Lectures

Lectures

Lectures Held on a Regular Basis

Please click here to find the lectures that are held on a regular basis. This information can be used to create a study plan. However, this information is tentative, i.e. always check the lectures of the current and next semester to keep the study plan up to date.

Lectures of the Current Semester

Module details on 'Advanced Topics in Data Mining I - Advanced Topics in Data Mining I'

CategoryData & Information
TypeSeminar
SiteHannover
LecturerProf. Dr. Ntoutsi, Eirini (Hannover)
Module Exam ID2007
ECTS-Credits5
Weekly Composition1
Required Hours of Work (presence / self-study)90 (90)
SemesterWinter
Teaching MethodsEach student chooses a research paper from a pool of related to the seminar’s topic papers. Students are expected to carefully read the paper as well as related work necessary to comprehend the topic. For each student there will be an advisor who guides the whole process and helps the student in case of difficulties and questions. The students are expected to write a report of their research topic and present their findings to the class. We will also include a peer evaluation component for both the report and the oral presentation in order to allow students to test their skills on how to review scientific work and scientific talks. There will be a few group meetings and regular meetings with the advisors through the semester. It is obligatory for the students to participate in those meetings and actively engage in discussions and Q&A sessions. The seminar requires a regular communication with the advisor. A preliminary schedule is as follows: 1) (class) Introductory lecture to the selected topic and presentation of the paper list. 2) (online) Paper selection (students send their preferences per email) and final paper assignment (per email). 3) (class) An introduction to the academic work on how to write, publish, and present your results. 4) (class) Short presentation of each paper to get a first idea of what everyone is working on. 5) (personal regular meetings) Meetings with students to discuss on their selected papers. 6) (class) Final presentation and delivery of the report.
Module DescriptionThe goal of the seminar is threefold: i) to introduce students to a research area ii) to critically read and analyze scientific papers iii) to present a research area in both written (report) and spoken forms (presentation and Q&A sessions). This seminar is dedicated to the discussion of selected topics in data mining, like certain techniques (e.g., clustering, classification, …), evaluation, mining for certain data types (e.g., timeseries, trajectories, …), feature selection etc. Each semester we will focus on a certain topic, this year focus is on: "discrimination discovery and fairness-aware data mining", a timely topic given the ongoing discussion on the risks of Artificial Intelligence, Big Data and Automatic Decision Making for unfairness and discrimination.
Module Outcomes- Reading and comprehending scientific work. - Discussing scientific work, its pros and cons and how can be improved or extended. - Reviewing known DM algorithms in the big data era and understanding what are their limitations and how one can overcome them - Presenting to the group and being able to answer questions. - Exchange ideas with your mentor
Recommended Literature European Union regulations on algorithmic decision-making and a "right to explanation“, Goodman and Flaxman, 2016. https://arxiv.org/abs/1606.08813 IEEE Ethically aligned design, A Vision for Prioritizing Human Wellbeing with Artificial Intelligence and Autonomous Systems, 2016. http://standards.ieee.org/develop/indconn/ec/ead_v1.pdf A Course on Fairness, Accountability and Transparency in Machine Learning, course syllabus maintained by Suresh Venkatasubramanian https://geomblog.github.io/fairness/ Data, Responsibly, Dagstuhl seminar, Abiteboul, Miklau, Stoyanovich and Weikum, 2016 http://drops.dagstuhl.de/opus/volltexte/2016/6764/
PrerequisitesStudents must know basic concepts of data mining/ machine learning (for example, through the Data Mining I lecture)
ExamWritten Report and Oral Presentation (30 minutes)
CommentsThe report, the final presentation, participation in the class and consistency will be taken into account for the final grade. The report on the assigned paper should be based on the original paper and follow-up research and should also take into account the seminar discussions. The report (3-5 pages) should be written in LaTex.

Available Course Modes

In the following document you can get an overview about the available course modes that are offered in the ITIS Master's program: Course Modes