Search

In order to evaluate the genome evolution and systematics, karyotype analysis of mitotic metaphase chromosomes in 51 taxa of Epimedium and two species of Vancouveria was conducted. The 53 taxa were clustered, based on their karyotype similarity coefficients. Results showed that the 53 taxa studied were all diploid with 12 chromosomes (2n = 2x = 12). Each taxon had one pair of satellites located on pair I of homologous chromosomes. Moreover, the karyotype types of the 53 taxa studied were all type 1A or 2A of Stebbins. It can be concluded that the karyotypes between species are indeed very similar and the genome of Epimedium was conservative in evolution. The cluster analysis of karyotype similarity coefficients could provide valuable clues for the systematics and taxonomy of Epimedium. Results of the cluster analysis strongly supported the previous taxonomic division of E. subg. Rhizophyllum and E. subg. Epimedium. The results also showed that the interspecific relationship was closely correlated with geographical distribution in E. subg. Epimedium and the taxa native to east Asia had the highest genetic diversity in Epimedium. Finally, the origin of the modern geographical distribution of Epimedium was inferred. Results of the present study have significant scientific values in further studies on resource utilisation, taxonomy and phylogeny in Epimedium.

A reconfigurable manufacturing system (RMS) built around a part family is a paradigm to meet dynamic demands and can provide the exact functionality and capacity of the associated part family. The reconfiguration of an RMS is a complex process, given the practical situation of the RMS structure. Furthermore, the reconfiguration complexity of an RMS determines the success of RMS implementation. The reconfiguration complexity can be evaluated using reconfiguration efforts. Eliminating parts involves more reconfiguration efforts compared to complete reconfiguration from a part family and will decrease the reconfiguration complexity of RMS implementation based on the corresponding part family. Therefore, a similarity coefficient that considers the reconfiguration efforts of RMS part family grouping is proposed in this paper. First, the definition of reconfiguration efforts is given based on the system configurations of each part, including actions such as function add, function delete, function swap, and function replace. Second, the common operation sequence (COS) and the longest common subsequence (LCS) between parts are analyzed. Third, the similarity between parts is calculated based on the reconfiguration efforts, the COS, and the LCS. The average linkage clustering algorithm is adopted for grouping the parts into families based on similarity. Finally, a case study is presented to implement the proposed part family grouping method and validate its effectiveness.

The objective of this study was to compared fourteen different similarity coefficients and their influence in sister line wheat cultivars clustering. Seventeen sister cultivars developed from two crosses were used and fingerprinted with 19 wheat microsatellite markers. Comparisons among the similarity coefficients were made using the Sperman correlation analysis, dendogram evaluation (visual inspection and consensus fork index - CIc), projection efficiency in a two-dimensional space, and groups formed by the Tocher optimization procedure. The Sperman correlation coefficients among the fourteen similarity coefficients were all high showing a strong association between them. The correlation coefficient between Dice and Kulczinski and Ochiai I as well as between Hamann and Simple matching and between Kulczinski and Ochiai I was equal to 1. Although visual estimation of the dendograms shows almost identical clustering structures, CIc indexes indicate that all coefficients are not identical. [Projekat Ministarstva nauke Republike Srbije, br. TR 31066]

ÖZETUZAKLIK VE BENZERLİK ÖLÇÜLERİNİN KÜMELEME SONUÇLARINAETKİSİKümeleme analizi, küme sayısı bilinmeyen nesneleri/değişkenleri benzerliklerine göre sınıflandırmaya yarayan bir çok değişkenli analiz tekniğidir. Kümeleme analizinde amaç nesneleri küme içerisinde çok benzer, kümeler arasında farklı olacak biçimde kümelemektir. Benzer olan nesneleri kümelere ayırmak için uzaklık/benzerlik ölçülerinden yararlanılmaktadır. Literatürde verinin türüne göre hesaplanan çeşitli uzaklık ve benzerlik ölçüleri mevcuttur. Uzaklık/benzerlik ölçüsünün seçiminden sonra araştırmanın amacına uygun olarak doğru kümeleme yönteminin seçilmesi gerekmektedir. Kümeleme analizi için bugün literatüre kazandırılmış çok sayıda teknik ve algoritma mevcut olmasına karşın bunları hiyerarşik ve hiyerarşik olmayan teknikler olarak iki başlık altında toplanmak mümkündür.Bu çalışmanın amacı uzaklık/benzerlik kavramlarını açıklayarak, mevcut uzaklık/benzerlik ölçülerini ortaya koyarak hangi durumda hangi uzaklık/benzerlik ölçüsünü kullanılacağına belirlemek, farklı kümeleme tekniği ve uzaklık/benzerlik ölçülerinin kullanılması durumunda elde edilen kümeleme sonuçlarını karşılaştırmaktır. Bu amaç doğrultusunda OECD ülkelerine belirlenen sosyo-ekonomik değişkenlere göre farklı kümeleme teknikleri ve uzaklık/benzerlik ölçülerine göre kümeleme analizi uygulanmış ve kümeleme analizinde sadece uzaklık ölçüsünü seçimi değil aynı zamanda tekniğin seçimi de önemli olduğu sonucuna ulaşılmıştır.ABSTRACTTHE EFFECT OF DISTANCE AND SIMILARITY MEASURES ON CLUSTERINGCluster analysis is a multivariate statistical technique that classifies objects in terms of their similarities, whose number of clusters are unknown. Aim of cluster analysis is classification of the objects in such a way that points in the same cluster have a high degree of similarity while the objects belonging the different cluster have a high degree of dissimilarity. Similarity or distance measures are used to group similar objects into cluster. In literature according to the type of data a variety of distance/similarity measures are avaliable. After selection the correct distance measure a clustering method must be selected in accordance with the purpose of study. Although today there are a large number of cluster analysis techniques and algorithms, it is possible to put them into under two headings as hierarchical and non hierarchical techniques.The aim of this study is to explain the concepts of distance/similarity, determine in which case in which distance /similarity measure to use and compare the outcomes of clustering that obtained the use of different clustering techniques and distance /similarity measures. For this purpose according to socio-economic variables different clustering techniques and distance/similarity measures applied to cluster OECD countries and it’s found that not only the measure of distance but also the selection of the technique is crucial in cluster analysis.

Cellular manufacturing is an effective alternative to batch-type production systems where different products are intermittently produced in small lot sizes with frequent setups, large in-process storage quantities, long production lead times, decreasing throughputs, and complex planning and control functions. An effective approach to forming manufacturing cells and introducing families of similar parts, consequently increasing production volumes and machine utilization, is the use of similarity coefficients in conjunction with clustering procedures. In a similarity coefficients based approach, the results of the clustering analysis depend on the minimum admissible level of similarity adopted for the generic group of clustered items. This is the so-called threshold value of group similarity. The aim of this paper is to identify effective values of the threshold value of group similarity to help practitioners and managers of manufacturing systems form machine groups and related part families. The proposed threshold values for a given similarity coefficient are based on calculation of the percentile of aggregations generated by the adopted clustering algorithm. The importance of the proposed measure of group similarity has been demonstrated by experimental analysis conducted on a large set of significant instances of the cell formation problem in the literature. This analysis can also support the best determination of this percentile-based cut value especially when the number of manufacturing cells is not known in advance.

Today’s distribution warehouses often need to process a far higher volume of smaller orders of multiple products which considerably increases logistics costs. They use the so-called order picking (OP) systems where products have to be picked from a set of specific storage locations by an OP process usually driven by production batches or customer orders. The OP is often very labour-intensive and its efficiency largely depends on the distance the order pickers have to travel, which therefore needs to be minimised. Minimising this distance is affected by several factors e.g. facility layout, shape of storage area, and especially the storage assignment strategy. Products that are frequently ordered together in multi-item, less than unit load customer orders should be stored near each other: this is the correlated storage assignment strategy. This study develops, tests and compares a set of different storage allocation rules based on the application of original similarity coefficients and clustering techniques. Lastly, a case study demonstrates the effectiveness of the proposed rules in minimising logistic costs.

Chenopodium ambrosioides L., conhecida no Brasil por suas propriedades medicinais e usada principalmente para o controle de verminoses intestinais, é pouco estudada quanto à diversidade genética. O objetivo deste trabalho foi avaliar a diversidade genética de 16 indivíduos de C. ambrosioides, provenientes de diferentes municípios da região cacaueira da Bahia, pela técnica de RAPD (DNA polimórfico amplificado ao acaso). Apenas 6,9% das 216 bandas RAPD amplificadas foram polimórficas e a análise de agrupamento evidenciou que não há formação de grupos por área de coleta. Portanto, há pequena variabilidade entre os materiais e esta variabilidade encontra-se distribuída entre as regiões amostradas. Chenopodium ambrosioides L. is known in many parts of Brazil for its medicinal properties, mainly used to control intestinal worms. Its genetic diversity is little studied. The objective of this work was to evaluate the genetic variability of 16 accessions of C. ambrosioides from the cocoa region of Bahia State, Brazil, by the RAPD technique (Random Amplified Polymorphic DNA). Only 6.9% of the 216 amplified RAPD bands were polymorphic and the pattern of dispersion of individuals showed no clustering related to sample site. Therefore, there is low variability among accessions and it is distributed among the accessions from the entire sampled region.

Cluster analysis or classification usually concerns a set of exploratory multivariate data analysis methods and techniques for finding a clustering structure on a dataset. That may refer either to groups of statistical data units or to groups of variables. In this work we deal with a generalization of this paradigm concerning clustering of complex data described by three different types of variables, frequently present in a three-way context. We obtain compatible versions of the same affinity coefficient for measuring similarity between statistical data units described by those three types of variables. A global generalized similarity coefficient is analyzed for such kind of mixed data, often arising in data mining or knowledge mining., This research was partially supported by FCT/POCTI/ and POCTI/FEDER, in the scope of CEAUL Research Project on Applied Multivariate Data Analysis and Modelling.

3rd SMTDA Conference Proceedings, 11-14 June 2014, Lisbon Portugal. Cluster analysis or "unsupervised" classification (from "unsupervised learning", in pattern recognition literature) usually concerns a set of exploratory multivariate data analysis methods and techniques for grouping either statistical data units or variables into groups of similar elements, that is finding a clustering structure in the data. Classical clustering methods usually work with a set of objects as statistical data units described by a set of homogeneous (that is, of the same type) variables in a two-way framework. This paradigm can be extended in such way that data units may be either simple / first-order elements (e.g., objects, subjects, cases) or groups of / second-order or more elements from a population (e.g., subsets, samples, classes of a partition) and/or descriptive variables may simultaneously be of different (e.g., binary, multi-valued, histogram or interval) types. Therefore, one has a complex and/or heterogeneous data set under analysis. In that case classification will often be carried out by using a three-way or a symbolic/complex approach. The present work synthesizes previous methodological results and shows several developments mostly regarding hierarchical cluster analysis of complex data, where statistical data units are described by either a homogeneous or a heterogeneous set of variables. We will illustrate that approach on a case study issued from the statistical literature. The methodology has been applied with success in a data mining context, concerning multivariate analysis of real-life data bases from economy, management, medicine, education and social sciences.