Authors: Giuseppe Amato (ISTI-CNR - Partner of AI4EU EU project, Italy), Vincent Oria (New Jersey Institute of Technology, USA), Miloš Radovanović (University of Novi Sad, Serbia)
Dates: October 2-4, 2019
Location: Newark, NJ (USA)
General Chair: Vincent Oria, New Jersey Institute of Technology (USA)
Program Committee Co-Chairs: Giuseppe Amato (ISTI-CNR - Partner of AI4EU EU project, Italy), Miloš Radovanovic (University of Novi Sad, Serbia)
The International Conference on Similarity Search and Applications (SISAP) is an annual forum for researchers and application developers in the area of similarity data management. It aims at the technological problems shared by numerous application domains, such as data mining, information retrieval, multimedia, computer vision, pattern recognition, computational biology, geography, biometrics, machine learning, and many others that make use of similarity search as a necessary supporting service.
From its roots as a regional workshop in metric indexing, SISAP has expanded to become the only international conference entirely devoted to the issues surrounding the theory, design, analysis, practice, and application of content-based and feature-based similarity search. The SISAP initiative has also created a repository serving the similarity search community, for the exchange of examples of real-world applications, the source code for similarity indexes, and experimental testbeds and benchmark data sets (http://www.sisap.org). The proceedings of SISAP are published by Springer as a volume in the Lecture Notes in Computer Science (LNCS) series.
The 2019 edition of SISAP was held at the New Jersey Institute of Technology in Newark, New Jersey, USA. Newark is an attractive location in the New York City metropolitan area with easy and convenient travel to and from the conference. The organization was smooth and with a strong technical program assembled by two co-chairs and sixty program committee members. Each paper was reviewed by at least three referees. SISAP 2019 received 42 papers and accepted 12 as full papers (28% acceptance rate). The program was completed with three keynote speakers of high calibre and one panel.
The first keynote speaker was Fabrizio Silvestri, a Software Engineer at Facebook London working in the Search Systems team. The Facebook AI team in London deals with applying artificial intelligence techniques to address societal problems such as the spread of online misinformation, or the integrity of election processes around the world. To do so, the team has developed a set of tools that exploit similarity search technologies to efficiently and effectively run a very high number of classification tasks on a massive set of data. Fabrizio Silvestri’s talk reviewed some of the problems studied and the solutions adopted.
The second keynote speaker was Alexander Tuzhilin, the Leonard N. Stern Professor of Business in the Department of Technology, Operations and Statistics at the Stern School of Business, NYU. Alex Tuzhilin discussed the role of similarity measures in recommender systems. Measures of similarity between users and between items to be recommended to the users lie at the core of many recommendation algorithms, and numerous metrics have been proposed in the recommender systems field since its inception. The talk explored the evolution of various similarity-based measures from the initial class of rating-based measures to the more recently proposed latent metrics and the metric learning methods. It also explored possible future research directions and novel applications of similarity measures in recommender systems.
The third keynote speaker was Dr. Cong Yu, a research scientist and manager at Google Research in New York City. Cong Yu leads the Structured Data Research Group. The group’s mission is to understand and leverage structured data on the Web to enhance user experience for Google products and has been responsible for several impactful products such as WebTables, Structured Snippets, and Fact-Checking at Google. Currently, his group focuses on technical research for news and has been partnering with journalists and policy advisors to combat online misinformation and improve news consumption. The ClaimReview structured data (http://schema.org/ClaimReview) is a successful example of such collaborations and powers various fact check features for Google. This talk described the genesis of ClaimReview and its role in combating online misinformation.
The SISAP 2019 panel was on Deep Learning meets Similarity Search. The panel was moderated by K. Selçuk Candan (Arizona State University, USA). The panellists were James Bailey (University of Melbourne, Australia), Ilaria Bartolini (University of Bologna, Italy), Michael Houle (National Institute of Informatics, Japan) and Stéphane Marchand-Maillet (University of Geneva, Switzerland).
As it is usually the case, SISAP 2019 included a program with papers exploring various similarity-aware data analysis and processing problems from multiple perspectives. The papers presented at the conference in 2019 studied the role of similarity processing in the context of metric search, visual search, nearest neighbour queries, clustering, outlier detection, and graph analysis. Some of the papers had a theoretical emphasis, while others had a systems perspective, presenting experimental evaluations comparing against state-of-the-art methods. An interesting event at the 2019 conference, as well as the two previous editions, was an electronic poster session that included all accepted papers. This component of the conference generated many lively interactions between presenters and attendees, to not only learn more about the presented techniques but also to identify potential topics for future collaboration.
In a tradition that began with the 2009 conference in Prague, extended versions of the top-ranked papers were invited for a Special Issue of the Information Systems journal. A shortlist for the best papers was created from those conference papers nominated by at least one of their 3 reviewers. An award committee of 3 researchers ranked the shortlisted papers, from which a final ranking was decided. The Best Paper Award was presented to Martin Aumüller and Matteo Ceccarello (IT University of Copenhagen, Copenhagen, Denmark) for the paper titled “The Role of Local Intrinsic Dimensionality in Benchmarking Nearest Neighbor Search” during the Conference Dinner. The best paper reconsiders common benchmarking approaches to nearest neighbour search and studies the effect of different local intrinsic dimensionality (LID) distributions on the running time performance of different implementations.
In addition to the excellent conference facilities at NJIT, we had several student volunteers who were ready to help ensure that the logistical aspects of the conference ran smoothly. Our conference banquet was held at the Newark Museum (https://www.newarkmuseum.org), the largest museum of the state of New Jersey. It holds major collections of American art, decorative arts, contemporary art, and arts of Asia, Africa, the Americas, and the ancient world. The participants were given a highlight tour of the museum prior to the banquet held in the Ballantine House. The Ballantine House is part of The Newark Museum since 1937, the house was designed a National Historic Landmark in 1985. Built in 1885 for Jeannette and John Holme Ballantine, of the celebrated Newark beer-brewing family, this brick and limestone mansion originally had 27 rooms, including eight bedrooms and three bathrooms.
SISAP 2019 demonstrated that the SISAP community has a strong stable kernel of researchers, active in the field of similarity search and to fostering the growth of the community. Organizing SISAP is a smooth experience thanks to the support of the Steering Committee and dedicated participants.
The SISAP 2019 Doctoral Symposium provided a forum for PhD students to present their research ideas and receive feedback from senior members of the research community. The Symposium fostered a collaborative environment with constructive discussions that benefited the students.
SISAP 2020 was supposed to be organized in Copenhagen by Martin Aumüller, Björn Þór Jónsson and Rasmus Pagh from the IT University of Copenhagen. But it will become a virtual event because of the COVID-19 pandemic. One of the major challenges of the SISAP conference series is to continue to raise its profile in the landscape of scientific events related to information indexing, database and search systems.