under field conditions. Thus, the objectives of this study were two-fold. First, to evaluate the potential of classic methods of discriminant analysis, as well as more sophisticated algorithms, for the classification of bare soil, annual
Francisca López-Granados, M. Teresa Gómez-Casero, José M. Peña-Barragán, Montserrat Jurado-Expósito and Luis García-Torres
Yun-wen Wang, Bruce L. Dunn and Daryl B. Arnall
assessing variability of reflectance measurements on plant growth and physiological conditions, using a statistical approach to extract useful information could be used to improve crop performance ( Parihar et al., 2003 ). Discriminant analysis, a
M. Teresa Gómez-Casero, Francisca López-Granados, José M. Peña-Barragán, Montserrat Jurado-Expósito, Luis García-Torres and Ricardo Fernández-Escobar
., 1999 ); decision tree technology in corn ( Goel et al., 2003 ); multivariate data analysis methods like principal components; and discriminant analysis in tomato ( Lycopersicon esculentum Mill.), wheat, and corn ( Girma et al., 2005 ; Karimi et al
Ettore Barone, Luigi Di Marco, Francesco P. Marra and Maria Sidari
Canonical discriminant analysis (CDA) of morphometric data of buds, leaves, and fruit, as well as isozyme analysis (esterase, peroxydase, and acid phosphatase) of leaf samples, were used to identify eight male pistachio selections and 10 female pistachio cultivars. According to the CDA, 77% and 93% of the total variance was summarized by the first three canonical discriminant functions for the female and male selections, respectively. Fruit characteristics, particularly fruit fresh and dry weights and fruit length, accounted for most of the discriminatory power for the female cultivars, while the dimensions of the leaves, principally leaf rachis length, were the most effective discriminating characters for the males. Isozyme analysis showed a higher degree of polymorphism in the male than the female genetic pool. Hence, using only three enzymes, it was possible to identify all of the male selections, but only 50% of the females. Peroxidase polymorphism clearly demonstrated the greater phylogenetic distance between `Kerman' and the local cultivars, as well as between `Cerasola', a quite different cultivar with a reddish hull, and the others tested. The combination of CDA and isozyme analysis enhanced the possibility of uniquely identifying the female cultivars.
potential information sources that consumers use in their purchase decisions ( Blackwell et al., 2006 ). Table 1 lists these question statements. Table 1. Summary of the discriminant analysis and t test performed to compare behavioral differences between
J.G. Cruz-Castillo, S. Ganeshanandam, B.R. MacKay, G.S. Lawes, C.R.O. Lawoko and D.J. Woolley
Processing plants requires that cultivars be categorized as either small, medium, or large peas to meet the different markets. A reliable nutrient diagnosis system based on sweet pea leaf analysis should be robust to the type of cultivar. The objective of this study was to determine whether the type of cultivar should be taken into account in producing the nutrient diagnosis. Proportions of peas in categories 1 (small) to 5 (large) were determined for 18 cultivars produced under commercial conditions over 3 years. Cluster analysis was conducted with the constraint of revealing three groups, as homogeneous as possible with regard to their proportions in the different categories. Three cultivars were identified as belonging to the small, nine to the medium, and six to the large group. The archetype of each group was characterized. The function discriminated among the cultivars perfectly along the canonical axes. However, no classification was possible when the nutrient composition variables (N, P, K, Ca, Mg, B, Fe, Mn, Zn) were used for discriminating cultivars' types. Hence, sweet pea cultivars of different types do not differ substantially in leaf composition.
M. Mcharo, D. LaBonte, R.O.M. Mwanga and A. Kriegner
Molecular markers linked to resistance to sweetpotato chlorotic stunt closterovirus [SPCSV (genus Crinivirus, family Closteroviridae)] and sweetpotato feathery mottle virus [SPFMV (genus Potyvirus, family Potyviridae)] were selected using quantitative trait loci (QTL) analysis, discriminant analysis and logistic regression. Eighty-seven F1 sweetpotato [Ipomoea batatas (L.) Lam.] genotypes from a cross of `Tanzania' and `Wagabolige' landraces were used to generate DNA marker profiles for this study. Forty-five of the clones were resistant to SPCSV while 37 were resistant to SPFMV. A combination of 232 amplified fragment length polymorphism (AFLP) markers and 37 random amplified polymorphic DNA (RAPD) markers obtained were analyzed to determine the most informative markers. All three statistical procedures revealed that AFLP marker e41m33.a contributed the greatest variation in SPCSV resistance and RAPD marker S13.1130 accounted for most of the variation in SPFMV resistance. The power of discriminant and logistic analyses is that you do not need a parent-progeny population. An evaluation of these two models indicated a classification and prediction accuracy rates of 96% with as few as four markers in a model. Both multivariate techniques identified one important discriminatory marker (e44m41.j) for SPCSV and two markers (e41m37.a and e44m36.d) for SPFMV that were not identified by QTL analysis.
Tzu-Fang Yeh and Li-Chun Huang
' responses to the extracted value factors. Discriminant analysis was then used to identify the main differences between the gender groups, as well as the geographic groups (rural area vs. urban area) for the consumption value of flowers. All statistical
Bridget K. Behe and Dennis J. Wolnick
We determined the influence of demographic characteristics and floral knowledge (measured as product experience) on the type of floral product purchased. A sample of 401 Pennsylvania residents was divided into fresh flower and flowering plant consumer segments. Results of discriminant analyses showed the two segments were moderately distinct. Purchasers of fresh flowers were younger and more likely employed outside the home than those who purchased flowering plants, but the latter had more blooming plants in their homes than did consumers of fresh flowers. Consumers of flowering plants and of fresh flowers did not differ in their level of floral knowledge or demographic characteristics. Minor differences were found between the two segments that were not substantial enough to justify distinct marketing strategies.