The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The basic features of some of the most versatile and popular open source frameworks for machine learning (TensorFlow, Deep Learning4j, and H2O) are considered and compared. Their comparative analysis was performed and conclusions were made as to the advantages and disadvantages of these platforms. The performance tests for the de facto standard MNIST data set were carried out on H2O framework for...
The hyper-parameter optimization of machine learning model is not a completely solved problem. The exquisite combination of artificial tuning and grid search may be a good choice in the area where the dimension of hyper-parameters is very low. But for high-dimensional hyper-parameter optimization problems, artificial tuning and grid search are obviously helpless. In this paper, we propose a quantum...
Restricted Boltzmann Machine (RBM) is the building block of Deep Belief Nets and other deep learning tools. Fast learning and prediction are both essential for practical usage of RBM-based machine learning techniques. This paper presents a concept named generalized redundancy elimination to avoid most of the the computations required in RBM learning and prediction without changing the results. It...
Ultrasonic NDE uses high frequency acoustic waves to evaluate materials, and often signal processing is required to detect echoes from defects in the presence of microstructure scattering noise. Scattering noise, also known as clutter, interferes with the flaw signal and cannot be completely eliminated by using classical signal processing methods such as band-pass filtering. In this paper, neural...
This article aims to present a strategy to properly distribute the graduates to complete a fourth year of after-Lyceum apprenticeship in the specialized field of Vocational Education and Training (VET) as it is implemented in Greece. The theoretical solution of the issue is presented, as well as the terms and conditions that should be met in order to create an on-line platform capable of organizing...
The problems of organization of engineering education on the basis of the competence approach are considered. It is argued that the ideas and principles of this approach only partially meet the needs of training an engineer in the conditions of the modern development of science, technique, technology, production and it manifests itself in a whole series of contradictions arising when projecting a...
The PID control algorithm is the most used industrial control method owing to its simplicity and ease of use. However, tuning PID parameters is not trivial and many methods have been reported in literature. This paper seeks to show a machine learning approach using multivariate regression with gradient descent and the normal equation. The first order cruise control system is used as an example and...
This work presents an embedded hardware architecture for real-time ultrasonic NDE applications that incorporate Hidden Markov Model (HMM) based statistical signal methods. Proposed algorithm is a combination of Discrete Wavelet Transform (DWT) for pre-processing A-scan signals and HMM for classification of the flaw presence. For this study, a MicroZed FPGA with Xilinx Zynq-7020 System-on-Chip (SoC)...
Fingerprints are one of the most popular biometrics traits. Feature extraction and vector creation are crucial in fingerprint-matching algorithms. For increasing the confidence of fingerprint recognition, different feature vector forms are considered in literature. In this paper, we introduce a complete (fully-implemented) algorithm for fingerprint recognition. The work describes image preprocessing...
A technique and algorithms for early detection of the started attack and subsequent blocking of malicious traffic are proposed. The primary separation of mixed traffic into trustworthy and malicious traffic was carried out using cluster analysis. Classification of newly arrived requests was done using different classifiers with the help of received training samples and developed success criteria.
Image classification domain has been an area which has attracted a lot of researchers over past years. Many classification methodologies for spatial image datasets has been developed. Artificial intelligence based approaches are getting popular now a days for getting the image classification task done in more efficient and correct way. The prime goal is to develop a classification mechanism which...
In traditional text sentiment analysis methods, text feature vector has the problem of high dimensionality and high sparseness. In view of this situation, we can cluster the similar words together and use the generated clusters to fit into a new dimension so that the text feature vector dimension will be decreased. By using Word2Vec tool and K-means clustering algorithm, this task can be completed...
In this paper, Self-adaptive Differential Evolutionary Extreme Learning Machine (SaDE-ELM) was proposed as a new class of learning algorithm for single-hidden layer feed forward neural network (SLFN). In order to achieve good generalization performance, SaDE-ELM calculates the error on a subset of testing data for parameter optimization. Since SaDE-ELM employs extra data for validation to avoid the...
Authors can be differentiated by their styles of writing. In this paper, we propose features which attempt to classify authors based on their writing styles. The features can be usage of parts of speech, punctuation marks, word lengths, sentence lengths, number of unique words used, etc. This concept is used in many fields like email classification, fraud detection, etc. We propose a module to extract...
Graduate employability is an increasingly major concern for academic institutions and assessing student employability provides a way of linking student skills and employer business requirements. Enhancing student assessment methods for employability can improve their understanding about companies in order to get suitable company for them. So, enhanced employability prediction of student can help them...
Tagging provides a convenient means to assign tokens of identification to research papers which facilitate recommendation, search and disposition process of research papers. This paper contributes a document centered approach for auto-tagging of research papers. The auto-tagging method mainly comprises of two processes:- classification and tag selection. The classification process involves automatic...
Security is obligatory for digital world. It requires robust and reliable security mechanisms which comprises irreplaceable identification of individual. Biometrics plays an important role in recognizing individual uniquely, furthermore iris based security is more impenetrable as compared to fingerprint based security. Also, human iris doesn't change with ageing and can be easily captured. Generic...
This paper focuses on the problem of machine learning classifier choice for network intrusion detection, taking into consideration several ensemble classifiers from the supervised learning category. We have evaluated Bagged trees, AdaBoost, RUSBoost, LogitBoost and GentleBoost algorithms, provided an analysis of the performance of the classifiers and compared their learning capabilities, taking for...
With the abundance of mobile connected devices and coexisting networks, localization solutions are inevitably subject to interference. In this paper, effects of interference on the performance of Fingerprinting Localization Algorithms (FPS) are studied both theoretically and through experimentation. The previously introduced theoretical framework based on Hypothesis Testing (HT) problem is employed...
The design of effective financial early warning algorithm is of great significance to the financial management of the company. The weak classification algorithm can be improved to a high classification algorithm with high recognition rate through the ensemble learning. The algorithm can overcome the drawback of low classification accuracy of single classifier. Therefore, this paper combines decision...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.