Open research questions in Advanced Clustering Algorithms Research

What the literature leaves open

• There is a need for quantitative morphological analysis of the active phase of sulfide catalysts from HRTEM images. Current methods are laborious and require manual annotation. There is a lack of publicly available labeled datasets for HRTEM images of sulfide catalysts.

  A Geometric Clustering Method of Lattice Fringes for Quantitative Assessment of the Active Phase Morphology of Sulfide Catalysts from HRTEM Images · 2026 · DOI

• The K-Means algorithm has limitations such as random initialization of centroids, which can lead to suboptimal clustering. The basic Hartigan–Wong algorithm is computationally expensive and may not always produce the optimal clustering. There is a need for improved initialization methods and the use of genetic concepts to improve clustering accuracy.

  Clustering Performance of a Recombinator Hartigan–Wong Algorithm · 2026 · DOI

• Despite their widespread use, limited research has explored how index performance varies across problem types, with traditional benchmarks focusing on ground-truth properties that cannot be known prior to clustering.

  Instance space of clustering validation measures · 2026 · DOI

• Future research will focus on exploring feature selection and reduction techniques to adapt the proposed algorithm for effective application to massive datasets characterized by an exceptionally high number of features.

  A novel fuzzy-entropy based online fuzzy C-Means clustering algorithm for massive data · 2025 · DOI

• Temporal data, obtained in the setting where it is only possible to observe one time point per experiment, is widely used in different research fields, yet remains insufficiently addressed from the statistical point of view.

  Joint Clustering With Alignment for Temporal Data in a One-Point-per-Experiment Setting · 2025 · DOI

• Moreover, since RVIs are computed using object dissimilarities, it remains unclear how they would need to be implemented for fair comparisons of different SPs.

  On the Use of Relative Validity Indices for Comparing Clustering Approaches · 2025 · DOI

• The model is more complex than other models, such as DLA. The process is not well defined for α < -2. The authors need to develop new mathematical techniques to study the behavior of the model.

  On One-Dimensional Cluster-cluster Model · 2026 · DOI

• The lack of a rigorous mathematical treatment of the Cluster-cluster model. The need for a more detailed analysis of the behavior of the model for different values of α.

  On One-Dimensional Cluster-cluster Model · 2026 · DOI

• The lack of efficient hardware acceleration for K-means algorithm. The need for a scalable and efficient system for large-scale data processing. The limitation of traditional CPU-based systems.

  FPGA-Based Hardware Accelerator for K-Means Clustering Algorithm · 2026 · DOI

• The proposed modification of the firefly algorithm proves particularly useful in scenarios where initialization sensitivity must be avoided, the number of clusters needs to be determined automatically, and clustering objectives extend beyond simple distance-based criteria. For future work, several directions can be explored to enhance the proposed approach.

  When Fireflies Cluster; Enhancing Automatic Clustering via Centroid-Guided Firefly Optimization · 2026

• The exponential growth of data over the past few years have posed significant challenges on its evaluation. The machine learning (ML) techniques working well in low-dimensional contexts are unable to perform simpler tasks on high-dimensional datasets. Thus, the efficiency of the existing algorithms is minimised and computational costs have increased.

  A Taxonomy of DBSCAN Variants Addressing Parameter Dependence and High-Dimensionality · 2026 · DOI

• Prior evaluation methods are insufficient for GSC. There is a need for a new methodology to evaluate explanations in GSC.

  An Evaluation Methodology for Explanations of Clustering Results in Textual Domain · 2026 · DOI

• The study highlights the need for a comparison of different clustering algorithms for this specific application.

  Application of Clustering Techniques for the Analysis and Classification of Firms in the Automotive Industry · 2026 · DOI

• Future research may focus on developing scalable and resource-aware clustering frameworks capable of operating in large-scale and dynamic environments, including emerging application domains such as the Internet of Nano Things (IoNT).

  A comprehensive survey on clustering algorithms: concepts, taxonomy with nature-inspired meta-heuristic approaches and performance metrics · 2026 · DOI

• The paper also identifies a need for a canonical-form algorithm that can compute a unique representative per orbit.

  Symmetry-Reduced Enumeration and Canonical Forms for Grid-Based Density Clustering Under the Hyperoctahedral Group · 2026 · DOI

• Our experiments on 185 publicly available mixed datasets show that the pairs of similarity measures recommended by SIMREC outperform the baseline pairs, including classically used pairs of similarity measures in the literature.

  Simrec: a similarity measure recommendation system for mixed data clustering algorithms · 2025 · DOI

• They primarily achieve data imputation through the generalization ability of deep models, but their model interpretability may be limited by the black-box nature.

  SOI-Net: Structural Optimization-Inspired Interpretable Network for Incomplete Multi-View Clustering · 2025 · DOI

• Despite its flexibility, a data driven embedding model may not be applicable when the raw data are unavailable due to privacy or security concerns, and it cannot be adapted to the increasingly common framework of transfer learning.

  Exploring the Relationship Between the Geometry of a Fixed Embedding of Image Data and Its Underlying Cluster Structure · 2024 · DOI

• Investigate the style matrix in IS-TSK-FC with deep learning models to more efficiently learn style feature representations of data.

  Interpretable style Takagi-Sugeno-Kang fuzzy clustering · 2026 · DOI

• Limited attention has been given to the interpretability of the clusters detected by most clustering algorithms. Most clustering algorithms focus on efficient cluster assignment, but few works have focused on cluster interpretability.

  Interpretable style Takagi-Sugeno-Kang fuzzy clustering · 2026 · DOI

• The study is motivated by the observation that the standard FCM model uses an equal metric contribution of all coordinates and therefore does not account for their different structural significance in forming cluster geometry.

  Conservative entropy-regularized feature-weighted fuzzy C-means clustering method · 2026 · DOI

• The ultimate goal of our study was to develop a system that combines static K-means and HAC classifiers. Using the strengths of multiple static classifiers, we aimed to improve overall performance and achieve a higher recognition rate.

  KMHC: A Parallel Hybrid Clustering Algorithm Integrating K-Means and Hierarchical Agglomerative Clustering with Voting-Based Consensus · 2026 · DOI

• We implemented the proposed system using 130 pairs of similarity measures (10 numerical and 13 categorical), 4 commonly used mixed data clustering algorithms (K-Prototypes, LSH-K-Prototypes, K-Medoids, and Hierarchical Clustering), and three cluster validity indices (Silhouette, Clustering Accuracy, and Adjusted Rand Index).

  Simrec: a similarity measure recommendation system for mixed data clustering algorithms · 2025 · DOI

• This work establishes RCI as a unified geometric clustering framework operating across four complementary structural layers: (i) Minimal-metric structural: generation of the radial base, restriction structure, admissible sieves, and certified sections from points and distances alone (Appendix 15); (ii) Local analytic: metric finite-difference curvature expansions, together with bi-Lipschitz spectral fidelity…

  Reverse Clustering Impact: Geometry-Driven Minimal Parameters for Clustering via Morse Theory · 2026 · DOI

• Applying the results to other clustering methods - Investigating the use of the hyperoctahedral group in other machine learning contexts - Extending the results to more general grid…

  Symmetry-Reduced Enumeration and Canonical Forms for Grid-Based Density Clustering Under the Hyperoctahedral Group · 2026 · DOI

Most-cited papers in Advanced Clustering Algorithms Research

• Clustering by Passing Messages Between Data Points · Science · 2007 · 5,531 citations
• Multidimensional Scaling by Optimizing Goodness of Fit to a Nonmetric Hypothesis · Psychometrika · 1964 · 5,470 citations
• Clustering by fast search and find of density peaks · Science · 2014 · 4,619 citations
• Hierarchical Clustering Schemes · Psychometrika · 1967 · 3,713 citations
• An Examination of Procedures for Determining the Number of Clusters in a Data Set · Psychometrika · 1985 · 2,770 citations
• Generalized Procrustes Analysis · Psychometrika · 1975 · 2,751 citations
• Chameleon: hierarchical clustering using dynamic modeling · Computer · 1999 · 1,538 citations
• A comprehensive survey of clustering algorithms: State-of-the-art machine learning applications, taxonomy, challenges, and future research prospects · Engineering Applications of Artificial Intelligence · 2022 · 877 citations
• The Analysis of Proximities: Multidimensional Scaling with an Unknown Distance Function. II · Psychometrika · 1962 · 872 citations
• An Examination of the Effect of Six Types of Error Perturbation on Fifteen Clustering Algorithms · Psychometrika · 1980 · 866 citations

Most recent work

• Contrastive prompt clustering for weakly supervised semantic segmentation · Expert Systems with Applications · 2026
• Tensorized Fine-Grained Incomplete Multiview Clustering via Intrinsic Structure Recovery · IEEE Transactions on Systems Man and Cybernetics Systems · 2026
• Tensorized Parameter-Free Multi-View Spectral Clustering Based on Fair Representation Learning · IEEE Transactions on Multimedia · 2026
• A Survey of Deep Graph Clustering: Taxonomy, Challenge, Application, and Open Resource · IEEE Transactions on Knowledge and Data Engineering · 2026
• One-Step Multi-View Clustering With Adaptive Low-Rank Anchor-Graph Learning · IEEE Transactions on Multimedia · 2026
• An upper bound on the silhouette evaluation metric for clustering · Pattern Recognition · 2026
• A rapid review of clustering algorithms · Array · 2026
• Deep Multi-View Clustering With Intra-View Similarity and Cross-View Correlation Learning · IEEE Transactions on Multimedia · 2026
• One-pass multiview clustering with anchor differentiation mechanism · Pattern Recognition · 2026
• Interpretable Subspace Clustering · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2026

Related topics in Computer Science