Summary of Towards Explainable Clustering: a Constrained Declarative Based Approach, by Mathieu Guilbert et al.
Towards Explainable Clustering: A Constrained Declarative based Approach
by Mathieu Guilbert, Christel Vrain, Thi-Bich-Hanh Dao
First submitted to arxiv on: 26 Mar 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The novel ECS framework for declarative clustering with Explainability-driven Cluster Selection integrates structural or domain expert knowledge expressed by means of constraints. It’s based on the notions of coverage and discrimination, which are formalized at different levels (cluster/clustering), each allowing for exceptions through parameterized thresholds. The method generates a set of partitions, computes frequent patterns for each cluster, prunes clusters that violate some constraints, and selects clusters and associated patterns to build an interpretable clustering. This last step is combinatorial and relies on a Constraint-Programming (CP) model to solve it. ECS can integrate prior knowledge in the form of user constraints, both before or within the CP model. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper aims to find a clustering that has high quality and is explainable. They want to create a good global explanation of a clustering that shows what each cluster is like and how it’s different from others. The team also wants to use expert knowledge to make the clustering better. They propose a new method called ECS (Explainability-driven Cluster Selection) that uses patterns to explain each cluster. The method has four steps: generate clusters, find patterns in each one, remove clusters that don’t follow rules, and pick the best clusters and patterns to show. |
Keywords
* Artificial intelligence * Clustering
