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,...

The Urban spatial structure is affected by spatial interactions among various activity locations, and land uses in the city over the transportation system. Each city has its unique circulation pattern of passengers and freight due to its unique geographic conditions and the distribution of locations of economic activities. In that sense, it is claimed in this chapter per the authors that urban ...

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...

Hierarchical clustering based on pairwise similarities is a common tool used in a broad range of scientific applications. However, in many problems it may be expensive to obtain or compute similarities between the items to be clustered. This paper investigates the hierarchical clustering of N items based on a small subset of pairwise similarities, significantly less than the complete set of N(N...

In the MediaEval 2016 Retrieving Diverse Social Images Task, we proposed a general framework based on agglomerative hierarchical clustering (AHC). We tested the provided credibility descriptors as a vector input for our AHC. The results on devset showed that this vector based on the credibility descriptors is the best feature, but unfortunately that is not confirmed on testset. To merge several...

The paper considers techniques for grouping objects that are described with many quantitative and qualitative attributes and may exist in several copies. Such multi-attribute objects may be represented as multisets or sets with repeating elements. Multiset characteristics and operations under an arbitrary number of multisets are determined. The various options for the objects’ aggregation (addi...

This paper presents the TITech summarization system participating in TAC2009. Specifically, we discuss our results for the Update track. We propose a new method for creating summaries by ordering sentences. After a draft summary is obtained, we conduct agglomerative hierarchical clustering on the sentences of the draft summary based on sentence associativity. Then we use a probabilistic method ...

In this paper, we consider that agents judge the feasible alternatives through linguistic terms –when they are confident in their opinions– or linguistic expressions formed by several consecutive linguistic terms –when they hesitate. In this context, we propose an agglomerative hierarchical clustering process where the clusters of agents are generated by using a distance-based consensus measure.