multiplier event were organized