Guilherme Sousa Alves

CPrefMiner: An Algorithm for Mining User Contextual Preferences Based on Bayesian Networks

In this article we propose CPrefMiner, a mining technique for learning a Bayesian Preference Network (BPN) from a given sample of user choices. In our approach, user preferences are not static and may vary according to a multitude of user contexts. So, we name them Contextual Preferences. Contextual Preferences can be naturally expressed by a BPN. The method has been evaluated in a series of experiments executed on synthetic and real-world datasets and proved to be efficient to discover user contextual preferences.

Visual Perception Similarities to Improve the Quality of User Cold Start Recommendations

Recommender systems are well-know for taking advantage of available personal data to provide us information that best fit our interests. However, even after the explosion of social media on the web, hence personal information, we are still facing new users without any information. This problem is known as user cold start and is one of the most challenging problems in this field. We propose a novel approach, VP-Similarity, based on human visual attention for addressing this problem. Our algorithm computes visual perceptions similarities among users to build a visual perception network. Then, this networked information is provided to recommender system to generate recommendations. Experimental results validated that VP-Similarity achieves high-quality ranking results for user cold start recommendation.

Field data and prediction models of pesticide spray drift on coffee crop

The objective of this work was to generate drift curves from pesticide applications on coffee plants and to compare them with two European drift-prediction models. The used methodology is based on the ISO 22866 standard. The experimental design was a randomized complete block with ten replicates in a 2x20 split-plot arrangement. The evaluated factors were: two types of nozzles (hollow cone with and without air induction) and 20 parallel distances to the crop line outside of the target area, spaced at 2.5 m. Blotting papers were used as a target and placed in each of the evaluated distances. The spray solution was composed of water+rhodamine B fluorescent tracer at a concentration of 100 mg L-1, for detection by fluorimetry. A spray volume of 400 L ha-1 was applied using a hydropneumatic sprayer. The air-induction nozzle reduces the drift up to 20 m from the treated area. The application with the hollow cone nozzle results in 6.68% maximum drift in the nearest collector of the treated area. The German and Dutch models overestimate the drift at distances closest to the crop, although the Dutch model more closely approximates the drift curves generated by both spray nozzles.

Distribution of glyphosate-resistant Amaranthus spp. in Nebraska

BACKGROUND Palmer amaranth (Amaranthus palmeri S. Wats.), common waterhemp (Amaranthus tuberculatus var. rudis), and redroot pigweed (Amaranthus retroflexus L.) are major weeds occurring in fields throughout Nebraska with recurrent grower complaints regarding control with glyphosate. The objective of this study was to investigate the frequency and distribution of glyphosate-resistant Palmer amaranth, common waterhemp, and redroot pigweed populations in Nebraska. The study also aimed to investigate how agronomic practices influence the occurrence of glyphosate resistance in the three Amaranthus species. RESULTS Glyphosate resistance was widespread in common waterhemp (81% of the screened populations), few Palmer amaranth populations were glyphosate-resistant (6% of the screened populations), whereas no glyphosate-resistant redroot pigweed populations were identified in Nebraska. Weed species, geographic region within the state, and current crop were the most important factors predicting the occurrence of glyphosate resistance in fields infested with Amaranthus species in Nebraska. CONCLUSION The intensive glyphosate selection pressure exerted in soybean (Glycine max) fields in eastern Nebraska is one of the major factors causing widespread occurrence of glyphosate resistance in common waterhemp in the state. The relatively low frequency of glyphosate-resistant Palmer amaranth in the state highlights the importance of the application timing and the adoption of multiple modes of action in weed management practices to delay the evolution of glyphosate resistance.

Spray Drift from Dicamba and Glyphosate Applications in a Wind Tunnel

With the recent introductions of glyphosate- and dicamba-tolerant crops, such as soybean and cotton, there will be an increase in POST-applied tank-mixtures of these two herbicides. However, few studies have been conducted to evaluate drift from dicamba applications. This study aimed to evaluate the effects of dicamba with and without glyphosate sprayed through standard and air induction flat-fan nozzles on droplet spectrum and drift potential in a low-speed wind tunnel. Two standard (XR and TT) and two air induction (AIXR and TTI) 110015 nozzles were used. The applications were made at 276 kPa pressure in a 2.2ms-1 wind speed. Herbicide treatments evaluated included dicamba alone at 560 g ae ha-1 and dicamba + glyphosate at 560 + 1,260 g ae ha-1. The droplet spectrum was measured using a laser diffraction system. Artificial targets were used as drift collectors, positioned in a wind tunnel from 2 to 12 m downwind from the nozzle. Drift potential was determined using a fluorescent tracer added to solutions, quantified by fluorimetry. Dicamba droplet spectrum and drift depended on the association between herbicide solution and nozzle type. Dicamba alone produced coarser droplets than dicamba + glyphosate when sprayed through air induction nozzles. Drift decreased exponentially as downwind distance increased and it was reduced using air induction nozzles for both herbicide solutions. Nomenclature: Dicamba; glyphosate; cotton, Gossypium hirsutum L.; soybean, Glycine max (L.) Merr.

VP-Rec: A Hybrid Image Recommender Using Visual Perception Network

A requirement for a great user experience is to meet the exact needs for the usage of a recommender system. Such systems need users historical preferences to reasonably perform, which might not be the case for a cold-start user. This paper presents VP-Rec, a hybrid image recommender system that addresses the new user cold-start problem. VP-Rec combines user visual perception and pairwise preferences as source of information to perform recommendations. First, we infer pairwise preferences from users ratings. Next, we build visual perception networks linking users according to their visual attention similarities. From these two inferred structures, we build consensual prediction models, so that when a new user enters the system, we capture his visual attention and choose the best model that fits him. The system has been tested on two image datasets, getting important improvements in terms of ranking quality (nDCG) when applied to new user cold-start scenario against state-of-art recommender systems.

Spray drift and droplet spectrum from dicamba sprayed alone or mixed with adjuvants using air-induction nozzles

The objective of this work was to evaluate the spray drift and droplet spectrum of dicamba applied alone or with potential drift-reducing adjuvants, using air-induction flat fan nozzles. Standard (XR and TT) and air-induction (AIXR and TTI) nozzles were evaluated in a wind tunnel. The adjuvants used were polymer, ammonium sulfate, vegetable oil, and phosphatidylcholine. The applications were conducted at 276 kPa pressure and 3.5 m s-1 wind speed. The droplet spectrum was measured using a laser diffraction system. Round strings were used as drift collectors, positioned perpendicularly to the wind direction, at 2, 3, 4, 5, 6, 7, and 12 m from the nozzle. Drift was calculated by quantifying, through fluorimetry, a fluorescent tracer added to each solution at 1 g L-1. Droplet spectrum and dicamba drift depend on the interaction between spray composition and nozzle type. Air-induction nozzles are more recommended for dicamba applications, especially the TTI nozzle. Polymer and ammonium sulfate increase droplet size in all nozzle types, which may reduce drift to nearby crops.

Dicamba Spray Drift as Influenced by Wind Speed and Nozzle Type

With the release of dicamba-resistant crops, it is necessary to understand how technical and environmental conditions affect the application of dicamba. This study sought to evaluate drift from dicamba applications through flat-fan nozzles, under several wind speeds in a wind tunnel. Dicamba applications were performed through two standard (XR and TT) and two air induction (AIXR and TTI) 110015 nozzles at 0.9, 2.2, 3.6 and 4.9ms-1 wind speeds. The applications were made at 276 kPa pressure and the dicamba rate was 561 g ae ha-1. The droplet spectrum was measured using a laser diffraction system. Artificial targets were used as drift collectors, positioned in a wind tunnel from 2 to 12m downwind from the nozzles. Drift potential was determined using a fluorescent tracer added to solutions, quantified by fluorimetry. The air induction TTI nozzle produced the lowest percentage of dicamba drift at 2.2, 3.6 and 4.9m s-1 wind speeds at all distances. Dicamba spray drift from XR, TT and AIXR nozzles increased exponentially as wind speed increased, whereas from TTI nozzle drift increased linearly as wind speed increased. Drift did not increase linearly as the volume percentage of droplets smaller than 100 μm and wind speed increased. Nomenclature: Dicamba.

Tensão superficial, potencial hidrogeniônico e condutividade elétrica de caldas de produtos fitossanitários e adjuvantes

A adicao de adjuvantes pode alterar as propriedades fisico-quimicas das caldas de produtos fitossanitarios, contribuindo com a qualidade da aplicacao. Assim, objetivou-se avaliar o comportamento da associacao entre 15 produtos fitossanitarios e quatro adjuvantes de uso agricola na tensao superficial, pH e condutividade eletrica da calda. As analises foram realizadas separadamente para tres categorias de produtos: inseticida, fungicida e herbicida. Foi utilizado o delineamento inteiramente casualizado (DIC), em esquema fatorial 5 x 6, com quatro repeticoes: cinco adjuvantes e seis produtos fitossanitarios. A concentracao das caldas foi estipulada adotando-se volume de calda de 200 L ha-1 e a dose recomendada pelos fabricantes. As caracteristicas avaliadas foram: tensao superficial, condutividade eletrica e pH da calda. As medias foram comparadas entre si pelo teste de Scott Knott, a 0,05 de significância. Os resultados dos produtos fitossanitarios sobre as caracteristicas fisico-quimicas mostraram-se dependentes dos adjuvantes e vice-versa, dada a interacao significativa entre os fatores. Dentre os adjuvantes, o que mais reduziu o pH da calda foi o fosfatidilcoline+acido propionico. Somente o herbicida clorimurom-etilico nao afetou a tensao superficial da calda, que foi reduzida pelos demais produtos. Todos os produtos afetaram a condutividade eletrica, sendo que os maiores aumentos foram obtidos pelos herbicidas 2,4-D dimetilamina e glifosato.

Deposição de calda em plantas de soja empregando pulverizador com barra convencional e barra auxiliar

O deposito das gotas pulverizadas no alvo e fundamental para o exito nas aplicacoes de produtos fitossanitarios. Este trabalho avaliou a deposicao de calda em plantas de soja e as perdas para o solo promovidas por um pulverizador de barra convencional, dotado de pontas de jato plano, e por um pulverizador com barra auxiliar de arrasto, dotado de pontas de jato conico vazio, variando os volumes de calda: 150 e 200 L ha -1 no sistema convencional e 50 e 100 L ha -1 no sistema com barra auxiliar. Foi adicionado o tracador Azul Brilhante a calda para ser detectado por espectrofotometria em folhas da metade superior e inferior da soja e em placas de Petri colocadas junto ao solo. Avaliou-se o espectro de gotas por meio de papeis hidrossensiveis. As pontas de jato conico vazio, empregadas na barra auxiliar, promoveram gotas mais finas e mais uniformes. A deposicao de calda nas plantas de soja utilizando-se a barra auxiliar com ponta de jato conico vazio no volume de 100 L ha -1 foi semelhante as obtidas com a barra convencional. Em relacao as perdas para o solo, nao houve diferenca entre as tecnologias adotadas.