A bat-neural network multi-agent system (BNNMAS) for stock price prediction: Case study of DAX stock price

Creating an intelligent system that can accurately predict stock price in a robust way has always been a subject of great interest for many investors and financial analysts. Predicting future trends of financial markets is more remarkable these days especially after the recent global financial crisis. So traders who access to a powerful engine for extracting helpful information throw raw data can meet the success. In this paper we propose a new intelligent model in a multi-agent framework called bat-neural network multi-agent system (BNNMAS) to predict stock price. The model performs in a four layer multi-agent framework to predict eight years of DAX stock price in quarterly periods. The capability of BNNMAS is evaluated by applying both on fundamental and technical DAX stock price data and comparing the outcomes with the results of other methods such as genetic algorithm neural network (GANN) and some standard models like generalized regression neural network (GRNN), etc. The model tested for predicting DAX stock price a period of time that global financial crisis was faced to economics. The results show that BNNMAS significantly performs accurate and reliable, so it can be considered as a suitable tool for predicting stock price specially in a long term periods.     

Light gradient boosting machine with optimized hyperparameters for identification of malicious access in IoT network

In this paper, an advanced and optimized Light Gradient Boosting Machine (LGBM) technique is proposed to identify the intrusive activities in the Internet of Things (IoT) network. The followings are the major contributions: i) An optimized LGBM model has been developed for the identification of malicious IoT activities in the IoT network; ii) An efficient evolutionary optimization approach has been adopted for finding the optimal set of hyper-parameters of LGBM for the projected problem. Here, a Genetic Algorithm (GA) with k-way tournament selection and uniform crossover operation is used for efficient exploration of hyper-parameter search space; iii) Finally, the performance of the proposed model is evaluated using state-of-the-art ensemble learning and machine learning-based model to achieve overall generalized performance and efficiency. Simulation outcomes reveal that the proposed approach is superior to other considered methods and proves to be a robust approach to intrusion detection in an IoT environment.     

Extreme learning machine and bayesian optimization-driven intelligent framework for IoMT cyber-attack detection

The Journal of Supercomputing - The Internet of Medical Things (IoMT) is a bionetwork of allied medical devices, sensors, wearable biosensor devices, etc. It is gradually reforming the healthcare...     

Optimization of process variables in drilling of carbon fiber reinforced plastics using various multi criteria decision making approaches

Carbon fiber-reinforced polymers are one of the lightweight materials used in structural design due to their exceptional mechanical performances. The drilling operation is indispensable as it facilitates the assembling of various manufactured components. However, drilling of fibrous laminates is deemed difficult in comparison to the traditional metals because of the anisotropic and non-homogeneous nature. The present work addresses the parametric effect on the drilled hole delamination and further reduces it with an optimal combination of parameters for multi-objectives using different multi-criterion decision-making techniques. Initially, the response surface-based regression model of delamination as a function of three static inputs has been developed, further revised with induced thrust as well as mean torque for the improvisation of the prediction capability. Finally, for the overall improvement, a decision-making model has been used that includes grey relation analysis, technique for order performance by similarity to ideal solution, and VIšekriterijumsko Kompromisno Rangiranje method. The delamination was found to be minimum at a low drill point angle (100°), high spindle rotation (2150 min⁻¹ ), and low feed rate (0.025 mm/rev) due to reduced thrust force. The mean absolute prediction error was significantly improved considering root mean square torque rather than axial thrust with process variables.     

Transducer invariant multi-class fault classification in a rotor-bearing system using support vector machines

Faults in a rotor-bearing system due to bearings and unbalance have been classified using support vector machines (SVMs). Vibration signals on a rotor-bearing system were measured simultaneously at five different rotating speeds using seven transducers. The most sensitive feature of the vibration signals has been determined using compensation distance evaluation technique. Multi-class SVMs classification algorithm was then implemented for classification of the faults by considering SVMs created by the possible combinations of the most two sensitive features for each type of fault. By using optimal SVM parameters, the effective location of transducer among seven transducers for best classification of the faults has been investigated and found that any transducer provides a classification of 75% or better and this classification rate increases when more transducers are considered. This paper provides a robust SVM based technique using only time domain data without any additional preprocessing for classifying bearing and unbalance faults.     

Smart SCADA system for urban air pollution monitoring

There are a number of different types of monitoring stations for checking air quality levels in urban environments. These monitoring stations usually just perform data acquisition of the measured values from the sensors and store them in the database. The processing of the measured results as well as the statistical analysis is mainly done in other places where the data come from various communication systems. Acquisition of the measured data is commonly done on-line while the processing and statistical analysis is performed off-line. As opposed to these measurement systems, this implemented device enables the acquisition and statistical processing of the measured data in real time and the results are instantly available to all users. The system indicates the air pollutants using the ARMA model. The transmission of information in the realized smart SCADA system is done by the Modbus protocol using shared variables which gives the whole system a stronger hierarchical structure.     

Application of Taguchi coupled Fuzzy Multi Attribute Decision Making (FMADM) for optimizing surface quality in turning austenitic and duplex stainless steels

In the present study, Taguchi approach is coupled with fuzzy-multiple attribute decision making methods for achieving better surface quality in constant cutting speed face turning of EN 1.4404 austenitic, EN 1.4462 standard duplex and EN 1.4410 super duplex stainless steels. Two typical multiple attribute decision making techniques were simultaneously adopted to determine multi-surface quality characteristics indices. The differences in rankings among derived indices are solved through converting each crisp values into trapezoidal fuzzy number and unifying them using fuzzy simple additive weight method. The fuzzy numbers are then deffuzified into crisp values employing techniques like; the spread, mode and area between centroid of centroids. Through this procedure, the decision maker is provided with necessary decision tools to optimize the cutting conditions with less sensitivity to the change of weights and no difference in ranking among the deffuzification techniques. Additionally, results of analyses of means and the validation experiments confirm that the optimum cutting conditions derived by this method produce far better surface finish than the best finish obtained during the course of experimentation. Analyses of variance results have shown the predominant effect of feed rate on surface quality. Finally, the collected chip at constant cutting speed and varying feed rates and depth of cuts has shown that friendlier-to-machine chips are obtained when machining austenitic stainless steels than duplex stainless steel grades.     

RF fingerprint measurements for the identification of devices in wireless communication networks based on feature reduction and subspace transformation

This document proposes a radiofrequency (RF) fingerprinting strategy for the proper identification of wireless devices in mobile and wireless networks. The proposed identification methods are based on the extraction of the preamble RF fingerprint of a device and its comparison with a set of already known device RF fingerprints. The identification method combines techniques for feature reduction such as Principal Component Analysis (PCA) and Partial Least Squares regression (PLS), both based on subspace transformation, along with a similarity-based analysis. In this work, a complete procedure for RF fingerprint data extraction and analysis is provided. In addition, some experimentation with commercial Wi-Fi devices is carried out for the methods validation.     

Nonlinear adaptive noise-induced algorithm and its application in penetration signal

A nonlinear adaptive noise induced algorithm with nonlinear weights was proposed to extract rigid body deceleration during penetration events; it has 3rd-order nonlinear weight, which ensures deceleration curve is smooth everywhere (not only continuous) and avoids sharp points (crucial for targets detection). In addition, an autocorrelation algorithm was improved by applying moving window method to be compared with the proposed nonlinear adaptive algorithm. By calculating penetration depth and Power Spectrum Density (PSD) of 4 deceleration time series, we show that the nonlinear adaptive algorithm more effectively reduces noise in deceleration for striking velocities between 538 and 800 m/s compared with Adaptive Paйta Criterion, moving window autocorrelation and wavelet algorithms. It is further shown that the proposed adaptive algorithm is of the same order as the other 3 methods in terms of computational complexity.