SEAnet: A Deep Learning Architecture for Data Series Similarity Search

2026-03-01

AI Summary

This work proposes the Deep Embedding Approximation (DEA) framework to address the limited performance of existing SAX-based indexing methods in similarity search over high-frequency, weakly correlated, or highly noisy time series. DEA is the first approach to integrate deep neural networks into symbolic aggregate approximation, introducing the SEAnet architecture with sum-of-squares preservation properties. It further incorporates the SEAtrans encoder and novel sampling strategies—SEAsam and SEAsamE—to enable efficient training and high-precision retrieval. Extensive experiments on seven synthetic and real-world datasets demonstrate that DEA significantly outperforms state-of-the-art methods, achieving notable advances in both time series representation quality and similarity search accuracy.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Time-Series/Data Streams
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

A key operation for massive data series collection analysis is similarity search. According to recent studies, SAX-based indexes offer state-of-the-art performance for similarity search tasks. However, their performance lags under high-frequency, weakly correlated, excessively noisy, or other dataset-specific properties. In this work, we propose Deep Embedding Approximation (DEA), a novel family of data series summarization techniques based on deep neural networks. Moreover, we describe SEAnet, a novel architecture especially designed for learning DEA, that introduces the Sum of Squares preservation property into the deep network design. We further enhance SEAnet with SEAtrans encoder. Finally, we propose novel sampling strategies, SEAsam and SEAsamE, that allow SEAnet to effectively train on massive datasets. Comprehensive experiments on 7 diverse synthetic and real datasets verify the advantages of DEA learned using SEAnet in providing high-quality data series summarizations and similarity search results.

Problem

• similarity search
• data series
• SAX-based indexes
• noisy data
• high-frequency data

Innovation

• Deep Embedding Approximation
• SEAnet
• Sum of Squares preservation
• SEAtrans
• Similarity Search