A Semi-supervised Clustering Method through Bottleneck Distance Exploration

2016 ◽  
Author(s):  
Yuan Yao ◽  
Yan Li ◽  
Ke Wang ◽  
Zhichao Huang ◽  
Yunming Ye
Keyword(s):  
Clustering Method ◽  
Supervised Clustering

Applying Supervised Clustering to Landsat MSS Images into GIS-Application

2013 ◽  
Vol 4 (3) ◽  
pp. 114-122
Author(s):  
Miguel Torres ◽  
Marco Moreno-Ibarra ◽  
Rolando Quintero ◽  
Giovanni Guzmán
Keyword(s):  
Supervised Classification ◽  
Satellite Images ◽  
Training Data ◽  
Clustering Method ◽  
Maximum Likelihood Classification ◽  
Gis Application ◽  
Supervised Clustering ◽  
Gis Applications ◽  
Geographic Environment ◽  
Decision Making Model

In this paper, the authors describe and implement an algorithm to perform a supervised classification into Landsat MSS satellite images. The Maximum Likelihood Classification method is used to generate raster digital thematic maps by means of a supervised clustering. The clustering method has been proved in Landsat MSS images of different regions of Mexico to detect several training data related to the geographic environment. The algorithm has been integrated into Spatial Analyzer Module to improve the decision making model and the spatial analysis into GIS-applications.

Detection of Leading Experts from ResearchGate

2018 ◽  
Vol 5 (3) ◽  
pp. 67-86
Author(s):  
Eya Ben Ahmed
Keyword(s):  
Social Media ◽  
Social Network ◽  
Technological Development ◽  
Clustering Method ◽  
Web Based ◽  
Supervised Clustering ◽  
And Performance

This article describes how thanks to the technological development, social media has propagated in recent years. The latter describes a range of Web-based platforms that enable people to socially interact with one another online. Several types of social media appeared. In this context, the author focuses on scientific social network which connects the researchers and allow them to communicate and collaborate online. In this paper, we, particularly, aim to detect the scientific leaders through firstly detect communities in social network then identify the leader of each group. To do this, the author introduces a new hierarchical semi-supervised clustering method based on ordinal density. The results of carried out experiments on real scientific warehouse have shown significant profits in terms of accuracy and performance.

scholarly journals Constrained Self-Supervised Clustering for Discovering New Intents (Student Abstract)

2020 ◽  
Vol 34 (10) ◽  
pp. 13863-13864
Author(s):  
Ting-En Lin ◽  
Hua Xu ◽  
Hanlei Zhang
Keyword(s):  
Prior Knowledge ◽  
Feature Engineering ◽  
Clustering Method ◽  
Dialogue System ◽  
Pairwise Constraints ◽  
Supervised Clustering ◽  
Benchmark Datasets

Discovering new user intents is an emerging task in the dialogue system. In this paper, we propose a self-supervised clustering method that can naturally incorporate pairwise constraints as prior knowledge to guide the clustering process and does not require intensive feature engineering. Extensive experiments on three benchmark datasets show that our method can yield significant improvements over strong baselines.

A Supervised Clustering Algorithm Based on Representative Points and its Application to Fault Diagnosis of Diesel Engine

2010 ◽  
Vol 121-122 ◽  
pp. 958-963 ◽  
Author(s):  
Yan Jun Pang ◽  
Wei Pan ◽  
Kai Di Liu
Keyword(s):  
Fault Diagnosis ◽  
Diesel Engine ◽  
Iterative Algorithm ◽  
Clustering Algorithm ◽  
Clustering Method ◽  
Weighted Distance ◽  
Supervised Clustering ◽  
Representative Points ◽  
Class Representative

By terms of extracting quantization values of each index making contributions to classification, this paper defines index classification weight; and also defines class representative points, weighted distance between samples and representative points; provides an iterative algorithm of searching class representative points, establishes a supervised clustering method based on representative points and it is apply into Fault diagnosis of Diesel Engine.

Kernal Based Semi-Supervised Clustering and its Application in Leave Recognition of Bauhinia Blakeana Leaves

2013 ◽  
Vol 756-759 ◽  
pp. 3849-3854
Author(s):  
Xi Yang Yang ◽  
Fu Sheng Yu
Keyword(s):  
Fuzzy Clustering ◽  
Principal Components ◽  
Leaf Spot ◽  
Clustering Algorithm ◽  
Projection Algorithm ◽  
Leaf Spot Disease ◽  
Clustering Method ◽  
Supervised Clustering ◽  
Spot Disease ◽  
Fuzzy Clustering Algorithm

A novel kernel based semi-supervised fuzzy clustering algorithm is proposed, and its iterative formula is given. This new algorithm can effectively improve the efficiency of the clustering algorithm. Combined with Fisher projection algorithm, two principal components are extracted from 7 hue statistics and 11 green value statistics, this new semi-supervised clustering method is applied to recognize the angular leaf spot disease of Bauhinia blakeana. The results showed that the consistent rate is 100% for the labeled leaves, and above 95% for other unlabeled leaves.

Sign in / Sign up

Export Citation Format

Share Document