SOCIAL THINGUM is Social Intelligent Learning Management System

SOCIAL THINGUM® is Social Intelligent Learning Management System for Demanding Users: highquality learning-support system with powerful use of Recommender Systems. www. socialthingum. com

SOCIAL THINGUM WWW. SOCIALTHINGUM. COM 2

Problem I need to create a course…. What to use? Where to find? Who to ask? Connexions ARIDIANE Open. Learn Share. TEC …. Merlot M. Mesiti, S. Valtolina, S. Bassis, EPIFANIA F. , B. Apolloni (2014). e-Teaching Assistant : A Social Intelligent Platform Supporting Teachers in the 3 Collaborative Creation of Courses. - CSEDU 2014 Barcellona, Spagna 2 -4 April 2014

Whoteach–RS Why the need of a Recommender System? Teaching requires the organization of effective ad hoc courses – Gathering adequate didactic resources To avoid frustration And waste of time Guide the user in the labyrinth of didactic material Main Target supporting and guiding the teacher in the organization of his/her courses Through 1. The reduction of the high number of available resources 2. The reduction of time and exhausting attempts to search teaching material 3. The creation of an effective and high-level course 13

WHOTEACH RESEARCH PROJECT OVERVIEW Main goal is to realize an original and innovative platform “Whoteach” with distinguishing components: Knowledge, Social Network, Intelligence. CASE STUDY &TARGET NETT EU PROJECT : “the aim of gathering a Social Network for improving the entrepreneurship teaching in the European educational system…. ” TARGET: demanding users who want to learn from teacher or expert 5

www. whoteach. it Creation of learning materials and classification using metadata and ontology from IEEE Social Intelligent Learning Management System Quality control of the materials made available to the community exchanging, sharing, commenting learning materials in a social context Recommender System to create new course using materials developed by other colleagues M. Mesiti, S. Valtolina, EPIFANIA F. , B. Apolloni (2014). Metaservizi per la produzione collaborativa di moduli didattici in ambienti sociali. TD TECNOLOGIE DIDATTICHE, ISSN: 1970 -061 X 6

Artificial Intelligence Recommender Systems (RSs) Recommendation Engine is your companion and advisor to help you make the right choices by providing you tailored options and creating a personalized experience for you. Michele Ballardin - Sviluppo di una piattaforma online per la valutazione di Recommender Systems 7

Whoteach–RS What’s new w. r. t. Classical rs? 1. Interactive ML paradigm Classical ML Data Interactive ML One-pass process Iterative process No users’ feedback Users control the behavior Long time to train the model Latency sensitive for training Numerical evaluation Friendly visualization eval. Model Human 4. Social Evaluation 2. Rule-based recommendation if-then rules • The if part states condition • The then part provides recommentation Why if-then rules? • Because they are one of the most expressive and humanreadable representation for hypotheses Course ratings by: • Experts • Teachers • Students suitably aggregated Apolloni et al. “Toward a human-centric intelligent didactical recommender system” 3. Fuzzy sets paradigm • Users are not crisp in their decisions and preferences • Different users interpret differently the same concepts not guys I like and not d so re lue like so b … me I agree Green is. best co the lor so far 1 4

www. whoteach. it THE OVERALL ARCHITECTURE Actors with different profiles Contributor Expert User e-Teaching Assistant ELearning system used to create courses by composing modules and didactic resources and to evaluate them Meta -services for filtering and validating modules and for fostering teachers’ involvement Browsing Recommender System Authoring Social Networks of Teachers and Students Evaluation Quality Search engine Metadata schema used for integrating heterogeneous LOs sources and for supporting the information retrieval strategy Local DB … DBs and Web Services distributed on the territory containing LOs DB DB WSs 2 9

OUTCOMES • Realization of a Advanced Learning Management System with: 1. 2. 3. 4. Advanced search engines Social Networks Quality control system Recommender System • Realization and Evaluation of a Recommender System (Epifania F. , P. Cremonesi (2012) User-centered evaluation of recommender systems with comparison between short and long profile CISIS 2012 IEEE; B. Apolloni, S. Bassis, M. Mesiti, S. Valtolina, F. Epifania. A Rule based recommender system WIRN 2015) 31

CONCLUSIONS • I have shown the requirement , design and realization of an innovative and original Social Intelligent Learning Management System • I have discussed a social Intelligent oriented solution for: – Exchanging high quality didactic materials – Creating courses through a computationally intelligent recommendation service • Realization of an innovative and original Recommender System (Apolloni et al 2015) – A modular, user-centric, interactive, highly interpretable RS conforming to the research project specifications and requirements – Analysis and evaluation • Who. Teach has already been tested for SIAM and Unimi, showing its potentialities in different learning environments. • We are developing and testing the scalability of the system in other contexts. 32

TEAM WWW. SOCIALTHINGUM. COM/CHI-SIAMO/ 12

Thank you and contact me ! Francesco Epifania, Ph. D. CEO of Social Things srl International Innovative ICT Startup http: //www. socialthingum. com S. O Viale Umbria 63, MIlano S. O Via Durando 38/a, Milano S. L. via De Rolandi 1 MIlano 333 3240632 02 2399 9153