We create a weighted lexical network derived from the cosine similarities of financial news feeds to compare two clustering methods, Newman's Modularity method and hierarchical clustering. We find that hierarchical clustering, clustering documents according to shared unique terms, shows results that are closer to expectation.

Clustering data has been of great interest to many researchers. Hierarchical clustering methods have been preferred because clusters can be visualized as a dendrogram. One of the problems of hierarchical clustering methods, however, is that the resulting dendrogram is not visually pleasing due to the scaling problem. Hence, a series of iterated logarithmic function is proposed so as to mitigate...

In the hierarchical clustering algorithms, it has become a basic difficult problem to determine the optimal clustering number in the dataset, as a result of the influence of outliers and noise points. Therefore, we propose a method to remove these interferential data in two stages in the hierarchical clustering algorithm, which is based on the traditional noise data removal method. Furthermore,...

Symbolic Data Analysis is based on a special descriptions of data – symbolic objects. Such descriptions preserve more detailed information about data than the usual representations with mean values. A special kind of symbolic object is also representation with distributions. In the clustering process this representation enables us to consider the variables of all types at the same time. We pres...

Abstract The vast majority of the existing social network-based group decision-making models require extra information such as trust/distrust, influence and so on. However, in practical process, it is difficult to get additional apart from opinions decision makers. For large-scale making (LSGDM) problem which makers articulate their preferences form comparative linguistic expressions, this pape...

In this paper it is shown a process to demarcate areas with analogous wind conditions. For this purpose a dispersion graph between wind directions will be traced for all stations placed in the studied zone. These distributions will be compared among themselves using the hierarchical clustering algorithm. This information will be used to build a matrix, letting us work with all relations simulta...

The Baire metric induces an ultrametric on a dataset and is of linear computational complexity, contrasted with the standard quadratic time agglomerative hierarchical clustering algorithm. In this work we evaluate empirically this new approach to hierarchical clustering. We compare hierarchical clustering based on the Baire metric with (i) agglomerative hierarchical clustering, in terms of algo...

â â  in this research, temporal and spatial behavior of subtropical high pressure was studied at the level of 600 hpa. this study has been done using daily data of geopotential height at 12 gmt in ncep/ncar database with spatial resolution of 2.5ã—2.5 degree in a 55 years period including 20089 days from january, 1st, 1951(dey, 11th, 1329) to december, 31st,2005(dey, 10th,1384). at first, the m...

Characterization and identification of multielement nanoparticles thanks to the use a spICP-ToF-MS coupled hierarchical agglomerative clustering (HAC).

In this paper we address the problem of building object class representations based on local features and fast matching in a large database. We propose an efficient algorithm for hierarchical agglomerative clustering. We examine different agglomerative and partitional clustering strategies and compare the quality of obtained clusters. Our combination of partitional-agglomerative clustering give...