Abstract:It has been a long time, since data mining technologies have made their ways to the field of data management. Classification is one of the most important data mining tasks for label prediction, categorization of objects into groups, advertisement and data management. In this paper, we focus on the standard classification problem which is predicting unknown labels in Euclidean space. Most efforts in Machine Learning communities are devoted to methods that use probabilistic algorithms which are heavy on Calculus and Linear Algebra. Most of these techniques have scalability issues for big data, and are hardly parallelizable if they are to maintain their high accuracies in their standard form. Sampling is a new direction for improving scalability, using many small parallel classifiers. In this paper, rather than conventional sampling methods, we focus on a discrete classification algorithm with O(n) expected running time. Our approach performs a similar task as sampling methods. However, we use column-wise sampling of data, rather than the row-wise sampling used in the literature. In either case, our algorithm is completely deterministic. Our algorithm, proposes a way of combining 2D convex hulls in order to achieve high classification accuracy as well as scalability in the same time. First, we thoroughly describe and prove our O(n) algorithm for finding the convex hull of a point set in 2D. Then, we show with experiments our classifier model built based on this idea is very competitive compared with existing sophisticated classification algorithms included in commercial statistical applications such as MATLAB.

Fast convex hull classifier based on random projection

Generalized Convexity-Based Inexact Projection Method for Multiple Kernel Learning

Convex hull properties and algorithms

Distance-Based Classifier Via the Kernel Trick.

A Fast Convex Hull Algorithm for Binary Image.

DataGrinder: Fast, Accurate, Fully non-Parametric Classification Approach Using 2D Convex Hulls

Fast learning algorithm for support vector domain description

Fast incremental learning algorithm of linear support vector machine based on hull vectors

GPU Accelerated Convex Hull Computation.

A Fast Convex Hull Algorithm for Ring-Distributed Planar Point Sets

Fast Fisher discriminant analysis with randomized algorithms.

Fast Fusion Clustering via Double Random Projection

Adaptive Quasiconformal Kernel Fisher Discriminant Analysis Via Weighted Maximum Margin Criterion

Matching Image Sets via Adaptive Multi Convex Hull

An Evaluation of GPU Filters for Accelerating the 2D Convex Hull

CudaChain: A Practical GPU-accelerated 2D Convex Hull Algorithm

Algorithmic Efficiency in Convex Hull Computation: Insights from 2D and 3D Implementations

Probabilistic Convex Hull Queries over Uncertain Data

Convex-hull Feature Adaptation for Oriented and Densely Packed Object Detection

Sparse Random Projection for Χ2 Kernel Linearization: Algorithm and Applications to Image Classification

Robust Grassmann manifold convex hull collaborative representation learning and its kernel extension for image set analysis