Evaluatoin of Agglomerative Hierarchical Clustering Methods
نویسندگان
چکیده
This paper describes the findings from evaluating the performance of agglomerative hierarchical cluster methods for determining seasonal factor groups. Seasonal factor groups are usually determined by traditional cluster analysis based on various similarity measures. Agglomerative hierarchical methods merge telemetry traffic monitoring sites (TTMSs) into groups according to their similarities. A wide variety of similarity measures may be used in cluster analysis. This study evaluated a total of eight agglomerative clustering methods: average linkage method, centroid method, EML method, flexible-beta method, McQuitty's similarity analysis method, median method, single linkage method, and Ward's minimum-variance method. Multi-year data collected between 1997 and 2000 from 21 TTMSs in Florida Department of Transportation (FDOT) District 4 were utilized in this study. The Pseudo F (PSF) statistic was employed as the criterion for determining the number of clusters. The average linkage, centroid, and single linkage methods were found to be more robust to outliers than the other methods. The study also found that the McQuitty's (MCQ) method performed better than the other methods on grouping TTMSs after outliers were eliminated. When the MCQ method was applied to analyze the historical data collected between 1997 and 1999, TTMSs were not consistently assigned to the same cluster group across years. Roadway functional classes were found to be insignificant in determining seasonal groups, while spatial location was a more significant factor because a TTMS tended to be clustered with those in its proximity.
منابع مشابه
A Comparative Agglomerative Hierarchical Clustering Method to Cluster Implemented Course
There are many clustering methods, such as hierarchical clustering method. Most of the approaches to the clustering of variables encountered in the literature are of hierarchical type. The great majority of hierarchical approaches to the clustering of variables are of agglomerative nature. The agglomerative hierarchical approach to clustering starts with each observation as its own cluster and ...
متن کاملAlgorithms for Model-Based Gaussian Hierarchical Clustering
Agglomerative hierarchical clustering methods based on Gaussian probability models have recently shown promise in a variety of applications. In this approach, a maximum-likelihood pair of clusters is chosen for merging at each stage. Unlike classical methods, model-based methods reduce to a recurrence relation only in the simplest case, which corresponds to the classical sum of squares method. ...
متن کاملMultiDendrograms: Variable-Group Agglomerative Hierarchical Clusterings
MultiDendrograms is a Java-written application that computes agglomerative hierarchical clusterings of data. Starting from a distances (or weights) matrix, MultiDendrograms is able to calculate its dendrograms using the most common agglomerative hierarchical clustering methods. The application implements a variable-group algorithm that solves the non-uniqueness problem found in the standard pai...
متن کامل2 Review of Agglomerative Hierarchical Clustering Algorithms
Hierarchical methods are well known clustering technique that can be potentially very useful for various data mining tasks. A hierarchical clustering scheme produces a sequence of clusterings in which each clustering is nested into the next clustering in the sequence. Since hierarchical clustering is a greedy search algorithm based on a local search, the merging decision made early in the agglo...
متن کاملInteractive Visualization of Hierarchical Clusters Using MDS and MST
In this paper, we discuss interactively visualizing hierarchical clustering using multidimensional scaling (MDS) and the minimal spanning tree (MST). We can examine the sequential process leading to agglomerative or divisive hierarchical clustering, compare the di erent agglomerative methods, and detect in uential observations better than is possible with dendrograms.
متن کامل