# The Minimum Description Length Principle for Pattern Mining: A Survey

arXiv: https://arxiv.org/abs/2007.14009 · HTML: https://mlsanity.com/paper/2007.14009/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Jul 2020.

## Built on by (most cited first, 13 of 13)

- [Robust subgroup discovery](https://mlsanity.com/paper/2103.13686/) (2021)
- [Conditional Density Estimation with Histogram Trees](https://mlsanity.com/paper/2410.11449/) (2024)
- [SkillZip: Evaluation-Free Skill Compression for Self-Evolving Agents by Discovering Reusable Structure](https://mlsanity.com/paper/2608.11079/) (2026)
- [MDL-based Compressing Sequential Rules](https://mlsanity.com/paper/2212.10252/) (2022)
- [SWoTTeD: An Extension of Tensor Decomposition to Temporal Phenotyping](https://mlsanity.com/paper/2310.01201/) (2023)
- [Probabilistic Truly Unordered Rule Sets](https://mlsanity.com/paper/2401.09918/) (2024)
- [Online Detection of Anomalies in Temporal Knowledge Graphs with Interpretability](https://mlsanity.com/paper/2408.00872/) (2024)
- [Role-based lateral movement detection with unsupervised learning](https://mlsanity.com/paper/2108.02713/) (2021)
- [KG-MDL: Mining Graph Patterns in Knowledge Graphs with the MDL Principle](https://mlsanity.com/paper/2309.12908/) (2023)
- [Data is Moody: Discovering Data Modification Rules from Process Event Logs](https://mlsanity.com/paper/2312.14571/) (2023)
- [Seqret: Mining Rule Sets from Event Sequences](https://mlsanity.com/paper/2505.06049/) (2025)
- [FedSLIM: Privacy-Preserving Federated MDL-Based Descriptive Pattern Mining Across Data Silos](https://mlsanity.com/paper/2607.23236/) (2026)
- [SkillZip Pro: Execution-Aware Dynamic Compression of Progressively Loaded Skills for Self-Evolving Agents](https://mlsanity.com/paper/2608.30785/) (2026)

## Abstract (arXiv)

This is about the Minimum Description Length (MDL) principle applied to pattern mining. The length of this description is kept to the minimum. Mining patterns is a core task in data analysis and, beyond issues of efficient enumeration, the selection of patterns constitutes a major challenge. The MDL principle, a model selection method grounded in information theory, has been applied to pattern mining with the aim to obtain compact high-quality sets of patterns. After giving an outline of relevant concepts from information theory and coding, as well as of work on the theory behind the MDL and similar principles, we review MDL-based methods for mining various types of data and patterns. Finally, we open a discussion on some issues regarding these methods, and highlight currently active related data analysis problems.