Motivated by the need for distributed learning and optimization algorithms with low communication cost, we study communication efficient algorithms for distributed mean estimation. Unlike previous works, we make no probabilistic assumptions on the data. We first show that for d dimensional data with n clients, a naive stochastic rounding approach yields a mean squared error (MSE) of ⇥(d/n) and ...

Query redirection provides a mechanism for BI Server to determine the set of logical table sources (LTS) applicable to a logical request whenever a request can be satisfied by more than one LTS.The Oracle BI repository shipped in Oracle Fusion applications contains metadata content for real-time reporting analysis (using Transactional Business Intelligence) and historical reporting (using BI Ap...

Image segmentation plays a significant role in computer vision. It aims at extracting meaningful objects lying in the image. Generally there is no unique method or approach for image segmentation. Clustering is a powerful technique that has been reached in image segmentation. The cluster analysis is to partition an image data set into a number of disjoint groups or clusters. The clustering meth...

In this work we introduce an anisotropic density-based clustering algorithm. It outperforms DBSCAN and OPTICS for the detection of anisotropic spatial point patterns and performs equally well in cases that do not explicitly benefit from an anisotropic perspective. ADCN has the same time complexity as DBSCAN and OPTICS, namely O(n log n) when using a spatial index, O(n2) otherwise. STKO@Geograph...

Abstract With the development of urbanization, problem urban traffic congestion is becoming more and serious. An improved k-means clustering algorithm was proposed to solve that traditional center could easily be affected by fall into local optimal solution. Based on big data New York City taxis, operational characteristics are analyzed. The experimental results show K-means has a better analys...

Data objects with mixed numeric and categorical attributes are commonly encountered in real world. The k-prototypes algorithm is one of the principal algorithms for clustering this type of data objects. In this paper, we propose an improved k-prototypes algorithm to cluster mixed data. In our method, we first introduce the concept of the distribution centroid for representing the prototype of c...

This paper presents an optimal and unsupervised satellite image segmentation approach based on Pearson system and k-Means Clustering Algorithm Initialization. Such method could be considered as original by the fact that it utilised K-Means clustering algorithm for an optimal initialisation of image class number on one hand and it exploited Pearson system for an optimal statistical distributions...

This paper reflects the results of an implementation of the K-means algorithm on U.N survey data on people’s priorities, organized by country. The dataset includes 16 features for each country, with each feature corresponding to a different societal issue. Each country has a rating in the range of [0, 1] that indicates how important a particular feature or issue is to that country’s people– the...

k-means++ is a seeding technique for the k-means method with an expected approximation ratio of O(log k), where k denotes the number of clusters. Examples are known on which the expected approximation ratio of k-means++ is Ω(log k), showing that the upper bound is asymptotically tight. However, it remained open whether k-means++ yields an O(1)-approximation with probability 1/poly(k) or even wi...

infiltration plays an important role in surface and subsurface hydrology and it is a key factor in the rainfall and runoff equations. the use of new approaches that have no limitations of common theoretical and empirical methods to determine infiltration relationships, will minimize the necessity of time consuming and costly experiments to determine permeability values and will make it possible...