Conference Program – SISAP2021
14th International Conference on Similarity Search and Applications
Dortmund, Germany, Sept 29 – Oct 01, 2021
Times are given in CEST, Central European Summer Time.
Wednesday, September 29th
| 13:30 to 14:00 |
Social Hangoutpre-conference hangout in Zoom (open breakout rooms for discussion) |
| 14:00 to 14:15 |
OpeningChair: Erich Schubert, TU Dortmund |
Similarity Search and Retrieval 1Chair: Richard Connor, University of St Andrews, Scotland | |
| 14:15 to 14:55 |
Miriama Jánošová, David Procházka and Vlastislav Dohnal Organizing Similarity Spaces using Metric Hulls |
|
Fabian Fier and Johann-Christoph Freytag Scaling Up Set Similarity Joins Using A Cost-Based Distributed-Parallel Framework | |
|
Erich Schubert A Triangle Inequality for Cosine Similarity | |
|
Felix Borutta, Peer Kröger and Matthias Renz A Cost Model for Reverse Nearest Neighbor Query Processing on R-trees Using Self Pruning (short paper) | |
| Break | |
Intrinsic DimensionalitySession chair: Arthur Zimek, University of Southern Denmark, Denmark | |
| 15:15 to 15:55 |
Miloš Savić, Vladimir Kurbalija and Miloš Radovanović Local Intrinsic Dimensionality and Graphs: Towards LID-aware Graph Embedding Algorithms |
|
Stephane Marchand-Maillet, Oscar Pedreira and Edgar Chavez Structural Intrinsic Dimensionality | |
|
James Bailey, Michael E. Houle and Xingjun Ma Relationships between Local Intrinsic Dimensionality and Tail Entropy | |
|
Michael E. Houle and Ken-ichi Kawarabayashi The Effect of Random Projection on Local Intrinsic Dimensionality | |
| Break | |
Doctoral SymposiumSession Chair: Ilaria Bartolini, University of Bologna, ItalyDS talks are 20 minutes live, with 10 minutes of discussion for additional feedback. | |
| 16:15 to 17:45 |
Marco Postiglione Towards an Italian Healthcare Knowledge Graph |
|
Luca Zecchini Progressive Query-driven Entity Resolution | |
|
Fabrizio Scarrone Discovering Latent Information from Noisy Sources in the Cultural Heritage Domain | |
| 18:00 to 19:00 |
Poster SessionPoster session in SISAP Adventure. |
| 19:00 to 20:00 |
Social HangoutHangout in SISAP Adventure. BYOD: Bring your own Drinks |
Thursday, September 30th
| 13:00 to 14:00 |
Poster SessionPoster session in SISAP Adventure. |
| 14:00 to 14:10 |
Welcome Back, Day 2Chair: Erich Schubert, TU Dortmund |
Similarity Search and Retrieval 2Chair: Björn Þór Jónsson, IT University of Copenhagen, Denmark | |
| 14:15 to 14:55 |
Jakub Lokoc and Tomáš Souček How Many Neighbours for Known-item Search? |
|
Lucia Vadicamo, Claudio Gennaro and Giuseppe Amato On Generalizing Permutation-Based Representations for Approximate Search | |
|
Terézia Slanináková, Matej Antol, Jaroslav Oľha, Vojtěch Kaňa and Vlastislav Dohnal Data-driven Learned Metric Index: an Unsupervised Approach | |
|
Maximilian Hünemörder, Peer Kröger and Matthias Renz Towards a Learned Index Structure for Approximate Nearest Neighbor Search Query Processing (short paper) | |
| Break | |
Similarity Search and Retrieval 3Session chair: Lucia Vadicamo, ISTI-CNR, Italy | |
| 15:15 to 15:55 |
Tomas Skopal, David Bernhauer, Petr Skoda, Jakub Klimek and Martin Necasky Similarity vs. Relevance: From Simple Searches to Complex Discovery |
|
Jonatan Møller Nuutinen Gøttcke, Arthur Zimek and Ricardo José Gabrielli Barreto Campello Non-Parametric Semi-Supervised Learning by Bayesian Label Distribution Propagation | |
|
Omid Jafari, Preeti Maurya, Khandker Mushfiqul Islam and Parth Nagarkar Optimizing Fair Approximate Nearest Neighbor Searches using Threaded B+-Trees | |
|
Magnus Lie Hetland and Halvard Hummel Fairest Neighbors: Tradeoffs Between Metric Queries (short paper) | |
| Break | |
Similarity ApplicationsSession chair: Matthias Renz, Christian-Albrechts-Universität zu Kiel, Germany | |
| 16:15 to 16:55 |
Vladimir Mic, Tomáš Raček, Aleš Křenek and Pavel Zezula Similarity Search for an Extreme Application: Experience & Implementation |
|
Maciej Gawinecki, Wojciech Szmyd, Urszula Żuchowicz and Marcin Walas What makes a good movie recommendation? Feature selection for Content-Based Filtering | |
|
Fernando Luque, Jorge L. López-López and Edgar Chavez Indexed Polygon Matching under Similarities | |
|
Ahmed Askar and Andreas Züfle Clustering Adverse Events of COVID-19 Vaccines across the United States | |
| 19:30 to 21:00 |
Online Social Event & Virtual City TourZoom, Open Breakout Rooms and a virtual city tour.BYOB: Bring Your Own Beverages |
Friday, October 1st
| 7:00 to 8:00 |
Poster SessionPoster session in SISAP Adventure. |
| 13:00 to 14:00 |
Social Hangoutpre-conference hangout in Zoom (open breakout rooms for discussion) |
| 14:00 to 14:10 |
Welcome Back, Day 3Chair: Erich Schubert, TU Dortmund |
Clustering and ClassificationSession chair: Miloš Radovanović, University of Novi Sad, Serbia | |
| 14:10 to 14:40 |
Erich Schubert, Andreas Lang and Gloria Feher Accelerating Spherical k-Means |
|
Erik Thordsen and Erich Schubert MESS: Manifold Embedding Motivated Super Sampling | |
|
Jonatan Møller Nuutinen Gøttcke and Arthur Zimek Handling Class Imbalance in k-Nearest Neighbor Classification by Balancing Prior Probabilities | |
| Break | |
Special Session: Similarity Search in Graph-Structured Data (SISEG)Session chair: Nils Kriege, University of Vienna, Austria | |
| 15:00 to 16:15 |
Franka Bause, David B. Blumenthal, Erich Schubert and Nils M. Kriege Metric Indexing for Graph Similarity Search |
|
Lucas Gnecco Heredia, Nicolas Boria, Sébastien Bougleux, Florian Yger and David Blumenthal The Minimum Edit Arborescence Problem and Its Use in Compressing Graph Collections | |
|
Mathias Fuchs and Kaspar Riesen Graph Embedding in Vector Spaces Using Matching-Graphs | |
|
Benjamin Paaßen An A*-algorithm for the Unordered Tree Edit Distance with Custom Costs (short paper) | |
|
Jakub Peschel, Michal Batko, Jakub Valcik, Jan Sedmidubsky and Pavel Zezula FIMSIM: Discovering Communities by Frequent Item-Set Mining and Similarity Search (short paper) | |
|
Panel Discussion All presenters of the SISEG session will participate in a panel discussion. | |
| 16:15 to 17:00 |
Closing and AwardsSession chair: Erich Schubert, TU Dortmund |
