Universit├Ąt Politehnica of Bucarest - Faculty of Automatic Control and Computers

Contact person:  Mihai Dascalu 

mikedascalu [at] yahoo.com

was head of class in 2009 (i.e., GPA 10/10; ranked 1st across specialization and university) at UPB, and currently holds two Master’s degrees, one in Internet systems engineering, UPB, and one in knowledge extraction, University of Nantes, and a double PhD with the highest distinctions in computer science (Excellent, UPB) and educational sciences (Très Honorable avec Felicitations, University Grenoble Alpes, France), with his thesis published as book in Springer, Studies in Computational Intelligence. He has experience in national and international research projects (H2020 RAGE - project director, FP7 LTfLL, FP7 ERRIC and CNCSIS K-TEAMS) with more than 80 published papers, including top computer education conferences (AIED, ITS, CSCL), renowned international conferences (ICALT, ICTAI, EC-TEL, ICWL, ISPDC, AIMSA) and journals (ijCSCL, Elsevier Computers in Human Behavior). Currently Mihai is a lecturer at UPB, responsible for the courses of Object Oriented Programming, Semantic Web Applications, and Data Mining and Data Warehousing. Complementary to his competencies in NLP, technology-enhanced learning (TEL) and discourse analysis, Mihai holds a multitude of professional certifications (e.g. PMP, PMI-RMP, PMI-ACP, CBAP, CISA, C|EH and CISSP) and extensive experience on strategic projects on non-refundable funds (EU, WB, USTDA). Moreover, Mihai has received the distinction “IN TEMPORE OPPORTUNO” in 2013 as the most promising young researcher in UPB and has obtained a Senior Fulbright scholarship.



Description of the smart city learning group operating in the affiliated institution

The University Politehnica of Bucharest (UPB) is the largest technical university in Romania, having over 22.000 students, with more than 2.000 students enrolled in the Computer Science and Engineering Department. The department also runs the National Center for Information Technology (NCIT), which is an EU Center of Excellence (EU-NCIT). It has 18 full professors and over 40 researchers and PhD students. The mission of NCIT is to promote advanced and inter-disciplinary research, to sustain the development of human resources in postgraduate educational programs. NCIT’s activity relies on a collaborative virtual environment using high-performance computing resources and computer-supported cooperative work tools. NCIT is actively involved in international cooperation with similar centers, partnerships with IT companies and development of national and international projects (ERRIC, LTfLL, EU-NCIT, COOPER, SINTEC, CoLaborator, IKF, AGCOR, EGEE, SEE-GRID, RoGRID, CODESTAR, REASON).




The UPB group has a broad research experience in Natural Language Processing (NLP), discourse analysis (DA) and learning analytics (LA), as well as Computer Supported Collaborative Learning (CSCL), marking the inter-disciplinary orientation of the ongoing researches. All the previous domains are directly linked with the scope of smart city learning by enabling a deeper understanding of discourse and of interactions among members.



Relevant achievements, best practices, products


ReaderBench – An educational software that uses natural language processing and text mining technologies to allow both tutors and learners to regularly track learning progress, providing multilingual support for English, French, and partially for Italian. The main areas covered by our system are derived from a unitary cohesion-based representation of discourse used to: a) identify reading strategies, b) evaluate textual complexity, and c) assess participation and collaboration. Developed in collaboration with University of Grenoble.

PolyCAFE – Software for providing automatic feedback and visualization services used for analyzing online chat and discussion forums in education contexts. Developed within the FP7 LTfLL project, works only for English.



On-going challenges


From a technical perspective, current challenges are being tackled with the development of new tools and their capabilities. The external resources made available to students and the collaboration frameworks could be automatically evaluated, in an effort to find ways to streamline tools (such as discourse analysis tools) and make them available to the student so they can find their own external resources (such as appropriate virtual communities). Potential further research questions are: How do students accept and use online communities as resources for knowledge sharing and construction? How does this discourse work and what is its effect on student’s academic performance? Therefore, we are investigating a design based research model that could change the design of the annotation collaboration framework to enhance the acceptance and continuance of the task, as well as stimulating a community.

Moreover, based on an in-depth discourse analysis framework we are striving to:

• highlight the problems associated with the bottom-up approaches currently used to benchmark city smartness and, as well, in the production of rankings of any sort;

• propose an alternative conceptual framework to monitor and benchmark the appeal and smartness of cities and territories based on the concept of flow derived from online published textual materials.



Relevant publications


• Dascalu, M., Dessus, P., Bianco, M., & Trausan-Matu, S. (2014). Are Automatically Identified Reading Strategies Reliable Predictors of Comprehension? In S. Trausan-Matu, K. E. Boyer, M. Crosby & K. Panourgia (Eds.), 12th Int. Conf. on Intelligent Tutoring Systems (ITS 2014) (pp. 456–465). Honolulu, USA: Springer.


• Dascalu, M., Stavarache, L.L., Trausan-Matu, S., Dessus, P., & Bianco, M. (2014). Reflecting Comprehension through French Textual Complexity Factors. In 26th Int. Conf. on Tools with Artificial Intelligence (ICTAI 2014) (pp. 615–619). Limassol, Cyprus: IEEE.


• Giovannella, C., Dascalu, M., & Scaccia, F. (2014). Smart City Analytics: state of the art and future perspectives. Interaction Design and Architecture(s) Journal – IxD&A, 20, 72–87. 


• Dascalu, M. (2014). Analyzing discourse and text complexity for learning and collaborating, Studies in Computational Intelligence (Vol. 534). Switzerland: Springer.


• Oca, A.M. Montes de, Nistor, N., Dascalu, M., & Trausan-Matu, S. (2014). Designing Smart Knowledge Building Communities. Interaction Design and Architecture(s) Journal – IxD&A, 22, 9–21. 


• Trausan-Matu, S., Dascalu, M., & Rebedea, T. (2014). PolyCAFe – Automatic support for the analysis of CSCL chats. International Journal of Computer-Supported Collaborative Learning, 9(2), 127–156. doi: 10.1007/s11412-014-9190-y


• Dascalu, M., Trausan-Matu, S., Dessus, P., & McNamara, D.S. (2015). Discourse cohesion: A signature of collaboration. In 5th Int. Learning Analytics & Knowledge Conf. (LAK'15) (pp. 350–354). Poughkeepsie, NY: ACM.


• Nistor, N., Trausan-Matu, S., Dascalu, M., Duttweiler, H., Chiru, C., Baltes, B., & Smeaton, G. (2015). Finding student-centered open learning environments on the internet: Automated dialogue assessment in academic virtual communities of practice. Computers in Human Behavior, 47(1), 119–127. doi: doi:10.1016/j.chb.2014.07.029


• Dascalu, M., Stavarache, L.L., Dessus, P., Trausan-Matu, S., McNamara, D.S., & Bianco, M. (in press). Predicting Comprehension from Students’ Summaries. In 17th Int. Conf. on Artificial Intelligence in Education (AIED 2015). Madrid, Spain: Springer.


• Dascalu, M., Trausan-Matu, S., Dessus, P., & McNamara, D.S. (in press). Dialogism: A Framework for CSCL and a Signature of Collaboration. In 11th Int. Conf. on Computer-Supported Collaborative Learning (CSCL 2015). Gothenburg, Sweden: ISLS.