Description
Book SynopsisThis book constitutes the refereed proceedings of the 16th International Conference on Similarity Search and Applications, SISAP 2023, held in A Coruña, Spain, during October 9–11, 2023.
The 16 full papers and 4 short papers included in this book were carefully reviewed and selected from 33 submissions. They were organized in topical sections as follows: similarity queries, similarity measures, indexing and retrieval, data management, feature extraction, intrinsic dimensionality, efficient algorithms, similarity in machine learning and data mining.
Table of ContentsKeynotes.- From Intrinsic Dimensionality to Chaos and Control: Towards a Unified Theoretical View.- The Rise of HNSW: Understanding Key Factors Driving the Adoption.- Towards a Universal Similarity Function: the Information Contrast Model and its Application as Evaluation Metric in Artificial Intelligence Tasks.-
Research Track.- Finding HSP Neighbors via an Exact, Hierarchical Approach.- Approximate Similarity Search for Time Series Data Enhanced by Section Min-Hash.- Mutual nearest neighbor graph for data analysis: Application to metric space clustering.- An Alternating Optimization Scheme for Binary Sketches for Cosine Similarity Search.- Unbiased Similarity Estimators using Samples.- Retrieve-and-Rank End-to-End Summarization of Biomedical Studies.- Fine-grained Categorization of Mobile Applications through Semantic Similarity Techniques for Apps Classification.- Runs of Side-Sharing Tandems in Rectangular Arrays.- Turbo Scan: Fast Sequential Nearest Neighbor Search in High Dimensions.- Class Representatives Selection in Non-Metric Spaces for Nearest Prototype Classification.- The Dataset-similarity-based Approach to Select Datasets for Evaluation in Similarity Retrieval.- Suitability of Nearest Neighbour Indexes for Multimedia Relevance Feedback.- Accelerating k-Means Clustering with Cover Trees.- Is Quantized ANN Search Cursed? Case Study of Quantifying Search and Index Quality.- Minwise-Independent Permutations with Insertion and Deletion of Features.- SDOclust: Clustering with Sparse Data Observers.- Solving k-Closest Pairs in High-Dimensional Data using Locality- Sensitive Hashing.- Vec2Doc: Transforming Dense Vectors into Sparse Representations for Efficient Information Retrieval.- Similarity Search with Multiple-Object Queries.- Diversity Similarity Join for Big Data.-
Indexing Challenge.- Overview of the SISAP 2023 Indexing Challenge.- Enhancing Approximate Nearest Neighbor Search: Binary-Indexed LSH-Tries, Trie Rebuilding, And Batch Extraction.- General and Practical Tuning Method for Off-the-Shelf Graph-Based Index: SISAP Indexing Challenge Report by Team UTokyo.- SISAP 2023 Indexing Challenge – Learned Metric Index.- Computational Enhancements of HNSW Targeted to Very Large Datasets.- CRANBERRY: Memory-Effective Search in 100M High-Dimensional CLIP Vectors.
