Fall 2016
Data Science Seminar
Lecturers: Mourad Khayati
Teaching language: English
Level: MSc students
Academic year: Fall 2016
Overview
The seminar on data science involves presentations that cover recent topics on data science. The area of this year’s seminar is predictive analytics. In the scope of this seminar, we investigate papers that describe algorithms and techniques to perform prediction and trend analysis on different representations of data inputs, e.g., graphs, time series, etc.
Structure
The goal for the students is to learn how to critically read and study research papers, how to describe a paper in a report, and how to present it in a seminar. Under supervision, students will select one paper to study, contrast and compare with related work. This seminar aims to help students to gather in-depth knowledge of an advanced topic and develop the skills required to describe a complex problem from the predictive analytics area in the form of both a presentation and a written report.
IMPORTANT NOTE: The papers will be distributed on a first come first serve basis.
Evaluation and Expectations
The final grade depends on the quality of the report, presentation, reproducibility experiments (if any) and active participation during the seminar. Each participant prepares a self contained report of max 10 pages and gives a presentation of 20 minutes. The report should describe in detail the proposed technique(s). The report might contain a small running example, counter example(s) and should explore the extreme cases where the proposed approach would perform best and worst. The reproducibility experiments consists on running the available code of the proposed system and making a 5 min demo about it.
Advices on how to:
IMPORTANT NOTE: Attendance is mandatory for the two class seminar sessions. The total number of participants will be limited to 10.
Schedule
Kickoff Meeting. Date: Tue, 27.09.2016, 14:00-16:00, room: A303
Setup and organization of seminar, and paper assignment
The presentation can be accessed through this link.
----------------------------------------------------------------------
Date: Tue, 1.11.2016
Report deadline Batch1
Date: Tue, 8.11.2016, all day, room: B312
Office meeting with students from Batch1
First Seminar Session. Date: Tue, 15.11.2016, 14:15-18:00, room: A303
Presentations of Batch1
----------------------------------------------------------------------
Date: Tue, 29.11.2016
Report deadline of Batch2
Date: Tue, 6.12.2016, all day, room: B312
Office meeting with students from Batch2
Second Seminar Session. Date: Tue, 13.12.2016, 14:15-18.00, room: A303
Presentations of Batch2
----------------------------------------------------------------------
Date: Tue, 10.01.2017 extended to Fri, 13.01.2017
Deadline final Report of Batch1 and Batch2
Paper Assignment
The papers will be distributed on a first come first serve basis. Please use the following link to select one paper among the list of papers.
Paper | Presentation Date | Presenter | First Report Deadline | Final Report Deadline |
(1) Link Prediction in Graph Streams | 15.11.2016 | Michael Jungo | 1.11.2016 | 10.01.2017 |
(2) GeoScope: Online Detection of Geo-Correlated Information Trends in Social Networks | 15.11.2016 | Tofunmi Ajayi | 1.11.2016 | 10.01.2017 |
(3) Latent Space Model for Road Networks to Predict Time-Varying Traffic | 15.11.2016 | Antonios Chaidaris | 1.11.2016 | 10.01.2017 |
(4) SMiLer: A Semi-Lazy Time Series Prediction System for Sensors + reproducibility | 15.11.2016 | Michael Zbinden | 1.11.2016 | 10.01.2017 |
(5) Predictive Tree: An Efficient Index for Predictive Queries On Road Networks | 15.11.2016 | Sammer Puran | 1.11.2016 | 10.01.2017 |
(6) Dynamic and Robust Wildfire Risk Prediction System: An Unsupervised Approach | 13.12.2016 | Igor Dundic | 29.11.2016 | 10.01.2017 |
(7) Online Anomaly Prediction for Robust Cluster Systems | 13.12.2016 | Michaël Diatta | 29.11.2016 | 10.01.2017 |
(8) Deep Learning Architecture with Dynamically Programmed Layers for Brain Connectome Prediction | 13.12.2016 | Luka Hamza | 29.11.2016 | 10.01.2017 |
(9) GLMix: Generalized Linear Mixed Models For Large-Scale Response Prediction | 13.12.2016 | Maryam Sadeghimehr | 29.11.2016 | 10.01.2017 |
(10) Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering + reproducibility | 13.12.2016 | Reto Schiegg | 29.11.2016 | 10.01.2017 |