Open Lectures

Challenges in Real-time Social
Media Analytics: What has changed

Abstract. We started research on social media analytics over 5 years ago. The essential focus of research at that time was live, big data, multimodal, multi-source and multilingual. It seems that 5 years have moved on, while the key research issues are still the same. So what has fundamentally changed? 5 years ago, the focus of social network service (SNS) is on user participation, interactions, co-viewing and co-authoring of contents, with research on analysis of SNS with the aim to understand, share and recommend. Now the key focus is still the same but with more emphasis on real-time aspect, which permits instant sharing and feedback. In fact the ability for real-time feedback by users triggers the possibility of co-creating and co-inventing of not just the content, but also the systems design and functionalities. Several examples are now available towards this direction. The key new research will correspondingly move beyond analysis, but more on predictive and prescriptive analytics.

Tat-Seng CHUA
National University of Singapore

Dr Chua is the KITHCT Chair Professor at the School of Computing, National University of Singapore. He was the Acting and Founding Dean of the School during 1998-2000. Dr Chua's main research interest is in multimedia information retrieval and social media analysis. In particular, his research focuses on the extraction, retrieval and question-answering (QA) of text, video and live media arising from the Web and social networks. He is the Director of a multi-million-dollar joint Center (named NExT) between NUS and Tsinghua University in China to develop technologies for live media search. The project will gather, mine, search and organize user-generated contents within the cities of Beijing and Singapore.

Dr Chua is the 2015 winner of the ACM SIGMM award for Outstanding Technical Contributions to Multimedia Computing, Communications and Applications. He is the Chair of steering committee of ACM International Conference on Multimedia Retrieval (ICMR) and Multimedia Modeling (MMM) conference series. Dr Chua is also the General Co-Chair of ACM Multimedia 2005, ACM CIVR (now ACM ICMR) 2005, ACM SIGIR 2008, and ACM Web Science 2015. He serves in the editorial boards of 4 international journals.

Dr Chua is the independent Director of a publicly listed company, and co-Founder of two technology startup companies in Singapore. He holds a PhD from the University of Leeds, UK.

Generation and analysis of speech sounds: a new approach

Abstract. The presentation is concerned with the specification and improvement of the traditional source-filter model of the human vocal tract proposed by G.Fant and analyzed by many scientists. The new method of recording the glottal wave synchronously with an output speech signal was employed to obtain the experimental material. The comparison of the recorded signals allowed analyzing the structure of the speech signal at different stages of its generation. As a result, the classic vocal tract model was specified by distinguishing a feedback component which formalizes the processes in the vocal tract as a complex acoustic nonlinear system. One of the functions of the component is to transform the acoustic energy from the articulation system upstream. The proposed approach is perspective for the synthesis of speech sounds with predefined acoustic properties that would be as simple and effective as formant synthesis.
It shows new perspectives of research on and description of the individual voice characteristics (speaker identification and verification tasks).
In our presentation the recording method will be described, perceptual experiment and the acoustic analysis results will be presented, the generated signals will be demonstrated.

Vera Evdokimova, Pavel Skrelin

Pavel Skrelin, Doctor of Philology, Professor, Head of the Department of Phonetics and Methods of Foreign Language Teaching of the St. Petersburg State University.
Member of the International Speech Communication Association, coordinator of its Special Interest Group on Russian speech analysis (SIGRU), member of the North-European Association for Language Technology. Member of the program committee of the following regular international conferences: TSD (Text, Speech, Dialogue), SPECOM (Speech and Computer), Phonetics without Borders.

Vera V.Evdokimova
PhD Theory of language
Senior teacher at the Department of Phonetics
Head of the Laboratory of Experimental Phonetics
Saint-Petersburg State University
Scientific interests: General and applied linguistics, Phonetics, Phonology, Speech acoustics, Voice source generation, Phonation, Automatic Speech synthesis, Speech recognition, Speaker identification and verification, Forensic speaker identification, Speech fatigue, Phonetic properties of singing,
Emotion recognition, Automatic phonetic transcribing system for Russian

Criminal Profiling in police practices,
science and fiction: from the origins to
future perspectives

Abstract. The inference of personal characteristics of the unknown author of a crime, relying on behavioral data coming mostly from the crime scene, is a hard and fascinating challenge for police forces, academic researchers and practitioners. The studies concerning the criminal behavior and the techniques developed with such purpose, generally known as Criminal Profiling, affected not only the police practices but also the collective imaginary about criminal analysis. But how well grounded is their theoretical frameworks, and how concretely useful the discipline turned out to be in criminal investigations? This talk summarises the development of the Criminal Profiling from its origins, considering the different approaches and outcomes of the main schools of profilers. Problems, opportunities and perspectives for the future will be discussed, with particular reference to possible Artificial Intelligence applications.

Tommaso Fornaciari
Italian National Police

Tommaso Fornaciari is a Police Officer Psychologist of Italian National Police. Since 2003 he worked at the Forensic Science Police Service, dealing with crime scene analysis, behavioral analysis and investigative data analysis, mostly regarding bloody murders. With the purpose of supporting the analysis of testimonies, in 2009 he began to attend the PhD school of the Center for Mind/Brain Sciences - CIMeC of the University of Trento, where he carried out a research project in forensic linguistics (Ph.D. in Cognitive and Brain Sciences, 2012). In particular, he applied computational techniques in order to detect deception in transcripts of hearings held in Italian Courts. He is going ahead with research in deception detection and currently he works at the Italian Ministry of Interior, where he is engaged in research and technological innovation for public security.

Topic modeling as a key technology for exploratory search and social media mining

Abstract. Topic modeling is a powerful tool for revealing an underlying thematic structure of large text collections. We consider topic modeling as a key technology for exploratory search, a rapidly evolving paradigm in information retrieval. In this application topic model must be well interpretable by humans, multimodal, multilingual, multigram, semi-supervised, fine-grained, hierarchical, temporal, incremental, etc. We develop a non-Bayesian regularization approach (ARTM) to satisfy a large number of requirements at once and build multi-objective topic models. We present the BigARTM.org open source project for regularized multimodal topic modeling of large collections. In our experiments, we show that BigARTM outperforms other popular packages in quality, runtime and functionality. As an application of ARTM we consider a project for semiautomatic monitoring of interethnic relations based on social media data.

Konstantin Vorontsov
Professor, Moscow Institute of Physics and Technology, Yandex School od Data Analysis.

The current state of m-Health : physician’s
and patient’s view.

Prof. Oleg Medvedev
School of Medicine, Lomonosov Moscow State University, Fruct MD

Head, Working Group “Telemedicine and Health Remote Monitoring” of the Ministry of Health of the Russian Federation.
Graduated from the I.Pavlov 1st Leningrad Medical Institute
Since 1992: Dean, Pharmacology Department, Fundamental Medicine Faculty, Lomonosov Moscow State University
During 10 years, he worked in the Institute of Pharmacology of the Academy of Medical Sciences and headed Pharmacology Department of the All-Union Cardiac Centre under the guidance of academician E.Chazov.
Oleg Medvedev’s scientific interests include pharmacology of cardiovascular system and antioxidants.
He authored over 200 publications in the leading national and foreign scientific journals. He renders joint research in USA, Japan, Germany, France, Australia and other countries.
Doctor of Medicine, Professor