scholarly journals RHOASo: An Early Stop Hyper-Parameter Optimization Algorithm

Mathematics ◽  
2021 ◽  
Vol 9 (18) ◽  
pp. 2334
Author(s):  
Ángel Luis Muñoz Castañeda ◽  
Noemí DeCastro-García ◽  
David Escudero García
Keyword(s):  
Parameter Optimization ◽  
Learning Algorithm ◽  
Optimization Algorithms ◽  
The Other ◽  
Machine Learning Algorithm ◽  
Asymptotic Functions ◽  
Benchmark Datasets ◽  
Tuning Process ◽  
Number Of Iterations ◽  
Conditional Optimization

This work proposes a new algorithm for optimizing hyper-parameters of a machine learning algorithm, RHOASo, based on conditional optimization of concave asymptotic functions. A comparative analysis of the algorithm is presented, giving particular emphasis to two important properties: the capability of the algorithm to work efficiently with a small part of a dataset and to finish the tuning process automatically, that is, without making explicit, by the user, the number of iterations that the algorithm must perform. Statistical analyses over 16 public benchmark datasets comparing the performance of seven hyper-parameter optimization algorithms with RHOASo were carried out. The efficiency of RHOASo presents the positive statistically significant differences concerning the other hyper-parameter optimization algorithms considered in the experiments. Furthermore, it is shown that, on average, the algorithm needs around 70% of the iterations needed by other algorithms to achieve competitive performance. The results show that the algorithm presents significant stability regarding the size of the used dataset partition.

Empirical Comparison of Various Discretization Procedures

1998 ◽  
Vol 12 (07) ◽  
pp. 1017-1032 ◽  
Author(s):  
Petr Berka ◽  
Ivan Bruha
Keyword(s):  
Machine Learning ◽  
Real World ◽  
Learning Algorithm ◽  
Machine Learning Algorithms ◽  
The Other ◽  
Machine Learning Algorithm ◽  
Empirical Comparison ◽  
Numerical Attributes ◽  
Real World Problems ◽  
Discretization Procedure

The genuine symbolic machine learning (ML) algorithms are capable of processing symbolic, categorial data only. However, real-world problems, e.g. in medicine or finance, involve both symbolic and numerical attributes. Therefore, there is an important issue of ML to discretize (categorize) numerical attributes. There exist quite a few discretization procedures in the ML field. This paper describes two newer algorithms for categorization (discretization) of numerical attributes. The first one is implemented in the KEX (Knowledge EXplorer) as its preprocessing procedure. Its idea is to discretize the numerical attributes in such a way that the resulting categorization corresponds to KEX knowledge acquisition algorithm. Since the categorization for KEX is done "off-line" before using the KEX machine learning algorithm, it can be used as a preprocessing step for other machine learning algorithms, too. The other discretization procedure is implemented in CN4, a large extension of the well-known CN2 machine learning algorithm. The range of numerical attributes is divided into intervals that may form a complex generated by the algorithm as a part of the class description. Experimental results show a comparison of performance of KEX and CN4 on some well-known ML databases. To make the comparison more exhibitory, we also used the discretization procedure of the MLC++ library. Other ML algorithms such as ID3 and C4.5 were run under our experiments, too. Then, the results are compared and discussed.

scholarly journals Hybrid task scheduling algorithm for high availability in cloud

2018 ◽  
Vol 7 (2.8) ◽  
pp. 455
Author(s):  
Snehaa K ◽  
Sivasankari S
Keyword(s):  
Service Providers ◽  
Learning Algorithm ◽  
Scheduling Algorithm ◽  
Cloud Service ◽  
High Availability ◽  
The Other ◽  
Machine Learning Algorithm ◽  
Task Scheduling Algorithm ◽  
Cloud Service Providers ◽  
Load Balancer

Today, wherever we go we will find one thing or the other that is running in cloud. It has become an essential component in our everyday life. Many cloud service providers have been extensively working on improving the performance of the service provided. Highavailability in cloud is a place where various cloud service providers are extensively researching. High availability in cloud can vary depending on the scenario it is considered under. In our case we are going to see this in the form of response time for the request to get processed and how we can improve it. We are proposing an algorithm which will ensure the above and also provide an efficient model for analysing the same. We will also be configuring Reinforcement machine learning algorithm for training the load balancer and the server to process the request faster in a short duration of time.

MAITS

2011 ◽  
pp. 189-208 ◽  
Author(s):  
Dharmendra Sharma
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Learning Algorithm ◽  
Computer Forensics ◽  
The Other ◽  
Security Requirements ◽  
Machine Learning Algorithm ◽  
Agent Based ◽  
Authentication And Authorization ◽  
Multi Agent

In this chapter, we propose a multi-agent-based information technology (IT) security approach (MAITS) as a holistic solution to the increasing needs of securing computer systems. Each specialist task for security requirements is modeled as a specialist agent. MAITS has five groups of working agents—administration assistant agents, authentication and authorization agents, system log *monitoring agents, intrusion detection agents, and pre-mortem-based computer forensics agents. An assessment center, which is comprised of yet another special group of agents, plays a key role in coordinating the interaction of the other agents. Each agent has an agent engine of an appropriate machine-learning algorithm. The engine enables the agent with learning, reasoning, and decision-making abilities. Each agent also has an agent interface, through which the agent interacts with other agents and also the environment.

scholarly journals Deep Multiphysics and Particle–Neuron Duality: A Computational Framework Coupling (Discrete) Multiphysics and Deep Learning

2019 ◽  
Vol 9 (24) ◽  
pp. 5369
Author(s):  
Alessio Alexiadis
Keyword(s):  
Machine Learning ◽  
First Principles ◽  
Learning Algorithm ◽  
The Other ◽  
Particle Methods ◽  
Machine Learning Algorithm ◽  
Computational Framework ◽  
In Series ◽  
Mathematical Algorithms ◽  
Modelling Techniques

There are two common ways of coupling first-principles modelling and machine learning. In one case, data are transferred from the machine-learning algorithm to the first-principles model; in the other, from the first-principles model to the machine-learning algorithm. In both cases, the coupling is in series: the two components remain distinct, and data generated by one model are subsequently fed into the other. Several modelling problems, however, require in-parallel coupling, where the first-principle model and the machine-learning algorithm work together at the same time rather than one after the other. This study introduces deep multiphysics; a computational framework that couples first-principles modelling and machine learning in parallel rather than in series. Deep multiphysics works with particle-based first-principles modelling techniques. It is shown that the mathematical algorithms behind several particle methods and artificial neural networks are similar to the point that can be unified under the notion of particle–neuron duality. This study explains in detail the particle–neuron duality and how deep multiphysics works both theoretically and in practice. A case study, the design of a microfluidic device for separating cell populations with different levels of stiffness, is discussed to achieve this aim.

scholarly journals Negation Detection on Mexican Spanish Tweets: The T-MexNeg Corpus

2021 ◽  
Vol 11 (9) ◽  
pp. 3880
Author(s):  
Gemma Bel-Enguix ◽  
Helena Gómez-Adorno ◽  
Alejandro Pimentel ◽  
Sergio-Luis Ojeda-Trueba ◽  
Brian Aguilar-Vizuet
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Random Fields ◽  
Conditional Random Fields ◽  
Learning Algorithm ◽  
Detection Task ◽  
The Other ◽  
Machine Learning Algorithm ◽  
Mexican Spanish ◽  
Negation Detection

In this paper, we introduce the T-MexNeg corpus of Tweets written in Mexican Spanish. It consists of 13,704 Tweets, of which 4895 contain negation structures. We performed an analysis of negation statements embedded in the language employed on social media. This research paper aims to present the annotation guidelines along with a novel resource targeted at the negation detection task. The corpus was manually annotated with labels of negation cue, scope, and, event. We report the analysis of the inter-annotator agreement for all the components of the negation structure. This resource is freely available. Furthermore, we performed various experiments to automatically identify negation using the T-MexNeg corpus and the SFU ReviewSP-NEG for training a machine learning algorithm. By comparing two different methodologies, one based on a dictionary and the other based on the Conditional Random Fields algorithm, we found that the results of negation identification on Twitter are lower when the model is trained on the SFU ReviewSP-NEG Corpus. Therefore, this paper shows the importance of having resources built specifically to deal with social media language.

scholarly journals BlindMobi: A system for bus identification, based on Bluetooth Low Energy, for people with visual impairment

2019 ◽  
Author(s):  
Hilson Andrade ◽  
David Borges ◽  
Leandro Bernardes ◽  
João Albuquerque ◽  
Abel Guilhermino Silva-Filho
Keyword(s):  
Mobile Device ◽  
Learning Algorithm ◽  
Detection System ◽  
Low Energy ◽  
The Other ◽  
Prototype System ◽  
Machine Learning Algorithm ◽  
Blind People ◽  
Urban Centers ◽  
Bluetooth Low Energy

This paper presents a bus detection system based on Bluetooth low energy (BLE) technology that aims to ease the traveling of blind people in large urban centers. The proposed system consists of two subsystems: one embedded hardware on the buses and the other running on the user mobile device. From the device embedded on the bus, BLE beacons containing itinerary information and bus acceleration are sent and read by the application running on the mobile device. Through this information and using the machine learning algorithm, the application is able to identify the approach and arrival of the bus, guiding the user. A complete prototype system has been constructed and tested to validate the proposed system, where a rating rate of 91.5% was obtained, indicating the feasibility of the proposal.

scholarly journals Improving the Chemical Selectivity of an Electronic Nose to TNT, DNT and RDX Using Machine Learning

Sensors ◽  
2019 ◽  
Vol 19 (23) ◽  
pp. 5207 ◽  
Author(s):  
Anton Gradišek ◽  
Marion van Midden ◽  
Matija Koterle ◽  
Vid Prezelj ◽  
Drago Strle ◽  
...  
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
The Other ◽  
Classification Model ◽  
Machine Learning Algorithm ◽  
Classification Models ◽  
Functionalized Surfaces ◽  
Other Substances ◽  
Chemical Selectivity ◽  
Excellent Accuracy

We used a 16-channel e-nose demonstrator based on micro-capacitive sensors with functionalized surfaces to measure the response of 30 different sensors to the vapours from 11 different substances, including the explosives 1,3,5-trinitro-1,3,5-triazinane (RDX), 1-methyl-2,4-dinitrobenzene (DNT) and 2-methyl-1,3,5-trinitrobenzene (TNT). A classification model was developed using the Random Forest machine-learning algorithm and trained the models on a set of signals, where the concentration and flow of a selected single vapour were varied independently. It is demonstrated that our classification models are successful in recognizing the signal pattern of different sets of substances. An excellent accuracy of 96% was achieved for identifying the explosives from among the other substances. These experiments clearly demonstrate that the silane monolayers used in our sensors as receptor layers are particularly well suited to selecting and recognizing TNT and similar types of explosives from among other substances.

scholarly journals Emotion Detection Using Deep Learning Algorithm

2021 ◽  
Vol 11 (4) ◽  
pp. 30-38
Author(s):  
Shital Sanjay Yadav ◽  
Anup S. Vibhute
Keyword(s):  
Neural Network ◽  
Learning Algorithm ◽  
Input Image ◽  
The Other ◽  
Emotion Detection ◽  
Image Size ◽  
Deep Learning Algorithm ◽  
Prime Task ◽  
Japanese Female ◽  
Benchmark Datasets

Automatic emotion detection is a prime task in computerized human behaviour analysis. The proposed system is an automatic emotion detection using convolution neural network. The proposed end-to-end CNN is therefore named as ENet. Keeping in mind the computational efficiency, the deep network makes use of trained weight parameters of the MobileNet to initialize the weight parameters of ENet. On top of the last convolution layer of ENet, the authors place global average pooling layer to make it independent of the input image size. The ENet is validated for emotion detection using two benchmark datasets: Cohn-Kanade+ (CK+) and Japanese female facial expression (JAFFE). The experimental results show that the proposed ENet outperforms the other existing methods for emotion detection.

scholarly journals Intelligent and Smart Irrigation System Using Edge Computing and IoT

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-16
Author(s):  
M. Safdar Munir ◽  
Imran Sarwar Bajwa ◽  
Amna Ashraf ◽  
Waheed Anwar ◽  
Rubina Rashid
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Irrigation System ◽  
The Other ◽  
Sensor Data ◽  
Final Decision ◽  
Machine Learning Algorithm ◽  
Android Application ◽  
Water Requirements ◽  
Sweet Water

Smart parsimonious and economical ways of irrigation have build up to fulfill the sweet water requirements for the habitants of this world. In other words, water consumption should be frugal enough to save restricted sweet water resources. The major portion of water was wasted due to incompetent ways of irrigation. We utilized a smart approach professionally capable of using ontology to make 50% of the decision, and the other 50% of the decision relies on the sensor data values. The decision from the ontology and the sensor values collectively become the source of the final decision which is the result of a machine learning algorithm (KNN). Moreover, an edge server is introduced between the main IoT server and the GSM module. This method will not only avoid the overburden of the IoT server for data processing but also reduce the latency rate. This approach connects Internet of Things with a network of sensors to resourcefully trace all the data, analyze the data at the edge server, transfer only some particular data to the main IoT server to predict the watering requirements for a field of crops, and display the result by using an android application edge.

Sign in / Sign up

Export Citation Format

Share Document