All school activities will be held in the Department of Mathematics and Mechanics, Ural state university (Tugeneva, 4). Lectures will be held in the Rm. 513, practical trainings — in Rm. 514, coffee breaks will be served in Rm. 507, luches — in the canteen in the ground floor.
|Sep 5, We||Sep 6, Th||Sep 7, Fr||Sep 8, Sa||Sep 9, Su||Sep 10, Mo||Sep 11, Tu||Sep 12, We|
|13.30-14.30||MSWS||YSC||YSC||city tour||YSC||search cup||departure|
Machine learning algorithms are widely used in web-related tasks, where due to the large scale and varying quality of data, adaptive techniques provide significant advantages over manual approaches. Examples of applications where learning methods have been very successful include learning ranking functions for search engines, detecting spam, clustering news articles, and learning hierarchies in online tagging systems. This course will provide a brief introduction into the general area of machine learning, show how important problems in web search and mining can be solved using machine learning techniques, and discuss problems and tradeoffs involved in applying machine learning approaches to web-scale datasets.
Andreas Rauber (Vienna University of Technology)
In this course we will take a closer look at the various areas, tasks, and methods that together form the field of music information retrieval (MIR).
We will start by considering the various types of data that are relevant for MIR activities, ranging from both symbolic as well as acoustic music data, via textual, up to image and video data. This will be followed by a brief overview of the overwhelming number of tasks and challenges in MIR to provide a thorough understanding of the problem domain and the interdisciplinary nature of this domain.
The core part of the course will then address a number of selected topics. Specifically, we will focus on various techniques for feature extraction from music, and their utilization for tasks such as retrieval, genre classification, chord detection, and others. We will also analyze and discuss the benefits of combining different modalities, such as textual and acoustic information, as well as the utilization of web information for these tasks. Last, but not least, we will take a closer look at a few applications, such as the PlaySOM and PocketSOM, that assist users in organizing their music collections, creating playlists on desktop computers as well as mobile phones. We will also review current music web portals and discuss future directions in music consumption and distribution.
The course will be acompanied by a range of practical exercises, allowing participants to analyze their own music collections and test the proposed mehods.
Alexander Sychov (Voronezh State University)
The course starts with the nature of the information retrieval in the context of the World Wide Web (WWW). Hyperlinks introducing in documents effects both documents representation and retrieval techniques. This effect is considered in the course. WWW formal description as directed graph, models and regularities are discussed. Further, the effect of hyperlink analysis on the relevance calculation and the crawling strategy is demonstrated. The additional topic of the course is the WWW self-organization and dynamics. For preliminary reading one may recommend presentation: http://company.yandex.ru/class/courses/sychev.xml
Nearest neighbors problem is formulated as follows: Given a set S of points in some space V (equipped with similarity function), construct a data structure which given any query point q from V finds the closest point in S to q. This kind of search problems arise in many areas: recommendation systems, text classification, personalized news aggregation and targeting on-line ads.
Course page (incl. slides and references)
Mikhail Ageev (Moscow State University)
This course will provide an introduction to the classical and modern problem statement for the text categorization tasks. We will show different techniques and methods for text categorization, based on machine learning and knowledge-based approach. We will also discuss the main problems of text categorization, which lead to erroneous categorization.
Igor Kuralenok (St.-Petersburg state university, Yandex)
See conference site for details.
Ivan Krasin (Google)
The lecture is dedicated to solving IR problems on large data sets. Map-Reduce technique is widely used by different subsystems of Google Search and makes it easy to implement programs which retrieve information from indexed web pages. The set of examples demonstrates use cases for Map-Reduce. The lecture ends with short talk about Google R&D in Russia.
Ilya Segalovich (Yandex)
1. What is morphology for in search tasks?
2. Mechanics beyond morphological analysis.
3. Dictionary-free morphology: a survey.
4. Applications: spell checker, Web search, etc.
Please send all inquiries to school[at]romip[dot]ru.