The Minimum Description Length Principle for Pattern Mining: A Survey
arXiv:2007.14009 · doi:10.1007/s10618-022-00846-z
Abstract
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.
arXiv commentData Min Knowl Disc (2022)
References in corpus (31)
- Maps of random walks on complex networks reveal community structure
- The information bottleneck method
- The map equation
- An information-theoretic framework for resolving community structure in complex networks
- Mapping change in large networks
- Multilevel compression of random walks on networks reveals hierarchical organization in large integrated systems
- Identifying modular flows on multilayer networks reveals highly overlapping organization in social systems
- Infomap Bioregions: Interactive mapping of biogeographical regions from species distributions
- Mapping higher-order network flows in memory and multilayer networks with Infomap
- A Philosophical Treatise of Universal Induction
- The Long and the Short of It: Summarising Event Sequences with Serial Episodes
- Exploring the solution landscape enables more reliable network community detection
- Tell Me Something I Don't Know: Randomization Strategies for Iterative Data Mining
- A Tutorial on Statistically Sound Pattern Discovery
- Interpretable multiclass classification by MDL-based rule lists
- Compression of Flow Can Reveal Overlapping-Module Organization in Networks
- Summarizing Data Succinctly with the Most Informative Itemsets
- What is Normal, What is Strange, and What is Missing in a Knowledge Graph: Unified Characterization via Inductive Summarization
- SSumM: Sparse Summarization of Massive Graphs
- A Subsequence Interleaving Model for Sequential Pattern Mining
- Finding Good Itemsets by Packing Data
- A Map Equation with Metadata: Varying the Role of Attributes in Community Detection
- Large-scale network motif analysis using compression
- Robust subgroup discovery
- Decomposable Families of Itemsets
- Probably the Best Itemsets
- Graph Similarity Description: How Are These Graphs Similar?
- Discovering outstanding subgroup lists for numeric targets using MDL
- Discovering Descriptive Tile Trees by Mining Optimal Geometric Subtiles
- A tutorial on MDL hypothesis testing for graph analysis
- Mining Periodic Patterns with a MDL Criterion