Research Seminars Archives >> 

November 

November,28th 

15:30

Room 55.309

Invited Research Seminar

Analyzing and Exploiting User-generated Listening Data for Listener Modeling and Music Recommendation

By Markus Schedl (Department of Computational Perception, Johannes Kepler University, Linz )

Host: Xavier Serra 

Abstract

Nowadays, music aficionados generate millions of listening events every day and share them via platforms such as Spotify, Last.fm, or Twitter. In 2016, the LFM-1b dataset (http://www.cp.jku.at/datasets/LFM-1b) containing more than 1 billion listening events of about 120,000 Last.fm users has been released to the research community and interested public. Since then, we performed various data analysis and machine learning tasks on these large amounts of user and listening data. The gained insights helped to develop new listener models and integrate them into music recommender systems, in an effort to increase personalization of the recommendations.
In this talk, I will report on our experiments with the LFM-1b dataset, focusing on the following topics:
- analyzing music taste around the world and clustering on the country level
- quantifying listener and country mainstreaminess
- music recommendation tailored to listener characteristics
- predicting user characteristics from music listening habits
- predicting country-specific genre preferences from cultural and socio-economic factors

November,29th 

15:30

Room 55.309

Invited Research Seminar

A brief 18 year career in music understanding

By Brian Whitman (formerly co-founder of The Echo Nest, Principal Scientist of Spotify)

Host: Xavier Serra 

Abstract

In 1999 I put a somewhat stalled electronic music career on hold to investigate how people were discovering new artists. I was especially interested in how very independent and niche music was getting noticed. Could the new methods of digital distribution at the time allow for greater scale and reach of all types of styles and genres of music? Could natural language processing, signal processing and machine learning play a role in understanding music? Two graduate degrees, a few stints in research labs, a startup, seventy employees, fifty customers, a large acquisition, 150 million active users of our technology and eighteen years later, I've left Spotify to focus on something new. I'm giving a brief talk to discuss what I've learned about the field, how big academic ideas translated into the marketplace, and what could be next. This will be a very informal group discussion.

November,30th 

15:30

Room 55.309

Invited Research Seminar

Competition policy and consumer outcomes - Assessing the impact of mobile consolidation on innovation and quality

By Xavier Pedrós  

Host: Miquel Oliver

Abstract

Competition authorities are required to assess mergers by looking at the likely effects of these on consumer outcomes. These include a range of characteristics that are important to consumers, such as price, quality and innovation. In practice, however, significant emphasis is put on (often short-term) prices, using tests that can easily trigger competition concerns in investment-intensive industries. Meanwhile, dynamic efficiencies and their possible effects on the quality of mobile services have often faced a strong burden of proof.

This seminar presents the relationship between market structure and consumer outcomes, focusing on the mechanisms by which a more concentrated market can deliver better quality and innovation outcomes. This seminar particularly presents an ex-post assessment of the impact of the 2012 Hutchison/Orange merger on innovation and quality. Using data on 4G networks rollout, download and upload speeds, this work shows that mobile mergers can induce significant dynamic efficiencies and realise direct benefits to consumers. This has relevant implications for competition policy and how quality aspects and efficiencies are taken into account.

Biography

Xavier Pedrós is Economist at GSMA Intelligence. GSMA Intelligence is the analysis and research arm of GSMA, an international organization of mobile operators. Xavier has carried out research in the area of competition and regulatory economics, with a focus on the impact of market structure and competition in mobile markets. He has co-authored a recently published assessment of the impact of mobile mergers on quality and innovation. He has also produced research in the areas of fiscal policy in mobile markets and assessments of the economic impact of mobile telecommunications. Xavier holds a bachelor degree in Political Science at Pompeu Fabra University and a Master´s degree on Economics, with a focus on Industrial Organization and Finance, at Université Catholique de Louvain-la-Neuve.

February,9th 

15:30

Room 55.309

Course (6 hours: 9th, 15th and 22nd of February)  

Phd Seminar: Statistical course and Design of Experiments

By Simone Tassani

The course of statistics aims to introduce a number of tools for master/Ph.D. students and post docs. The presented tools will play a role in planning many kind of studies, properly analyse the results and understand if data analysed by other researchers are or not reliable.

The course will start with a brief digression over the several implications that bad statistics have today over the scientific society and why every researcher should know the basic concepts behind a statistical analysis.

 

February,15th 

15:30

Room 55.309

Course (6 hours: 9th, 15th and 22nd of February) 

Phd Seminar: Statistical course and Design of Experiments

By Simone Tassani

The course of statistics aims to introduce a number of tools for master/Ph.D. students and post docs. The presented tools will play a role in planning many kind of studies, properly analyse the results and understand if data analysed by other researchers are or not reliable.

The first part of the course will introduce General Linear Modelling and its most common applications:

  1. Linear regression

  2. T-test

  3. F-test

  4. Analysis of Variance (ANOVA)

Monofactorial and multifactorial analysis will be presented, together with the definition of Type I and Type II error, multiple comparison errors and tests for multiple comparison.

In the case of multifactorial analysis, the concept of interaction among the factors will also be presented and how to use interaction as estimator of the error in absence of repetitions.

This will lead to the presentation of some examples of Design of Experiment (Latin and Greek-Latin squares) for the reduction of the number of experiments.

February,22nd 

15:30

Room 55.309

Course (6 hours: 9th, 15th and 22nd of February) 

Phd Seminar : Statistical course and Design of Experiments

By Simone Tassani

The course of statistics aims to introduce a number of tools for master/Ph.D. students and post docs. The presented tools will play a role in planning many kind of studies, properly analyse the results and understand if data analysed by other researchers are or not reliable.

The second part of the course will apply the theory presented in the first part in order to present different kind of advanced Design of Experiments.

Power analysis for the a-priori identification of the required sample size. The complete and fractioned factorial projects, the “confounding” i.e. concept of confused effect, determination of blocs in functions of confused effects and presentation of the Taguchi’s Method.

The course will than introduce the analysis with more than 2 levels and how to evaluate linear and quadratic component of the data.

It will finally conclude with the Response Surface Methodology for the analysis of continuous variables.

If the seminar is ofered via streaming in: 

- Room 55.309 follow this link

- Auditorium follow this link

If you are interested in giving a Research Seminar or you would like to invite a speaker please fill in the following form RSDetails Form .