14
Algorithms Available
Loading…
Explore, analyze, and visualize your data with 14 state-of-the-art clustering algorithms
At a glance
Algorithms Available
Dimensional Support
Cluster Detection
Upload Your Data
Getting Started Quick Guide
Upload a CSV or Excel (.xlsx) file with your numerical features. Headers are automatically detected.
Set the number of clusters, preprocessing options, and algorithm-specific parameters.
Explore interactive visualizations, metrics, and detailed cluster analysis.
Choose an algorithm based on your data characteristics and goals
Partitioning (Centroid-Based) 4 algorithms
The most widely-used clustering algorithm. It partitions data into k clusters by iteratively assigning points to the nearest centroid and updating centroids until convergence.
The Classic Partitioning Algorithm · Fast · Scalable · Simple · Well-understood · Outlier: Centroid Distance · Best for: Large datasets with spherical clusters
Try it now
Similar to K-Means but uses actual data points (medoids) as cluster centers instead of means. More robust to outliers and noise in the data.
Robust to Outliers · Robust · Interpretable · Any distance metric · Outlier: Medoid Distance · Best for: Data with outliers or categorical features
Try it now
A non-parametric algorithm that doesn't require specifying the number of clusters. It finds clusters by shifting points towards the mode (highest density) of the data.
No K Required · Automatic k · Mode-seeking · Flexible shapes · Outlier: Mode Distance · Best for: Unknown number of clusters, image segmentation
Try it now
Unlike hard clustering, FCM allows data points to belong to multiple clusters with varying degrees of membership. Perfect for overlapping or ambiguous cluster boundaries.
Soft Clustering · Soft membership · Overlapping clusters · Fuzzifier control · Outlier: Membership Analysis · Best for: Overlapping clusters, uncertainty quantification
Try it now
Density-Based 3 algorithms
Discovers clusters of arbitrary shape based on density. Can identify noise points and doesn't require specifying the number of clusters beforehand.
Density-Based Spatial Clustering · Arbitrary shapes · Noise detection · No k required · Outlier: k-Distance · Best for: Geographic data, anomaly detection
Try it now
Hierarchical extension of DBSCAN that handles varying density clusters. Currently the gold standard for density-based clustering.
The Modern Standard · Variable density · Soft clustering · Robust · Outlier: GLOSH Score · Best for: Complex real-world datasets
Try it now
Creates an ordering of the database that captures density-based clustering structure. Produces reachability plots for visual cluster analysis.
Ordering Points Algorithm · Reachability plot · Variable density · Hierarchical view · Outlier: Reachability Ratio · Best for: Exploratory analysis, varying density clusters
Try it now
Hierarchical 3 algorithms
Builds a hierarchy of clusters by progressively merging the closest pairs. Produces a dendrogram that can be cut at any level.
Bottom-Up Hierarchical · Dendrogram · Flexible linkage · Any k post-hoc · Outlier: Centroid Distance · Best for: Taxonomy creation, gene expression analysis
Try it now
Balanced Iterative Reducing and Clustering using Hierarchies. Designed for very large datasets with limited memory.
Scalable Hierarchical Clustering · Memory efficient · Incremental · Large-scale · Outlier: Subcluster Distance · Best for: Very large datasets, streaming data
Try it now
A divisive (top-down) hierarchical clustering algorithm that repeatedly bisects clusters using K-Means until the desired number of clusters is reached.
Divisive Hierarchical · Top-down · Divisive · Hierarchical structure · Outlier: Inertia-Weighted · Best for: Document clustering, when hierarchy matters
Try it now
Probabilistic & Neural 4 algorithms
Models data as a mixture of Gaussian distributions. Provides soft cluster assignments and can model elliptical clusters of different sizes.
Probabilistic Clustering · Probabilistic · Elliptical clusters · BIC/AIC selection · Outlier: Log-Likelihood · Best for: Statistical modeling, density estimation
Try it now
Uses eigenvalues of a similarity matrix to reduce dimensionality before clustering. Excellent for non-convex cluster shapes.
Graph-Based Approach · Graph theory · Non-convex shapes · Similarity-based · Outlier: Embedding Distance · Best for: Image segmentation, community detection
Try it now
Identifies exemplars among data points by passing messages between pairs. Automatically determines the number of clusters.
Message-Passing Algorithm · Auto k · Exemplar-based · Message passing · Outlier: Exemplar Distance · Best for: Finding representative examples, face clustering
Try it now
A neural network approach that grows nodes to represent data topology. Adapts structure during training to match data distribution.
Self-Organizing Network · Topology learning · Adaptive · Neural network · Outlier: Quantization Distance · Best for: Topology preservation, continuous learning
Try it now
Start with K-Means for a quick baseline, then try Fuzzy C-Means if you suspect overlapping clusters.