The class imbalance problem has been a hot topic in the machine learning community in recent years. Nowadays, in the time of big data and deep learning, this problem remains in force. Much work has been performed to deal to the class imbalance problem, the random sampling methods (over and under sampling) being the most widely employed approaches. Moreover, sophisticated sampling methods have been developed, including the Synthetic Minority Over-sampling Technique (SMOTE), and also they have been combined with cleaning techniques such as Editing Nearest Neighbor or Tomek's Links (SMOTE+ENN and SMOTE+TL, respectively). In the big data context, it is noticeable that the class imbalance problem has been addressed by adaptation of traditional techniques, relatively ignoring intelligent approaches. Thus, the capabilities and possibilities of heuristic sampling methods on deep learning neural networks in big data domain are analyzed in this work, and the cleaning strategies are particularly analyzed. This study is developed on big data, multi-class imbalanced datasets obtained from hyper-spectral remote sensing images. The effectiveness of a hybrid approach on these datasets is analyzed, in which the dataset is cleaned by SMOTE followed by the training of an Artificial Neural Network (ANN) with those data, while the neural network output noise is processed with ENN to eliminate output noise; after that, the ANN is trained again with the resultant dataset. Obtained results suggest that best classification outcome is achieved when the cleaning strategies are applied on an ANN output instead of input feature space only. Consequently, the need to consider the classifier's nature when the classical class imbalance approaches are adapted in deep learning and big data scenarios is clear.

No abstract

In this work, plasma-nitrided AISI 316L stainless steel samples were performed by ion nitriding process under pulsed direct current (DC) discharge at different current densities (1 to 2.5 mA/ cm 2). The effect of nitriding current density on the size of crystalline coherently diffracting domains (crystallite size) and strain grade was investigated using X-ray diffraction (XRD) coupled with Williamson-Hall method. Additionally, hardness and wear resistance of the nitriding layer were characterized using a Vickers indenter and pin-on-disk technique respectively. Results showed a decrease in crystallite size from 99 nm for untreated samples to 1.4 nm for samples nitrided at 2.5 mA/cm 2 promoted both: an increase in hardness from 226 HV25g to 1245 HV25g, and a considerably decrease in volume loss by wear effect.

No abstract

Precipitation hardening aluminum alloys are used in many industries due to their excellent mechanical properties, including good weldability. During a welding process, the tensile strength of the joint is critical to appropriately exploit the original properties of the material. The welding processes are still under study, and gas metal arc welding (GMAW) in pulsed metal-transfer configuration is one of the best choices to join these alloys. In this study, the welding of 6061 aluminum alloy by pulsed GMAW was performed under two heat treatment conditions and by using two filler metals, namely: ER 4043 (AlSi 5 ) and ER 4553 (AlMg 5 Cr). A solubilization heat treatment T4 was used to dissolve the precipitates of β"phase into the aluminum matrix from the original T6 heat treatment, leading in the formation of β-phase precipitates instead, which contributes to higher mechanical resistance. As a result, the T4 heat treatment improves the quality of the weld joint and increases the tensile strength in comparison to the T6 condition. The filler metal also plays an important role, and our results indicate that the use of ER 4043 produces stronger joints than ER 4553, but only under specific processing conditions, which include a moderate heat net flux. The latter is explained because Mg, Si and Cu are reported as precursors of the production of β"phase due to heat input from the welding process and the redistribution of both: β" and β precipitates, causes a ductile intergranular fracture near the heat affected zone of the weld joint.