Oberseminar: Data Mining

About the oberseminar

Many sessions take place in form of a reading group, where we discuss the assigned paper(s). Everyone should read the paper(s) before the meeting to be able to actively participate in discussions. Sometimes we have talks by guests, PhD students or MSc and BSc students.

Feel free to join, for a detailed schedule and up-to-date information please register via StudIP.



Organisation

• Contact person: Prof. Dr. Eirini Ntoutsi
• Place: Library, Appelstrasse 4, 2nd floor (Building: 3703).
• Time: 10:00-11:00 (during the semester, see detailed schedule below)

Schedule WS19/20

• Friday 18.10.2019: Dry run bachelor thesis presentation by Philip Ossenkopp on ``Dealing with Concept Drifts via Weightless Neural Networks'' & discussion on the related paper WCDS: A Two-Phase Weightless Neural System for Data Stream Clustering.
• Friday 25.10.2019: CIKM paper presentation by Vasilis on AdaFair: Cumulative Fairness Adaptive Boosting.
• Friday 1.11.2019: Bachelor thesis presentation by Philip Ossenkopp on ``Dealing with Concept Drifts via Weightless Neural Networks''.
• Friday 8.11.2019: Paper discussion: KNN classifier with self adjusting memory for heterogeneous concept drift.
• Friday 15.11.2019: no meeting (Workshop in Bielefeld)
• Friday 22.11.2019: no meeting (ITN coordinator's day in Brussels)
• Friday 29.11.2019: Amir's ongoing work & paper discussion on ensemble learning over streams: (1) A Drift-based Dynamic Ensemble Members Selection using Clustering and (2) Mitigating Concept Drift via Rejection.
• Friday 06/12/2019: Invited talk Maryam Tavakol: Contextual and Sequential Recommendations.
• Friday 13.12.2019: no meeting (Workshop in Hannover)
• Friday 20/12/2019: Arjun's ongoing work & paper discussion on dynamic update of word embeddings: (1) Dynamic Word Embeddings and (2) Dynamic word embedding for semantic discovery.
• Friday 10/01/2019: Paper discussion: Evolving Fuzzy Min–Max Neural Network Based Decision Trees for Data Stream Classification .
• Friday 17/01/2020: Invited talk: Alvaro Veizaga Campero: ML for requirement engineering specifications.
• Friday 24/01/2020: Tai ongoing work
• Friday 30/01/2020: Felipe ongoing work

Schedule SS19

• Friday 19/07/2019: Dry run DEXA presentation by Vasilis on ``Fairness-enhancing interventions in stream classification''
• Friday 12/07/2019: Dry run ECMLPKDD workshop presentation by Tai on ``A Neighborhood-augmented LSTM Model forTaxi-Passenger Demand Prediction''
• Friday 05/07/2019: We meet at 14:00 at the Multimedia Room, 15th Floor, L3S to attend the Just Machine Learning talk by Prof. Dr. Tina Eliassi-Rad.
• Friday 28/06/2019: Paper discussion TBA
• Friday 14.06.2019: We meet at 14:00 at the Multimedia Room, 15th Floor, L3S to attend the Designing human-machine collaborations – how to effectively interweave human and algorithmic activities talk by Prof. Claudia Müller-Birn.
• Friday 07.06.2019: We meet at 14:00 at the Multimedia Room, 15th Floor, L3S to attend the Mosaics in Big Data talk by Prof. Volker Markl, TU Berlin.
• Friday 31.05.2019: Dry run bachelor thesis presentation by Niels on ``Detecting tendon failure from acoustic emission data''
• Friday 24.05.2019: Paper discussion Heterogeneous Ensemble for Feature Drifts in Data Streams
• Friday 17.05.2019: Paper discussion Interpretable Predictions of Tree-based Ensembles via Actionable Feature Tweaking
• Friday 26.04.2019: Tai presents his work on mobility prediction. Bahman his results from the NYC taxi dataset.
• Friday 03.05.2019: Paper discussion Matrix profile and Super efficient cross-correlation algorithm (Eirini at SDM).
• Friday 12.04.2019: Paper discussion A Confidence-Based Approach for Balancing Fairness and Accuracy