• the Data AnalyTics, Understanding,
    Mining, and Management Lab

    EXPLORE THE LAB

DATUM

Welcome to the Data AnalyTics, Understanding, Mining, and Management (DATUM) Lab at The Ohio State University, led by Professor John Paparrizos. Our research focuses on advancing the next generation of data-intensive and machine learning applications. We specialize in developing adaptive solutions for managing and analyzing structured and unstructured data, including relational data, time series, multimedia, text, graphs, web data, and IoT data.

At DATUM Lab, we are dedicated to advancing data management, data mining, machine learning, and artificial intelligence. We welcome researchers and students interested in these fields to join us in exploring the frontiers of data science.

Research Interests

Machine Learning

Data Mining

Database

Conferences/Journals

SIGMOD, SIGKDD,

NeurIPS, ICDE,

VLDB, ...

Current Projects

SIGNAL Library

Chameleon

SHIELD

people
Prof. John Paparrizos
Assistant Professor
CSE @ OSU
CSD @ AUTH

News

Awards · Papers · Events

Read more

2026.05

Three papers and one demo accepted at ACM SIGMOD 2026!

  1. HYDRA: A Multi-Level Hierarchy-Driven Approach for Robust Anomaly Detection in Time Series
  2. The Power of Anomaly Detection in Predictive Maintenance: [Experiments & Analysis]
  3. MUFASA: Fast and Accurate Multivariate Time-Series Clustering
  4. GlassboxAD: An Interactive System for Dissecting Hierarchical Time-Series Anomaly Detection

2026.02

One tutorial was presented at ACM WSDM 2026!

  1. A Comprehensive Guide to Time-Series Anomaly Detection

2025.11

Our comprehensive time-series intelligence platform (TimeSeries.org) is comming soon!

2025.08

One tutorial was presented at ACM KDD 2025!

  1. Advances in Time-Series Anomaly Detection: Algorithms, Benchmarks, and Evaluation Measures

2025.08

Four papers and two demos accepted at VLDB 2025!

  1. TSB-AutoAD: Towards Automated Solutions for Time-Series Anomaly Detection
  2. Time-Series Clustering: A Comprehensive Study of Data Mining and Deep Learning Methods
  3. Beyond Compression: A Comprehensive Evaluation of Lossless Floating-Point Compression
  4. BURST: Rendering Clustering Techniques Suitable for Evolving Streams
  5. EasyAD: A Demonstration of Automated Solutions for Time-Series Anomaly Detection
  6. SAIL: A Voyage to Symbolic Approximation Solutions for Time-Series Analysis

2025.07

Two tutorials accepted at ACM SIGKDD 2025 and IJCAI 2025!

  1. Advances in Time-Series Anomaly Detection: Algorithms, Benchmarks, and Evaluation Measures

2025.06

Humbled and honored to receive the 2025 ACM SIGMOD Test-of-Time Award for k-Shape!

2025.05

Three papers and a demo accepted at ACM SIGMOD 2025:

  1. SPARTAN: Data-Adaptive Symbolic Time-Series Approximation
  2. Understanding the Black Box: A Deep Empirical Dive into Shapley Value Approximations
  3. A Structured Study of Multivariate Time-Series Distance Measures
  4. ShapX Engine: A Demonstration of Shapley Value Approximations

2024.10

One paper accepted at NeurIPS 2024:

  1. The Elephant in the Room: Towards A Reliable Time-Series Anomaly Detection Benchmark

SPONSORS