---

The objective of this study was to classify the Shoosh Aquifer to several zones with different water quality in Khuzestan Province, Iran. In this regard, the performance of classification methods (Discriminant function and Cluster analysis) for the classification of groundwater based on the level of pollution with an emphasis on the problem of over-fitting in training data were considered. An over-fitted model will generally have poor predictiveperformance, as it can exaggerate minor fluctuations in the data. Cluster Analysis(CA) was adopted to spatially explain the similarity of sampling stations with respect to measured parameters. Three methods for variable selection were used including regularized discriminant analysis, principal component analysis and Wilks's lambda method. The best algorithm for variable selection was Wilks'lambda which resulted in reducing the generalization error of the test sample to 0.1 for leave-one-out and 4-fold cross-validation. The second best performed algorithm was regularized discriminant function with 0.167 and 0.133 misclassification error for the two above-mentioned methods, respectively. Principal component analysis did not proved to be a promising algorithm for variable selection in the classification methods.

---

This study was conducted to evaluate genetic variation among 70 sunflower recombinant inbred lines (RIL) derived from the crosses PAC2 × RHA266 together with parents based on seed morphological traits by using a rectangular lattice design with two replications. Seed morphological such as kernel length, kernel width, kernel diameter, 100-kernel weight, percentage of hull, percentage of dehulled kernel and seed yield per plant was measured. Analysis of variance revealed significant differences among lines for the studied traits. The highest coefficient of phenotypic variation was observed for seed yield per plant (23.42) and the lowest one was observed for percentage of dehulled kernel (1.37). The highest heritability was observed for 100-kernel weight (0.995) and kernel width (0.990) and the lowest one was observed for the yield per plant (0.521). The highest correlation coefficients were observed between kernel diameter and kernel width (0.908). Principal component analysis reduced the seed characteristics traits to 2 components explaining 81% accumulative variance. By using Ward clustering method based on seed morphological traits the 72 studied sunflower lines were classified into six groups.

---

Drought is known as an important factor limiting growth and product of field crops in most parts of the world and Iran. In the present work, the genetic diversity of 100 inbred lines of sunflower was investigated based on agro-morphogical characters with simple lattic design with two replications under normal and drought stress conditions. Combined analysis of variance revealed significant differences among lines for most of studied traits. Uneder normal condition, the highest coefficient of genetic variation was observed for stem diameter and the lowest one observed for relative water content. In drought stress condition, the highest coefficient of genetic variation was observed for seed yield per plant and the lowest one observed for days to flowering. The results of correlation analysis showed that there is significant and positive correlation between seed yield per plant with most of the studied traits in both stress conditions. Stepwise regression analysis revealed that under drought stress condition 73.9 percent of seed yield per plant variation was expailed by heed diameter, leaf width and petiol length and in normal condition 73.6 yield grain per plant variation explained by head diameter and plant height. Cluster analysis grouped lines into 4 clusters in each one of normal and drought conditions but the distruption of lines within groups were differents depending to stress environment that present the genetic variability for drought tolerance in sunflower lines.

---

This study aims to explain the role of children in family purchase decisions based on the parent clusters with a market-oriented approach. Accordingly, the present study is exploratory in terms of objective and mixed and applied in terms of result. In the first step, 30 statements were extracted using in-depth interviews that after examining the face and content validity, and exploratory factor analysis in the pre-test, these statements were reduced to 29 and the final questionnaire was developed. To determine the sample size, the experimental rules were used, therefore given the 29 resulting statements, the sample size was obtained 290 that a total of 400 questionnaires with a share of 25% were distributed among four urbanite groups of parents in Mazandaran, among which 311 questionnaires were usable. To achieve the parents’ attitude factors, exploratory factor analysis was applied and after removing four statements during various stages, 25 remaining questions were classified in six factors. In the second step, using K--mean cluster analysis, three parent clusters (traditional, halfway and moderns) resulted from the hierarchical cluster analysis approach were analyzed based on six attitudinal factors. Results showed that in traditional, halfway and modern clusters, three factors are of paramount importance including interaction, product and time. Finally, by matching the demographic characteristics of the statistical sample with three parental clusters, recommendations were presented with a market-oriented approach among four types of footwear and clothing, educational, toys, food and drink products.

---

Aims: In the present study, the morphometric and meristic traits of Alburnus chalcoides from the Chelvand (Astara), Khalesara (Talesh), Siahdarvishan (Anzali wetland), Sefid and Pol (Guilan Province), Tonekabon and Babol (Mazandaran Province) rivers were compared.
Materials & Methods: For this purpose, a total of 170 specimens were collected from above-mentioned rivers. In the Lab some 39 morphometric and 10 meristic traits were recorded. All data were analyzed by using One-Way ANOVA, Duncan test, Kruskal-Wallis, principal component analysis (PCA), canonical variate analysis (CVA) and cluster analysis.
Finding: The results showed a significant difference in 36 morphometric and five meristic traits among the studied populations. CVA result separated the studied populations in the southern parts of the Caspian Sea basin. Also, cluster analysis placed the Chelvand River population in a distinct clade. 
Conclusion: This study showed high diversity of morphometric compared to meristic traits and better performance of these traits in population's distinction.

---

---

In this study, the morphological variation of Oxynoemacheilus bergianus was studied in the different rivers of the Namak Lake and Caspian Sea basins using traditional morphometric method. For this purpose, a total of 76 specimens were collected from eight river systems and after fixation into 4% buffered formalin, transfered to the laboratory, a total of 31 morphological characteristics were measured using digital calipers. After standardization, the morphometric data were analyzed using multivariate analysis including principal component analysis (PCA), canonical variate analysis with p-value obtained from MANOVA (MANOVA/CVA) and cluster analysis (CA). The results showed significant differences in 24 traits between the studied populations (P<0.05), which anal fin depth and the ventral-anal fin distances were main discriminative ones. CVA analysis was able to separate the studied populations. Also, CA placed the Gharesu and Sefid populations in a clade and separate from other populations. The observed differences may be related to phenotype plasticity in response to environmental conditions.

---

---

This study was conducted to investigate the morphological variation of Planiliza abu in the Tireh (Tigris Basin), Kor (Kor River Basin) and Jegin (Hormozgan Basin) rivers using traditional (TM) and geometric morphometric (GM) methods. For this purpose, a total of 62 specimens were collected using electrofishing device and Salik net. In the Lab, 21 morphometric traits were measured. Then, to extract the morphological data in the geometric method, 16 landmark-points were defined and digitized on the photographs taken from the left side of fish using tpsDig2 software. The results showed that the studied populations had significant differences in 7 morphometric traits (P<0.05). The differences in the geometric method were those of the head size, body depth, pectoral fin position and caudal peduncle length. Based on the results, GM method showed higher accuracy to reveal the morphological variations in the generalist species of Planiliza abu, which can inhabit a wide range of habitats.

---

This paper used the relevant data to study the frequency of thunderstorm occurrence in Tabriz as a sample of North West Iran’s climate. Where required, the methods such as Spearman rank correlation, linear and sixth order polynomial trend, as well as the cluster analysis according to the average linkage (within group) has also been applied utilizing the block distance standard for the monthly classification of the thunderstorm occurrence. The obtained results showed that the month of May, during various monthly time scales, and the spring season in seasonal intervals, undergo the majority of thunderstorm occurrences. The methodological statistics show that, secondary climatic features like hails, heavy showers, floods, and lightening are dominant climatic features in Tabriz during the thunderstorm occurrence. The above-mentioned climatic hazards are part of the characteristics of Tabriz climate and all of North West Iran, which bring about many economic, social and environmental damages, particularly for the farmers and animals farms. The probability distribution of thunderstorm occurrence in Tabriz indicates the irreversibility of their occurrence chiefly in short time return periods. Therefore, thunderstorms and their resulting secondary features are regarded to be a permanent hazard in Tabriz.

---

This study was conducted to assess genetic diversity among the 32 tall fescue half-sib families using a randomized complete block design (RCBD) experiment in the four replicates. Based on analysis of variance, significant differences were observed among studied genotypes at the probability of 1% for plant height, canopy diameter, days to heading, days to pollination, crown diameter, fresh forage yield, dry forage yield, number of stem and seed yield in first harvest and in canopy diameter, crown diameter, fresh forage yield and dry forage yield in second harvest. Based on the results of mean comparisons, highest dry forage yield in the first harvest was obtained in genotype 32 by 758.5 grams. Principal component analysis by considering eigenvalues greater than one, caused to introduction of three components which determined 80.5% of the variation among the samples. In cluster analysis, the greatest of distinction between the groups was achieved with three clusters, and by cutting the dendrogram genotypes in three groups. According to the results, the third cluster was superior to other two clusters in terms of most traits. The genotypes of third cluster, according to the value of this cluster in terms of forage yield and seed yield will be of particular importance in breeding programs. In the breeding of cross-pollinated forage crops, success in selection depends on creating diversity by genetic recombination and achievement of heterosis. Due to the distance between groups 1 and 3, probably the most successful crosses will be achieved among genotypes of these two groups.

---

 Identifying resistant genotypes is necessary to control wheat take-all disease Gaeumannomyces graminis var. tritici. In this study, 30 bread wheat genotypes were evaluated under greenhouse and field conditions. The genotypes were evaluated with fifteen molecular markers (SSR and specific primers for translocation wheat-rye). The genotypes were divided into four groups based on disease severity (the greenhouse) and agronomic traits (the field). Chi-square results showed the interactions for these groupings. The correlation between disease severity and agronomic traits indicated that plant resistance is strongly dependent on plant yield. Based on cluster analysis for molecular data (based on simple matching similarity coefficient and UPGMA method), genotypes were separated into resistant and susceptible ones. The correlation between disease severity and amplified loci showed that disease resistance is interactive with xbarc232, xbarc124, and gpw95001 markers. Resistance to take-all disease is probably associated with the interaction of several genes. These results add significant information to our knowledge of the chromosomal location of genes for the take-all disease.

---

ABSTRACT Rosa damascena has attracted considerable attention in horticulture, biochemistry and in pharmacology because of the fragrance of the flowers and the high content of its bio-logically active substances. There is limited information available on factors controlling flower yield and its components. The present study reports an investigation on flower yield and the various factors affecting it, in Iranian Rosa damascena using sixteen geno-types collected from different regions in Iran. These genotypes were planted at Isfahan Agriculture Research Center, using a randomized complete block design with three repli-cations in 2005. Twelve yield determining characters including flower yield per plant, number of flowers per plant, fresh flower weight, plant height, canopy diameter, length and width of bud, number of petal per flower, length and width of receptacle, fresh weight of petal as well as flowering period were recorded. Rosa damascena genotypes sig-nificantly varied for all the traits except for the fresh weight of petals per flower. Pheno-typic and genotypic coefficients of variation for flower yield per plant (48.03%, 36.49%), number of flowers per plant (40.65%, 26.99%), number of petals per flower (37.56%, 32.31%) were higher than the coefficients for other tested traits. Cluster analysis revealed that Khuzestan and Shiraz 2 genotypes were the most related ones, while the most inde-pendent ones were the western and eastern Azerbaijan genotypes. Results of stepwise re-gression analysis showed that 90 percent of total variation of flower yield per plant could be explained by the number of flowers per plant. This suggested that number of flowers per plant is the most important component determining flower yield per plant. According to principle component analysis, parents with a high standing of the following traits: fresh weight of flower, number of petals per flower and bud width, may well be employed for hybridization in a breeding program. Regression and cluster analyses proved to be the appropriate multivariate analyses for an identification of Rosa damascena genotypes pos-sessing the desirable characters for hybridization to develop improved cultivars.

---

Market segmentation by artificial neural networks has no deep root in the history. Generally, this ever developing approach has started since several years ago, and developed to other marketing areas. Now, beside statistical techniques, it is considered as one of the most popular methods in Custamer classification. In Due to the necessity of recognizing the target market for a specific company, a need for the usage of an effective approach for customers grouping was recognized, in the Present research, and finally cluster analysis with SOM neural networks, was selected, and used for customers clustering. Firstly, beneficent criteria for market segmentation were identified, and then a proper, questionnaire was designed. After gathering the questionnaires and collecting the data, using artificial neural networks, the customers were clustered, and the obtained, results were analyzed. At the end, the Findings of this method were compared with those of the traditional methods for clusteringusing K-means.

---

Identifying Inefficient Bank Branches Using DEA and Use of Integrated Strategies in Order to Increase Performance Branches Naser Hamidi1, Reza Akbari Shemirani2, Safar Fazli3 1. Assistant Professor, Department of Managemen and Accounting, Islamic Azad University Qazvin Branch, Qazvin, Iran 2. M. A , Department of Managemen and Accounting , Islamic Azad University Qazvin Branch, Qazvin, Iran 3. Assistant Professor, Department of Social Sciences, Imam Khomeini International University, Qazvin, Iran Received: 27 /10/2010 Accept: 9/3/2011 Among the various economic and financial organizations, banks are considered and discussed as one of the most important pillars of any economic system. Therefore, considering the important role of banks in the developed countries and their multiple branches, bank branch performance measurement is important. The main purpose of this work is measure branch performance and the use of integration strategy to obtain efficient branches. Accordingly, first the efficiency of bank branches conceptual system was defined. Then using the action plan were inputs and outputs determined. In the next stage, Tehran province branch efficiency was calculated through Data Envelopment Analysis model as non-radial (SBM) to determine the inefficient branches. Based on the bank merger policy, the clusters were determined and the branches in each cluster were merged in a binary form. The final stage of the integration of non-radial (SBM) to assess the performance of branches and compare it with their initial performance.

---

In thirty-three wild cherry (Prunus avium L.) accessions from Central Serbia, tree, leaf, fruit, and stone morphological characters were evaluated during three consecutive years. The goal was to detect relationships between the genotypes and to identify the most useful traits for discrimination among them. The study revealed a high variability in the set of the evaluated wild cherry accessions and considerable differences were found among them in all studied attributes. The majority of important correlations were determined among variables representing fruit and leaf size, and variables related to color. Cluster analysis distinguished wild cherry accessions into two distinct groups. In PCA, fruit and leaf traits such as leaf length and width, and fruit height, width, and weight, and skin flesh and juice color were predominant in the first two components, indicating that they were useful for the assessment of wild cherry germplasm characterization. These results indicate that these accessions must be conserved as valuable genetic resources to enrich the cherry gene pool and can be used for improving breeding efficiency of important horticultural traits worldwide.

---

Classification of statistical elements is one of the challenging areas in management science. This subject has changed to an interesting research areas. Although methods of cluster analysis and discriminant analysis are used as common methods in the classification, there is a doubt about their application due to high statistical errors of the methods. In this paper, it is tried to combine analysis approach of statistical discrimination and OR technique and a new method titled goal discriminant analysis is developed. Four discriminant analysis methods titled FLDF, FG, GP1 and GP2 are applied in this paper. In order to evaluate its efficiency in management science area, the fourfold technique has been employed in 5 managerial case studies. The results show that the FLDF method, which is a discriminant analysis method, is more efficient than other methods. Moreover, goal discriminant methods have more efficiency in management classification with over two groups.

---

In the present study, random amplified polymorphic DNA (RAPD) and inter-simple sequence repeat (ISSR) markers were used to estimate genetic diversity and relationships among 35 cluster bean [Cyamopsis tetragonoloba (L.) Taub] genotypes. For RAPD analysis, 20 random primers were used which amplified 164 bands, 147 of which were polymorphic with an average polymorphism of 90.3%. The amplified products varied in size from 250 to 2,050 bp. For ISSR analysis, 10 primers were used which produced 105 bands, 102 of which were polymorphic (97%). The size of amplified bands ranged from 450 to 3,500 bp. The efficiency of primers in generating sufficient information for genetic diversity analysis was computed using discriminatory power (D_j), which ranged from 0.40 to 0.98 for RAPDs and 0.44 to 0.99 for ISSRs. Jaccard similarity coefficients were used to estimate the genotypic association with each other, which varied from 0.38 to 0.91 for RAPDs and from 0.20 to 0.88 for ISSRs. Cluster analysis indicated that all 35 genotypes could be distinguished by both RAPD and ISSR markers. Both of the methods (RAPD and ISSR) showed significant correlation (r= 0.69), implying their equal importance in cluster bean diversity analysis.

---

The objectives of the present study were to evaluate spring wheat recombinant inbred lines (RILs) of diverse origin by estimating genetic parameters viz., variability, character association, cluster analysis, and principal component analysis (PCA) for spot blotch resistance and yield components at BHU Agricultural Research Farm during 2010-2011. Grain yield per plot was significantly and positively associated with biomass, 1,000-grain weight, harvest index, chlorophyll content, and grains per spike at genotypic level. The line 65 exhibited lowest mean of AUDPC value (632) indicating its potential as resistant parent. Cluster analysis grouped all the 324 spring wheat lines into 19 clusters using Ward’s method. Extreme divergence was observed among clusters. By using D²-statistics, the highest inter cluster distance (584.72) was found between Clusters VIII and XIX. Cluster VIII recorded highest mean values for chlorophyll content, peduncle length, bio-mass, grains per spike, 1000-grain weight and grain yield. The major contributing trait towards genetic divergence was found to be AUDPC (60.36%). First 5 principal components (PC1, PC2, PC3, PC4 and PC5) accounted for proportionate values of 20.66, 17.96, 15.07, 8.28, and 7.38%, respectively, contributing 69.35% of the total variability. The second PCs had high positive PC value for plant height, biomass, and 1,000-grain weight. The breeding objectives of the present experiment was to identify genetically diverse wheat spot blotch resistant RILs for developing high yielding spot blotch resistant cultivars especially adopted to south Asia in future breeding programs.

---

Wheat gluten is one of the most important plant proteins of fairly low price consisting of glutenins and  gliadins. The hydrophobic nature and low water solubility of gluten has limited its applications in many food and non-food products. In order to improve the applications of gluten, it was first fractionated into its main components; gliadins and gluteneins and then they were acetylated using acetic anhydride. In the next stage, the rhelogical properties of these proteins at different pH values (3, 6 and 9) were compared with their corresponding controls. The pH had a great impact on the degree of acetylation. Glutenin had the highest, while gliadin had the lowest degree of acetylation at different pH values. Moreover, all these proteins were de-aceytelated at pH 3. Hardness, gumminess, chewiness and elasticity of the acetylated gluten and glutenin decreased significantly at pH 3.