Salesian Journal on Information Systems

2-9

Authors: NEIVA, A. C. B.; LINDEN, R.; GUEDES, V. G.; MASSA, C. L. S.

Maintenance is an important step that must be performed as soon as possible, to minimize the impact of failures. In this paper we discuss the concept of power curve fitting and deviation calculation to detect failures that cause the wind turbines to misperform (the so-called Type I failures). As mathematics and software applications evolution for predictive maintenance is growing fast, simple considerations will be made to enable the development of an advanced individual power curve, so the real time turbine performance analysis may be as accurate as possible. Past environmental and production data is subjected to outlier filtration and statistical analysis, prior to creation of a best fit that would allow for future failures to be detected early. The innovation relies on segregating a power curve for each directional sector and on the correction of the influence of the temperature on the power production, achieving an advanced power curve that really mimics as close as possible the real operational output.

31-50

Authors: LIMA, S. M. L.; SILVA, H. K. L.; LUZ, J. H. S ; et al.

Each second, 8 (eight) new malwares (malicious + software) are launched. Based on this finding, our goal is to propose an antivirus endowed with artificial intelligence, which can identify malware through models based on neural networks of rapid training and high accuracy. The proposed antivirus is equipped with morphological extreme learning machines (mELMs), which are inspired by the image processing theory of Mathematical Morphology. The results shown here are compared with classical approaches and evaluated through widely used classification metrics. On average, the antivirus proposed can distinguish malware from benign applications in 99.80% of cases, with a training time of 9.32 seconds. We found an average ratio between percentage accuracy and training time in reverse order. The proposed antivirus is 96 times better than state-of-the-art Deep Learning.

2-18

Authors: FERREIRA JR, D. A., CABRAL, J. P. O., MACEDO, C. J. A. et al.

The Internet of Things (IoT) is based on objects ("things"), which are able to communicate and transfer data. The fact generates an increase in the flow of data traffic over the network. In order to support this increase, the heterogeneity of devices and the geographic distribution have the need for an infrastructure which can handle this demand. The Institute of Informatics of the Federal University of Goias has proposed a software-defined IoT gateway thats supports for various wireless communication technologies and fog/cloud environment integration. A prototype was deployed for the queue at the University Restaurant using a single gateway IoT, which was afterwards evaluaed with good performance. In this article, the results of the evaluation of the prototype are presented, as well as the expansion of the system in a pilot project at Campus Samambaia (UFG), with the implementation of multiple gateways IoT, aiming to collect data on the amount of organic waste , monitoring of garbage levels in selective bins, temperature, humidity, gas level and video monitoring. The experiments carried out were intended to analyze the working and the capacity of the system to manage the data flow, from the sensors to the user's application.

28-41

Authors: NASCIMENTO, P. S.; KOMATI, K. S.; ANDRADE, J. O.

This paper describes the proposal of an archtecture and the comparison of regression methods for the analysis of personal loan proposals using machine learning algorithms. The case study presented here deals with data from a financial company that has a historical database of credit analyses, formed by the client's profile information, his previous relationship history with the financial company and the loan proposal data. It is important that the proposed architecture achieves the following objectives: decrease the cost associated with accessing credit protection agency's data, improve the accuracy compared to the existing process, allow the IT team keep a control variable over the approval or not of the proposal, and keep current employees' jobs. In order to meet all these requirements, a differential of this proposal is the usage of two stages of analysis, one without the data from credit protection agencies and the second with this type of data (if necessary). Another differential is that regression models will be used, whose results are converted into discrete results via comparison with denial/approval thresholds. In the first step, the result is approved { when it is above a pass threshold, denied { when it is below a fail threshold, or proceed to the second step. The second stage process is similar, except that cases that do not qualify as approved or denied continue for manual analysis. The methodology comprises exploratory analysis and pre-processing of data, calibration, validation and tests in two regression models: Random Forest and Partial Least Squares. The results of the experiments showed that the Random Forest achieved results that were better than both the Partial Least Squares and the existing system in the financial company. For this configuration, the first stage of the classification process is able to classify 86.56% of the proposals without manual intervention, and in the second stage, 4.04% are also classified automatically, in addition to reaching 97% accuracy at the end of the second stage.

56-65

Authors: BEZERRA, W.R., KOCH, F.L., WESTPHALL, C.B.

This work provides the basis to understand and select Cloud Computing models applied for the development of IoT solutions using Low-Power Wide Area Network (LPWAN). Cloud Computing paradigm has transformed how the industry implement solution, through the commoditization of shared IT infrastructures. The advent of massive Internet of Things (IoT) and related workloads brings new challenges to this scenario demanding malleable configurations where the resources are distributed closer to data sources. We introduce an analysis of existing solution architectures, along with an illustrative case from where we derive the lessons, challenges, and opportunities of combining these technologies for a new generation of Cloud-native solutions.