Publications

This volume contains the refereed proceedings of the 6th International Conference on Analysis of Images, Social Networks, and Texts (AIST 2017)1. The previous conferences during 2012–2016 attracted a significant number of students, researchers, academics, and engineers working on interdisciplinary data analysis of images, texts, and social networks.   The broad scope of AIST made it an event where researchers from different domains, such as image and text processing, exploiting various data analysis techniques, can meet and exchange ideas. We strongly believe that this may lead to cross fertilisation of ideas between researchers relying on modern data analysis machinery. Therefore, AIST brought together all kinds of applications of data mining and machine learning techniques. The conference allowed specialists from different fields to meet each other, present their work, and discuss both theoretical and practical aspects of their data analysis problems. Another important aim of the conference was to stimulate scientists and people from industry to benefit from the knowledge exchange and identify possible grounds for fruitful collaboration.   The conference was held during July 27–29, 2017. The conference was organised in Moscow, the capital of Russia, on the campus of Moscow Polytechnic University.   This year, the key topics of AIST were grouped into six tracks: 1. General topics of data analysis chaired by Sergei Kuznetsov (Higher School of Economics, Russia) and Amedeo Napoli (LORIA, France) 2. Natural language processing chaired by Natalia Loukachevitch (Lomonosov Moscow State University, Russia) and Alexander Panchenko (University of Hamburg, Germany) 3. Social network analysis chaired by Stanley Wasserman (Indiana University, USA) 4. Analysis of images and video chaired by Victor Lempitsky (Skolkovo Institute of Science and Technology, Russia) and Andrey Savchenko (Higher School of Economics, Russia) 5. Optimisation problems on graphs and network structures chaired by Panos Pardalos (University of Florida, USA) and Michael Khachay (IMM UB RAS and Ural Federal University, Russia) 6. Analysis of dynamic behaviour through event data chaired by Wil van der Aalst (Eindhoven University of Technology, The Netherlands) and Irina Lomazova (Higher School of Economics, Russia)   One of the novelties this year was the introduction of a new specialised track on process mining (Track 6).

In this paper we show that for a given co-authorship network we could construct a recommender system for searching collaborators with similar research interests defined via keywords and topic modelling. We suggest new link embedding method and evaluate our model on National Research University Higher School of Economics (NRU HSE) co-authorship network.

Co-authorship networks contain invisible patterns of collaboration among researchers. The process of writing joint paper can depend of different factors, such as friendship, common interests, and policy of university. We show that, having a temporal co-authorship network, it is possible to predict future publications. We solve the problem of recommending collaborators from the point of link prediction using graph embedding, obtained from co-authorship network. We run experiments on data from HSE publications graph and compare it with relevant models.

This paper is devoted to mathematical modelling of the progression considering stages of breast cancer. Given the relation between primary tumor (PT) and metastases (MTS), the problem of discovering breast cancer (BC) process seems to be twofold: firstly, it is im- portant to describe the whole natural history of BC to understand the process as a whole; secondly, it is necessary to predict the period of a clinical MTS manifestation. In order to understand growth processes of BC on each stage CoMBreC was proposed as a new research tool. The CoMBreC is threefold: CoMPaS (stages I-II), CoM-III (stage III) and CoM-IV (stage IV). A new model rests on exponential growth model and complementing formulas. For the first time, it allows us to calculate different growth periods of PT and MTS in patients with/without lymph nodes MTS: 1) non-visible period for PT; 2) non- visible period for MTS; 3) visible period for MTS. Calculations via CoMBreC correspond to survival data considering stage of BC. It may help to improve predicting accuracy of BC process using an original mathematical model referred to CoMBreC and corresponding software. Consequently, thesis concentrated on: 1) modelling the whole natural history of PT and MTS in patients with/without lymph nodes MTS; 2) developing adequate and precise CoMBreC that reflects relations between PT and MTS; 3) analysing the CoMBreC scope of application. The CoMBreC was implemented to iOS application as a new predictive tool: 1) is a solid foundation to develop future studies of BC models; 2) does not require any expensive diagnostic tests; 3) is the first predictor of survival in breast cancer that makes forecast using only current patient data.

The goal of this research is to improve the accuracy of predicting the breast cancer (BC) pro- cess using the original mathematical model referred to as CoMPaS. The CoMPaS is the original mathematical model and the corresponding software built by modelling the natural history of the primary tumor (PT) and secondary distant metastases (MTS), it reflects the relations between the PT and MTS. The CoMPaS is based on an exponential growth model and consists of a system of determinate nonlinear and linear equations and corresponds to the TNM classification. It allows us to calculate the different growth periods of PT and MTS: 1) a non-visible period for PT, 2) a non-visible period for MTS, and 3) a visible period for MTS. The CoMPaS has been validated using 10-year and 15-year survival clinical data con- sidering tumor stage and PT diameter. The following are calculated by CoMPaS: 1) the number of doublings for the non-visible and visible growth periods of MTS and 2) the tumor volume doubling time (days) for the non-visible and visible growth periods of MTS. The diameters of the PT and secondary distant MTS increased simultaneously. In other words, the non-visible growth period of the secondary distant MTS shrinks, leading to a decrease of the survival of patients with breast cancer. The CoMPaS correctly describes the growth of the PT for patients at the T1aN0M0, T1bN0M0, T1cN0M0, T2N0M0 and T3N0M0 stages, who does not have MTS in the lymph nodes (N0). Additionally, the CoMPaS helps to con- sider the appearance and evolution period of secondary distant MTS (M1). The CoMPaS correctly describes the growth period of PT corresponding to BC classification (parameter T), the growth period of secondary distant MTS and the 10-15-year survival of BC patients considering the BC stage (parameter M).

In this paper, we consider new formulation of graph embedding algorithm, while learning node and edge representation under common constraints. We evaluate our approach on link prediction problem for co-authorship network of HSE researchers’ publications. We compare it with existing structural network embeddings and feature-engineering models.

Formal concepts and closed itemsets proved to be of big importance for knowledge discovery, both as a tool for concise representation of association rules and a tool for clustering and constructing domain taxonomies and ontologies. Exponential explosion makes it difficult to consider the whole concept lattice arising from data, one needs to select most useful and interesting concepts. In this paper interestingness measures of concepts are considered and compared with respect to various aspects, such as efficiency of computation and applicability to noisy data and performing ranking correlation.

Nowadays,  mind  mapping  is  rather  popular  educational  technique.  Like any other learning tools, mind maps became  a part of modern educational  trends like blended learning and computer-supported collaborative learning. Lots  of mind mapping software tools are adopted to teaching and learning routines  such as educational content delivery or assessment. This paper focuses on the  additional automatic evaluation of digital educational mind maps gained from the  existing procedures of assessments. The review of automatic graders which support the evaluation process demonstrates that some systematical work is done in  automation grading by comparing students’ mind maps with a template. But lots  of questions about automatic mind maps’  scoring by retrieving  the  data from a  scored mind map are still open. This paper introduces the automatic grader for  educational mind maps (AGEMM) which acts like a teacher’s assistant and calculates several quantitative metrics. The AGEMM is implemented as a web-service and interacted with mind maps prepared in the Coggle web-service through  its API. The AGEMM is adopted to a bachelor course. Results demonstrate that  scores from the AGEMM may be transformed to scales or criterial levels which  are  used  to  evaluation.  Moreover,  the  AGEMM  application  revealed  several  problems and shew lines of development which we discuss in the paper.

In this paper, we present novel winning team predicting models and compare the accuracy of the obtained prediction with TrueSkill model of ranking individual players impact based on their impact in team victory for the two most popular online games: Dota 2 and Counter-Strike: Global Offensive.

This volume contains proceedings of the first Workshop on Data Analysis in Medicine held in May 2017 at the National Research University Higher School of Economics, Moscow. The volume contains one invited paper by Dr. Svetla Boytcheva, 6 regular contributions and 2 project proposals, carefully selected and reviewed by at least two reviewers from the international program commit- tee. The papers accepted for publication report on different aspects of analysis of medical data, among them treatment of data on particular diseases (Consoli- dated mathematical growth model of Breast Cancer CoMBreC, Artificial neural networks for prediction of final height in children with growth hormone deficiency), methods of data analysis (analysis of rare diseases, methods of machine learning and Big Data, subgroup discovery for treatment optimization), and instrumental tools (explanation-oriented methods of data analysis in medicine, information support features of the medical research process, modeling frame- work for medical data semantic transformations, radiology quality management and peer-review system). Organizers of the workshop would like to thank the reviewers for their careful work and all contributors and participants of the workshop.

In this talk, we show a realistic post-processing rendering based on generative adversarial network CycleWGAN. We propose to use CycleGAN architecture and Wasserstein loss function with additional identity component in order to transfer graphics from Grand Theft Auto V to the older version of GTA video-game, Grand Theft Auto: San Andreas. We aim to present the application of modern art style transfer and unpaired image-to-image translations methods for graphics improvement using deep neural networks with adversarial loss.

Co-authorship networks contain hidden structural patterns of research collaboration. While some people may argue that the process of writing joint papers depends on mutual friendship, research interests, and university policy, we show that, given a temporal co-authorship network, one could predict the quality and quantity of future research publications. We are working on the comparison of existing graph embedding and feature engineering methods, presenting combined approach for constructing co-author recommender system formulated as link prediction problem. We also present a new link embedding operator improving the quality of link prediction base don embedding feature space. We evaluate our research on a single university publication dataset, providing meaningful interpretation of the obtained results.

Modern co-authorship networks contain hidden patterns of researchers interaction and publishing activities. We aim to provide a system for selecting a collaborator for joint research or an expert on a given list of topics. We have improved a recommender system for finding possible collaborator with respect to research interests and predicting quality and quantity of the anticipated publications. Our system is based on a co-authorship network derived from the bibliographic database, as well as content information on research papers obtained from SJR Scimago, staff information and the other features from the open data of researchers profiles. We formulate the recommendation problem as a weighted link prediction within the co-authorship network and evaluate its prediction for strong and weak ties in collaborative communities.

Modern medicine aspire to improve the effectiveness of treatment for some diseases through, so called, personalized medicine. However, totally personalized medicine or personalized treatment of even one disease is a very ambitious goal. Subgroup analysis of patients is a preliminary step to the total personalization. Several completely different views on the principles and usefulness of subgroup analysis for treatment personalization exist. This paper is limited to data-driven subgroup discovery, when collected data analyzed for significant treatment-biomarker interactions in post-hoc manner, and presents a brief overview of key methods for this type of subgroup analysis.

In clinical trials comparing experimental and control treatment the effect of treatment often depends on the range of patient’s characteristics (biomarkers) such as clinical, anthropological, genetic, psychological, social characteristics and others. Personalized medicine aims at finding such dependencies to tailor treatment strategies to a patient. This paper presents an overview of the approaches to data analysis of clinical trials intended for identification of influential biomarkers and subgroups of patients, where experimental and control treatment differ significantly in efficiency.