Generating Optimum Number of Clusters Using Median Search and Projection Algorithms

2010 ◽  
Author(s):  
Suresh L. ◽  
Jay B. Simha ◽  
Rajappa Veluru
Keyword(s):  
Optimum Number ◽  
Projection Algorithms ◽  
Number Of Clusters

An Optimizing Method of Competitive Neural Network

2011 ◽  
Vol 467-469 ◽  
pp. 894-899
Author(s):  
Hong Men ◽  
Hai Yan Liu ◽  
Lei Wang ◽  
Yun Peng Pan
Keyword(s):  
Neural Network ◽  
Network Structure ◽  
Clustering Analysis ◽  
Learning Rate ◽  
Optimum Number ◽  
Number Of Clusters ◽  
Competitive Neural Network ◽  
Output Neurons ◽  
Simulation Results ◽  
Strong Ability

This paper presents an optimizing method of competitive neural network(CNN):During clustering analysis fixed on the optimum number of output neurons according to the change of DB value，and then adjusted connected weight including increasing ,dividing , delete. Each neuron had the different variety trend of learning rate according with the change of the probability of neurons. The optimizing method made classification more accurate. Simulation results showed that optimized network structure had a strong ability to adjust the number of clusters dynamically and good results of classification.

A new validity clustering index-based on finding new centroid positions using the mean of clustered data to determine the optimum number of clusters

2021 ◽  
pp. 116329
Author(s):  
Ahmed Khaldoon Abdalameer ◽  
Mohammed Alswaitti ◽  
Ahmed Adnan Alsudani ◽  
Nor Ashidi Mat Isa
Keyword(s):  
Clustered Data ◽  
Optimum Number ◽  
Number Of Clusters ◽  
The Mean

Baffin Bay/Nares Strait surface (seafloor) sediment mineralogy: further investigations and methods to elucidate spatial variations in provenance

2019 ◽  
Vol 56 (8) ◽  
pp. 814-828 ◽  
Author(s):  
John T. Andrews
Keyword(s):  
Strong Association ◽  
Optimum Number ◽  
Baffin Island ◽  
Number Of Clusters ◽  
Data Transformations ◽  
West Greenland ◽  
Weight Percentage ◽  
Nares Strait ◽  
Significant Difference ◽  
Sediment Mineralogy

The goal of the paper is to ascertain whether there are significant regional variations in sediment mineral composition that might be used to elucidate ice sheet histories. The weight percentages of nonclay and clay minerals were determined by quantitative X-ray diffraction. Cluster analysis, an unsupervised learning approach, is used to group sediment mineralogy of 263 seafloor/core top samples between ∼80°N and 62°N. The optimum number of clusters, based on 30 indexes, was three for the weight percentage data but varied with data transformations. Maps of the distribution of the three mineral clusters or facies indicate a significant difference in weight percentages between samples from the West Greenland and Baffin Island shelves. However, several indexes support a larger number of clusters and similar analyses of the spatial distribution and defining minerals of nine mineral facies indicated a strong association with the original three clusters and with broad geographic designations (i.e., West Greenland shelf, Baffin Island fiords, etc). Classification Decision Tree analysis indicates that this difference is primarily controlled by the percentages of plagioclase feldspars versus alkali feldspars.

scholarly journals Implementasi Metode Improved K-Means dengan Algoritma Dbscan untuk Pengelompokan Film

2020 ◽  
Vol 6 (01) ◽  
pp. 1-8
Author(s):  
Muhammad Muhajir ◽  
Annisa Ayunda Permata Sari
Keyword(s):  
Film Industry ◽  
Optimal Number ◽  
Optimum Number ◽  
Number Of Clusters ◽  
The Past ◽  
Box Office ◽  
Dbscan Algorithm ◽  
The People ◽  
The World ◽  
Optimal Number Of Clusters

The Indonesian film industry continues to experience an increase seen from the number of films that appear in theaters today with a box office increase of 28 percent each year in the past four years. Internet Movie Database (IMDb) is a website that provides information about films around the world, including the people involved in it from actors, directors, writers to makeup artists and soundtracks. In this case the researcher wants to conduct research on the characteristics of the film and the factors that make a film to be included in the IMDb Top 250. The data used in this study uses scraped data from the website. The method used is a non-hierarchical clustering method, namely kmeans and Dbscan. Where the Dbscan algorithm is used to determine the optimum number of clusters then proceed by grouping data based on centroids with k-means algorithm. From the analysis it was found that the factors that could influence a film included in the IMDB Top 250 were duration, number of votes, and films directed by Rajkumar Hirani and the optimal number of clusters using Dbscan algorithm obtained six clusters. With the improved k-means algorithm, the accuracy value for the cluster results is 87.2%.

scholarly journals Analisis Cluster Data Interkomparasi Anak Timbangan dengan Algoritma Self Organizing Maps

2021 ◽  
Vol 7 (2) ◽  
Author(s):  
Arif Fajar Solikin ◽  
Kusrini Kusrini ◽  
Ferry Wahyu Wibowo
Keyword(s):  
Data Mining ◽  
Statistical Test ◽  
Optimum Number ◽  
Data Normalization ◽  
Self Organizing Map ◽  
Self Organizing Maps ◽  
Number Of Clusters ◽  
Map Algorithm ◽  
Cluster Data ◽  
Self Organizing

Intercomparison was conducted to determine the ability and the performance of the laboratory. Intercomparison results are usually expressed in the range of En ratio values (En ?|1|) which express the equivalence of one laboratory with other laboratories. If the laboratory is declared unequal, then it needs to identify the source of the problem by itself. To make it easier, it can be done by Clustering which is one of the data mining techniques. Clustering is done by applying a self organizing map algorithm on the KNIME (Konstanz Information Miner) analytic tools. Several experiments were carried out with different layer size and data normalization status from one experiment to another experiment. The results were analyzed through pseudo F statistical test and icdrate test. The largest pseudo F statistic value was obtained from the 8th experiment (setting the layer size 2x2 without data normalization) with a pseudo F statistic value of 167.53 for 1kg artifacts and a Pseudo F statistic value of 104.86 for 200 g artifacts where the optimum number of clusters are 4. The smallest icdrate value was obtained from the 5th experiment (setting the 2x3 layer size without data normalization) with an icdrate value of 0.0713 for 1kg artifacts and icdrate value of 0.2889 for 200g artifacts with the best number of clusters being 6. From 12 laboratories can be grouped into 6 groups where each group has the same identification. There are groups 1, 3 and 6 have 1 member, while groups 2, 4 and 5 have 3 members.

Sign in / Sign up

Export Citation Format

Share Document