Statistical Testing Framework for Clustering Pipelines by Selective Inference

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Miyata, Yugo, Shiraishi, Tomohiro, Nishino, Shuichi, Takeuchi, Ichiro
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915971469410304
author Miyata, Yugo
Shiraishi, Tomohiro
Nishino, Shuichi
Takeuchi, Ichiro
author_facet Miyata, Yugo
Shiraishi, Tomohiro
Nishino, Shuichi
Takeuchi, Ichiro
contents A data analysis pipeline is a structured sequence of steps that transforms raw data into meaningful insights by integrating multiple analysis algorithms. In many practical applications, analytical findings are obtained only after data pass through several data-dependent procedures within such pipelines. In this study, we address the problem of quantifying the statistical reliability of results produced by data analysis pipelines. As a proof of concept, we focus on clustering pipelines that identify cluster structures from complex and heterogeneous data through procedures such as outlier detection, feature selection, and clustering. We propose a novel statistical testing framework to assess the significance of clustering results obtained through these pipelines. Our framework, based on selective inference, enables the systematic construction of valid statistical tests for clustering pipelines composed of predefined components. We prove that the proposed test controls the type I error rate at any nominal level and demonstrate its validity and effectiveness through experiments on synthetic and real datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2603_18413
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Statistical Testing Framework for Clustering Pipelines by Selective Inference
Miyata, Yugo
Shiraishi, Tomohiro
Nishino, Shuichi
Takeuchi, Ichiro
Machine Learning
A data analysis pipeline is a structured sequence of steps that transforms raw data into meaningful insights by integrating multiple analysis algorithms. In many practical applications, analytical findings are obtained only after data pass through several data-dependent procedures within such pipelines. In this study, we address the problem of quantifying the statistical reliability of results produced by data analysis pipelines. As a proof of concept, we focus on clustering pipelines that identify cluster structures from complex and heterogeneous data through procedures such as outlier detection, feature selection, and clustering. We propose a novel statistical testing framework to assess the significance of clustering results obtained through these pipelines. Our framework, based on selective inference, enables the systematic construction of valid statistical tests for clustering pipelines composed of predefined components. We prove that the proposed test controls the type I error rate at any nominal level and demonstrate its validity and effectiveness through experiments on synthetic and real datasets.
title Statistical Testing Framework for Clustering Pipelines by Selective Inference
topic Machine Learning
url https://arxiv.org/abs/2603.18413