Methods based on kernel density estimation have been successfully applied for various data mining tasks. Their natural interpretation together with suitable properties make them an attractive tool among others in clustering problems. In this paper, the Complete Gradient Clustering Algorithm has been used to investigate a real data set of grains. The wheat varieties, Kama, Rosa and Canadian, characterized by measurements of main grain geometric features obtained by X-ray technique, have been analyzed. The proposed algorithm is expected to be an effective tool for recognizing wheat varieties. A comparison between the clustering results obtained from this method and the classical k-means clustering algorithm shows positive practical features of the Complete Gradient Clustering Algorithm.

The objective of the study has been to examine whether the freestyle dressage classes are judged reliably and to elaborate a method assessing the consistency of judging. The data contained 13 000 marks of the ten best-ranked horses at nine Grand Prix classes and four Intermediate I classes from ten international competitions. The marks concerned 119 horses in total and were given by 37 judges. A method for evaluating the consistency of judging has been proposed. The index of disagreement (ID) assesses the disagreement of ranking by an individual judge relative to the general ranking based on the sum of marks awarded by five judges. The results show that the mean ID of individual judges is highly differentiated. The ID is influenced by the judge's position in the arena. The consistency of judging was lower in Intermediate I Freestyle Tests than in Grand Prix Freestyle Tests and it varied in different competitions. The conclusion of the study is that the results of the freestyle dressage classes are often biased. The present system of judging should be permanently checked. The offered method of evaluating agreement between judges' rankings may help to improve the consistency of judging and the reliability of the horses' scores.

Summary The impact of fertilizer application on soil aggregate stability is of increasing interest to soil scientists. Aggregate water stability depends primarily on soil organic matter. We studied silty loam and loamy sand aggregates from three long‐term fertilizer treatments (control, pig manure and NPK) which significantly altered the quantity of organic matter. A new approach to examining aggregate stability was used: soil aggregates were immersed in methanol‐water solutions with methanol at 0, 20, 40 and 60% concentration (C), and non‐disrupted aggregates were separated after 30 minutes. The aggregate resistance R(C) against each solution was taken as the percentage of stable aggregates. Overall resistance of the aggregates was taken as the Rtot parameter given by the product of four R(C) values. The R(C) values of all aggregates were positively correlated with C. The R(60) values were independent of the applied fertilizer. The R(0) value for silty loam aggregates ranged from 28% (pig manure) to 7% (NPK), while that for loamy sand equalled 7–9% in all cases. The R(20) values were most effective at differentiating the soils and the fertilizer treatments. However, the Rtot value was a better indicator of aggregate stability. Greater differences in stabilities of aggregates were noted in loamy sand. Mineral fertilizer application seemed to decrease aggregate resistance in both soils. Total organic carbon and nitrogen content in all non‐disrupted aggregates were negatively correlated with methanol concentration (C) of the solution applied for aggregate separation. The largest decrease was for the pig manure treatment, and the smallest was for the control. Porosity and pore size distributions of the aggregates were derived from micro‐tomography and approximated to lognormal pore size distributions. Larger porosities and pores were found in water‐stable aggregates than in methanol‐stable aggregates. It seems that the dominant mechanisms for aggregate instability during fast wetting were not related only to the pore air compression, but to weakening of attractive forces between aggregate particles by water.

No abstract

No abstract