https://github.com/memgonzales/meta-learning-clustering
Presented at the 2022 IEEE Region 10 Conference (TENCON 2022). Our main contribution is twofold: (1) the construction of a meta-learning model for recommending a distance metric for k-means clustering and (2) a fine-grained analysis of the importance and effects of the meta-features on the model's output
https://github.com/memgonzales/meta-learning-clustering
clustering distance-metric k-means k-means-clustering machine-learning meta-features meta-learning random-forest
Last synced: 3 months ago
JSON representation
Presented at the 2022 IEEE Region 10 Conference (TENCON 2022). Our main contribution is twofold: (1) the construction of a meta-learning model for recommending a distance metric for k-means clustering and (2) a fine-grained analysis of the importance and effects of the meta-features on the model's output
- Host: GitHub
- URL: https://github.com/memgonzales/meta-learning-clustering
- Owner: memgonzales
- License: mit
- Created: 2022-07-05T10:14:01.000Z (almost 3 years ago)
- Default Branch: master
- Last Pushed: 2024-05-05T09:59:21.000Z (about 1 year ago)
- Last Synced: 2025-01-12T17:23:38.351Z (5 months ago)
- Topics: clustering, distance-metric, k-means, k-means-clustering, machine-learning, meta-features, meta-learning, random-forest
- Language: Jupyter Notebook
- Homepage: https://doi.org/10.1109/TENCON55691.2022.9978037
- Size: 97.9 MB
- Stars: 3
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# Distance Metric Recommendation for $k$-Means Clustering: A Meta-Learning Approach
![badge][badge-jupyter]
![badge][badge-python]
![badge][badge-pandas]
![badge][badge-numpy]
![badge][badge-scipy]
**This work was accepted for paper presentation at the 2022 IEEE Region 10 Conference ([TENCON 2022](https://www.ieeer10.org/wp-content/uploads/2021/03/2022-TENCON-Hong-Kong-Section-updated2.pdf)), held virtually and in-person in Hong Kong:**
- The final version of our paper (as published in the conference proceedings of TENCON 2022) can be accessed via this [link](https://ieeexplore.ieee.org/abstract/document/9978037).
- Our preprint can be accessed via this [link](https://github.com/memgonzales/meta-learning-clustering/blob/master/Distance%20Metric%20Recommendation%20for%20k-Means%20Clustering%20A%20Meta-Learning%20Approach.pdf).
- Our TENCON 2022 presentation slides can be accessed via this [link](https://github.com/memgonzales/meta-learning-clustering/blob/master/Presentation%20Slides.pdf).
- Our [dataset of datasets](https://github.com/memgonzales/meta-learning-clustering/tree/master/dataset_of_datasets) is publicly released for future researchers.
- Kindly refer to [`0. Directory.ipynb`](https://github.com/memgonzales/meta-learning-clustering/blob/master/0.%20Directory.ipynb) for a guide on navigating through this repository.
If you find our work useful, please consider citing:
```
@INPROCEEDINGS{9978037,
author={Gonzales, Mark Edward M. and Uy, Lorene C. and Sy, Jacob Adrianne L. and Cordel, Macario O.},
booktitle={TENCON 2022 - 2022 IEEE Region 10 Conference (TENCON)},
title={Distance Metric Recommendation for k-Means Clustering: A Meta-Learning Approach},
year={2022},
pages={1-6},
doi={10.1109/TENCON55691.2022.9978037}}
```
This repository is also archived on [Zenodo](https://doi.org/10.5281/zenodo.7880146).
## Description
**ABSTRACT:** The choice of distance metric impacts the clustering quality of centroid-based algorithms, such as $k$-means. Theoretical attempts to select the optimal metric entail deep domain knowledge, while experimental approaches are resource-intensive. This paper presents a meta-learning approach to automatically recommend a distance metric for $k$-means clustering that optimizes the Davies-Bouldin score. Three distance measures were considered: Chebyshev, Euclidean, and Manhattan. General, statistical, information-theoretic, structural, and complexity meta-features were extracted, and random forest was used to construct the meta-learning model; borderline SMOTE was applied to address class imbalance. The model registered an accuracy of 70.59%. Employing Shapley additive explanations, it was found that the mean of the sparsity of the attributes has the highest meta-feature importance. Feeding only the top 25 most important meta-features increased the accuracy to 71.57%. The main contribution of this paper is twofold: the construction of a meta-learning model for distance metric recommendation and a fine-grained analysis of the importance and effects of the meta-features on the model’s output.
**INDEX TERMS:** meta-learning, meta-features, $k$-means, clustering, distance metric, random forest
## Authors
- Mark Edward M. Gonzales
[email protected]
- Lorene C. Uy
[email protected]
- Jacob Adrianne L. Sy
[email protected]
- Dr. Macario O. Cordel, II
[email protected]
This is the major course output in a machine learning class for master's students under Dr. Macario O. Cordel, II of the Department of Computer Technology, De La Salle University. The task is to create a ten-week investigatory project that applies machine learning to a particular research area or offers a substantial theoretical or algorithmic contribution to existing machine learning techniques.
[badge-jupyter]: https://img.shields.io/badge/Jupyter-F37626.svg?&style=flat&logo=Jupyter&logoColor=white
[badge-python]: https://img.shields.io/badge/python-3670A0?style=flat&logo=python&logoColor=white
[badge-pandas]: https://img.shields.io/badge/Pandas-2C2D72?style=flat&logo=pandas&logoColor=white
[badge-numpy]: https://img.shields.io/badge/Numpy-777BB4?style=flat&logo=numpy&logoColor=white
[badge-scipy]: https://img.shields.io/badge/SciPy-654FF0?style=flat&logo=SciPy&logoColor=white