Navegando por Autor "Rossoni, Diogo Francisco"

Agora exibindo 1 - 2 de 2

Spatial variability of pores in oxidic latosol under a conservation management system with different gypsium doses
(Editora UFLA, 2014-10) Carducci, Carla Eloize; Oliveira, Geraldo César; Curi, Nilton; Rossoni, Diogo Francisco; Costa, Alisson Lucrécio; Heck, Richard Jonh
Soil structure is modify when subjected to the agricultural process, i.e., a new spatial organization of the pores system is formed, with relation to the physical quality of it. Thus the aim of this work was to visualize and quantify, through X-ray CT scan, the pores distribution in an oxidic Latosol submitted to a conservation management system with different gypsum doses. Three random trenches were dug lengthwise along the plant row in a very clayey gibbsitic dystrophic Red Latosol, subjected to the following gypsum levels: G0: absence of gypsum; G7: 7 Mg ha-1 and G28: 28 Mg ha-1 of additional gypsum, applied to the surface of the plant row. Undisturbed soil samples were collected in plexiglass tubes at depths of 0.20-0.34, 0.80-0.94 and 1.50-1.64 m after six years of coffee cultivation for quantification of 3D pores obtained by X-ray CT scan. The spatial variability of the soil structure was evaluated by semivariograms generated by 3D images in grayscale. Distribution of the detectable pore diameter was conducted by data mining. Statistical analyzes employed packages 'geoR' to semivariogram and 'randomForest' for data mining in R language. A greater spatial continuity of the pores occurred in the G7 at the three depths. The combined effects of the management system promoted a greater spatial variability of the soil structure in the G28 treatment. Based on geostatistical analyses, it can be infer that the adoption of the system under study promoted changes in the pore network in all directions (X, Y and Z), however with better pores continuity in the vertical direction(Z).
Unsupervised classification of specialty coffees in homogeneous sensory attributes through machine learning
(Editora UFLA, 2020) Ossani, Paulo César; Rossoni, Diogo Francisco; Cirillo, Marcelo Ângelo; Borém, Flávio Meira
Brazil is the largest exporter of coffee beans, 29% world exports, 15% this volume in specialty coffees. Thereby researches are done, so that identify different segments in the market, in order to direct the end consumer to a better quality product. New technologies are explored to meet an increasing demand for high quality coffees. Therefore, in this article has an objective to propose the use of machine learning techniques combined with projection pursuit in the construction of unsupervised classification models, in a sensory acceptance experiment, applied to four groups of trained and untrained consumers, in four classes of specialty coffees in which they were evaluated sensory characteristics: aroma, body coffee, sweetness and general note. For evaluating classifier performance, in the data with reduced dimension, all instances were used, and considering four groupings, the models were adjusted. The results obtained from the groupings formed were compared with pre-established classes to confirm the model. Success and error rates were obtained, considering the rate of false positives and false negatives, sensitivity and classification methods accuracy. It was concluded that, machine learning use in data with reduced dimensions is feasible, as it allows unsupervised classification of specialty coffees, produced at different altitudes and processes, considering the heterogeneity among consumers involved in sensory analysis, and the high homogeneity of sensory attributes among the analyzed classes, obtaining good hit rates in some classifiers.