Periodic Reporting for period 2 - MATHISIS (Managing Affective-learning THrough Intelligent atoms and Smart InteractionS)
Reporting period: 2017-07-01 to 2019-03-31
MaTHiSiS which stands for ‘Managing Affective-learning THrough Intelligent atoms and Smart InteractionS’ is a project centered on exciting new software that promotes a new way of thinking about education and training. MaTHiSiS intends to provide high support for students increasing their engagement while learning, providing extra motivation and helping students with learning difficulties. MaTHiSiS works on tablets, phones, interactive whiteboards and personal computers – and even robots! Teachers and trainers can design their courses in the form of graphs where they attach the learning content with various difficulty levels. MaTHiSiS is able to provide each of their learners with a learning path matched to their individual skills and needs, personalised based on their individual profile and their previous learning achievement. MaTHiSiS monitors students’ progress and automatically adapts - via artificial intelligence features - as they interact with the system to suit their emotions and their abilities. The aim is to provide a unique personal experience tailored to each student specific needs, engage and motivate them to learn in new and more effective ways. See: www.mathisis-project.eu Twitter:@MaTHiSiSProject, Facebook:@mathisis.project.
Apart from the MaTHiSiS platform as a whole, the project has delivered individual standalone components which function within the MaTHiSiS system but can also be commercialized separately. They are the following: Affect Recognition, Learning Games Creation Tool, Learning Content Editor and Learning Analytics and Visualization tools. The MaTHiSiS platform and the individual results have been packaged for their distribution and published under the MaTHiSiS Web site (http://www.mathisis-project.eu/results ). The initial steps towards commercial activities have already been initiated through the Trademarkt (™) registration and through distribution of the system and individual components through third party platforms and marketplaces (RAGE).
The project has been promoted through 1) the MaTHiSiS web site in seven languages (English, Spanish, Greek, French, Lithuanian, Italian, German); 2) peer communications directly - with over 10; remotely - with more than 90 SMEs providing educational tools and solutions. 3) targeted communication activities with 30 education/training representatives of public administration at a local, regional and national level, in order to invite and attract policy makers; 4) 31 skill acquisition events with training sessions targeting tutors/learners. 5) 9 special sessions/workshops co-located at conferences targeting developers /SMEs; 6) Presence at 4 tradeshows; 7) 2 hackathon events targeting developers’ community; 8) participation in more than 35 events representing the project, raising awareness among the general public, aiming to inform and invite diverse stakeholder groups ; 9) 9 publications in peer-review journals and 22 Publications on international conferences and workshops, focusing on initiating the thematic building throughout the academic communities. 10) Constantly feeding social networks, taking advantage of the MaTHiSiS consortium existing wide networks, thus broadcasting project’s value proposition; 11) Two promotional videos (https://www.youtube.com/watch?v=IEE3j4nr8-w and https://youtu.be/tbgGnj6LvTU); 12) Pilots with schools, industries and other educations organisations, involving all four MaTHiSiS cases (ASC, PMLD MEC, ITC and CGDLC), thus having the opportunity to engage schools, learners and tutors with the MaTHiSiS ecosystem, in a more systematic way.
• A Cloud Learner Space (CLS)
• A robust goal-based learning experience design mechanism
• An elaborate authoring tool (Learning Content Editor - LCE).
• A sophisticated front-end for tutors and independent learners
• A privacy safeguarding and ethics-compliant User Space (US) in the CLS.
• A break-through Sensorial Component (SC).
• A cutting-edge component for affect recognition based on Interactions with the end user devices (platform agents) (IPA)
• Innovative research components for affect recognition based on keyboard/mouse interactions
• A highly innovative multimodal fusion mechanism (AIRlib),
• This mechanism enables a personalized reaction system (DSS).
• A non-linear learning scenario deployment engine (Learning Graph Engine - LGE)
• A sophisticated decision-making engine
• Support of a variety of learning user devices (platform agents – PAs)
• A cutting-edge synchronous collaboration component,
• Versatile front-end components (Experiencing Service (ES) clients – desktop and mobile
• A mechanism for knowledge transfer
• Lastly, around 130 Learning Materials have been delivered: 10 TurtleBot, 40 NAO, 54 Native Android and 28 Web-Based.