Determining the Number of Neurons in Artificial Neural Networks for Approximation, Trained with Algorithms Using the Jacobi Matrix

How can we determine the optimal number of neurons when constructing an artificial neural network? This is one of the most frequently asked questions when working with this type of artificial intelligence. Experience has brought the understanding that it takes an individual approach for each task to specify the number of neurons. Our method is based on the requirement of algorithms looking for a minimum of functions of type 𝑺􁈺𝒛􁈻 􀵌 Σ 􁈾𝝋𝒊 𝒎 􁈺𝒛 􁈻􁈿𝟐 𝒊􀭀𝟏 that satisfy the inequality 𝒑 􀵑 𝒎, where p is the dimensionality of the argument z, and m is the number of functions. Formulas for an upper limit of the required neurons are proposed for networks with one hidden layer and for networks with r hidden layers with an equal number of neurons.

Download Full-text

New Method for Locker Door with Face Security System Using Backpropagation Algorithm

International Journal of Engineering & Technology ◽

10.14419/ijet.v7i2.13.16931 ◽

2018 ◽

Vol 7 (2.13) ◽

pp. 402

Author(s):

Y Yusmartato ◽

Zulkarnain Lubis ◽

Solly Arza ◽

Zulfadli Pelawi ◽

A Armansah ◽

...

Keyword(s):

Neural Network ◽

Neural Networks ◽

Pattern Recognition ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Human Brain ◽

Back Propagation ◽

The Face ◽

Artificial Neural ◽

Hidden Layer

Lockers are one of the facilities that people use to store stuff. Artificial neural networks are computational systems where architecture and operations are inspired by the knowledge of biological neurons in the brain, which is one of the artificial representations of the human brain that always tries to stimulate the learning process of the human brain. One of the utilization of artificial neural network is for pattern recognition. The face of a person must be different but sometimes has a shape similar to the face of others, because the facial pattern is a good pattern to try to be recognized by using artificial neural networks. Pattern recognition on artificial neural network can be done by back propagation method. Back propagation method consists of input layer, hidden layer and output layer.

Download Full-text

Sustainability through the Use of Modern Simulation Methods—Applied Artificial Intelligence

Sustainability ◽

10.3390/su11082384 ◽

2019 ◽

Vol 11 (8) ◽

pp. 2384 ◽

Cited By ~ 1

Author(s):

Constantin Ilie ◽

Catalin Ploae ◽

Lucia Violeta Melnic ◽

Mirela Rodica Cotrumba ◽

Andrei Marian Gurau ◽

...

Keyword(s):

Neural Network ◽

Artificial Intelligence ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Simulation Methods ◽

Cognitive Response ◽

European Economic ◽

Artificial Neural ◽

Training Error

As the transformative power of AI crosses all economic and social sectors, the use of it as a modern technique for the simulation and/or forecast of various indicators must be viewed as a tool for sustainable development. The present paper reveals the results of research on modeling and simulating the influences of four economic indicators (the production in industry, the intramural research and development expenditure, the turnover and volume of sales and employment) on the evolution of European Economic Sentiment using artificial intelligence. The main goal of the research was to build, train and validate an artificial neural network that is able to forecast the following year’s value of economic sentiment using the present values of the other indicators. Research on predicting European Economic Sentiment Indicator (ESI) using artificial neural networks is a starting point, with work on this subject almost inexistent, the reason being mainly that ESI is a composite of five sectoral confidence indicators and is not thought to be an emotional response to the interaction of the entrepreneurial population with different economic indicators. The authors investigated, without involving a direct mathematical interaction among the indicators involved, predicting ESI based on a cognitive response. Considering the aim of the research, the method used was simulation with an artificial neural network and a feedforward network (structure 4-9-6-1) and a backward propagation instruction algorithm was built. The data used are euro area values (for 19 countries only—EA19) recorded between 1999 and 2016, with Eurostat as the European Commission’s statistical data website. To validate the results, the authors imposed the following targets: the result of the neural network training error is less than 5% and the prediction verification error is less than 10%. The research outcomes resulted in a training error (after 30,878 iterations) of less than 0.099% and a predictive check error of 2.02%, which resulted in the conclusion of accurate training and an efficient prediction. AI and artificial neural networks, are modeling and simulation methods that can yield results of nonlinear problems that cover, for example, human decisions based on human cognitive processes as a result of previous experiences. ANN copies the structure and functioning of the biological brain, having the advantage through learning and coaching processes (biological cognitive), to copy/predict the results of the thinking process and, thus, the process of choice by the biological brain. The importance of the present paper and its results stems from the authors’ desire to use and popularize modern methods of predicting the different macroeconomic indices that influence the behavior of entrepreneurs and therefore the decisions of these entrepreneurs based on cognitive response more than considering linear mathematical functions that cannot correctly understand and anticipate financial crises or economic convulsions. Using methods such as AI, we can anticipate micro- and macroeconomic developments, and therefore react in the direction of diminishing their negative effects for companies as well as the national economy or European economy.

Download Full-text

DEVELOPMENT AND MODELING OF AN ARTIFICIAL NEURAL NETWORK FOR THE AI-DRIVER IN THE COMPOSITION OF AN UNMANNED VEHICLE

Bulletin of Belgorod State Technological University named after V G Shukhov ◽

10.34031/2071-7318-2020-5-8-101-110 ◽

2020 ◽

Vol 5 (8) ◽

pp. 101-110

Author(s):

A. Vlasov ◽

T. Kruglova

Keyword(s):

Neural Network ◽

Artificial Intelligence ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Control Systems ◽

Unmanned Vehicles ◽

Unmanned Vehicle ◽

Artificial Neural ◽

Training Error

Improving control systems for unmanned vehicles is the most urgent task in robotics. The use of such a tool as artificial neural networks can solve problems with intelligent and adaptive control. The existing concept of AI driver (driver with artificial intelligence) implies a system capable of controlling the speed and position of an unmanned vehicle in space. This article proposes a method for developing an artificial neural network for an AI-driver, taking into account the appearance of obstacles in the path of an unmanned vehicle, compiling an empirical database for training, and modeling the developed system to obtain both a control signal and a trajectory. The proposed system consists of two artificial neural networks that divide the task of driving an unmanned vehicle into two sub-tasks: processing data from rangefinders and generating a speed setting signal for the left and right drives. This approach reduces the retraining of the neural network and allows you to get a smaller training error. The use of artificial intelligence will make it possible to increase the functionality and reliability of control systems for unmanned vehicles.

Download Full-text

Jaringan Syaraf Tiruan Untuk Pengenalan Citra Sidik Jari Pada Smart Home Berbasis MYSQL

Jurnal Teknologi Informasi dan Ilmu Komputer ◽

10.25126/jtiik.2018551032 ◽

2018 ◽

Vol 5 (5) ◽

pp. 597

Author(s):

Nur Yanti ◽

Fathur Zaini Rachman ◽

Nurwahidah Jamal ◽

Era Purwanto ◽

Fachrurozy Fachrurozy

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Data Base ◽

Monitoring System ◽

Smart Home ◽

Fingerprint Image ◽

Artificial Neural ◽

Hidden Layer

Sistem keamanan yang bertujuan sebagai sistem monitoring pada smart home seperti memonitoring pengguna laboratorium, perpustakaan, atau ruangan penyimpanan dan peminjaman peralatan praktek di program studi suatu kampus, ruang penyimpanan senjata, hingga rumah tinggal, memerlukan sekuritas yang handal untuk memudahkan identifikasi pengguna ruangan atau pencegahan dari tindak pencurian, maka dirancang sistem monitoring melalui pengenalan citra sidik jari menggunakan sensor ZFM60, jaringan syaraf tiruan dan MySQL. Tujuannya agar di dapat pola yang relevan dari citra dan mengeliminasi informasi atau variabel yang tidak relevan. Metode yang digunakan yaitu experimental, terdiri dari pengumpulan data sidik jari, perancangan sistem pengolahan citra, pembuatan dan pengujian hardware dan software, serta implementasi sistem. Hasil proses pengenalan atau klarifikasi citra sidik jari melalui GUI Matlab, nilai error hasil pengolahan dan pelatihan citra sidik jari dengan jaringan syaraf tiruan, digunakan sebagai ciri citra dan disimpan sebagai data base pada MySQL, kemudian dibandingkan dengan nilai error citra sidik jari baru yang di klarifikasi. Nilai citra yang dapat dikenali berada diantara -0,0005 hingga 0,0005, diluar batas tersebut merupakan citra yang tidak dikenali. Selisih (nilai error) antara ciri citra yang tersimpan pada data base dan ciri citra yang diklarifikasi menghasilkan nilai error yang kecil yaitu < 0.0005, menunjukkan jaringan syaraf tiruan backpropagation handal diimplementasikan pada pengenalan sidik jari untuk melatih pola citra dari sidik jari. Konfigurasi jaringan yaitu maksimal epoch = 3000, learning rate = 1, target error = 0.1, hidden layer = 17. Pelatihan jaringan syaraf tiruan pada konfigurasi tersebut menghasilkan nilai error terkecil dari ciri citra sebesar 0.0000085. Abstract The security system that aims as a monitoring system in smart home such as monitoring laboratory users, libraries, or storage rooms and borrowing practical equipment in the study program of a campus, weapons storage room, to a residence, requires reliable securities to facilitate identification of room users or prevention from theft, it is designed a monitoring system through fingerprint image recognition using ZFM60 sensors, artificial neural networks and MySQL. The goal is to get relevant patterns from the image and eliminate irrelevant information or variables. The method used is experimental, consisting of fingerprint data collection, image processing system design, hardware and software manufacturing and testing, and system implementation. The result of the process of recognition or clarification of fingerprint images through the Matlab GUI, the error value of processing and training of fingerprint images with artificial neural networks, is used as a feature of the image and stored as a data base on MySQL, then compared with the error value of the new fingerprint image that is clarified. The recognizable image value is between -0,0005 to 0,0005, beyond this limit is an unrecognized image. The difference (error value) between the characteristics of the image stored in the data base and the clarified image feature produces a small error value of <0.0005, indicating a reliable backpropagation artificial neural network is implemented in fingerprint recognition to train the image pattern of fingerprints. Network configuration is maximum epoch = 3000, learning rate = 1, target error = 0.1, hidden layer = 17. Artificial neural network training in the configuration produces the smallest error value of the image characteristics of 0.0000085.

Download Full-text

Optimal Artificial Neural Network Type Selection Method for Usage in Smart House Systems

Sensors ◽

10.3390/s21010047 ◽

2020 ◽

Vol 21 (1) ◽

pp. 47

Author(s):

Vasyl Teslyuk ◽

Artem Kazarian ◽

Natalia Kryvinska ◽

Ivan Tsmots

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Input Data ◽

Optimization Criterion ◽

Fuzzy Input ◽

Different Types ◽

Smart House ◽

Artificial Neural

In the process of the “smart” house systems work, there is a need to process fuzzy input data. The models based on the artificial neural networks are used to process fuzzy input data from the sensors. However, each artificial neural network has a certain advantage and, with a different accuracy, allows one to process different types of data and generate control signals. To solve this problem, a method of choosing the optimal type of artificial neural network has been proposed. It is based on solving an optimization problem, where the optimization criterion is an error of a certain type of artificial neural network determined to control the corresponding subsystem of a “smart” house. In the process of learning different types of artificial neural networks, the same historical input data are used. The research presents the dependencies between the types of neural networks, the number of inner layers of the artificial neural network, the number of neurons on each inner layer, the error of the settings parameters calculation of the relative expected results.

Download Full-text

Optimization Artificial Neural Network Using Artificial Bee Colony in Letter Recognition Classification

JELIKU (Jurnal Elektronik Ilmu Komputer Udayana) ◽

10.24843/jlk.2020.v08.i04.p13 ◽

2020 ◽

Vol 8 (4) ◽

pp. 469

Author(s):

I Gusti Ngurah Alit Indrawan ◽

I Made Widiartha

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Networks ◽

Classification Accuracy ◽

Artificial Bee Colony Algorithm ◽

Artificial Bee Colony ◽

Letter Recognition ◽

Bee Colony ◽

Artificial Neural ◽

Hidden Layer

Artificial Neural Networks or commonly abbreviated as ANN is one branch of science from the field of artificial intelligence which is often used to solve various problems in fields that involve grouping and pattern recognition. This research aims to classify Letter Recognition datasets using Artificial Neural Networks which are weighted optimally using the Artificial Bee Colony algorithm. The best classification accuracy results from this study were 92.85% using a combination of 4 hidden layers with each hidden layer containing 10 neurons.

Download Full-text

FEATURES OF DIAGNOSTIC ARTIFICIAL NEURAL NETWORKS FOR HYBRID EXPERT SYSTEMS

Digital Technologies ◽

10.33243/2313-7010-26-36-46 ◽

2019 ◽

Vol 26 ◽

pp. 36-46

Author(s):

S. KONOVALOV ◽

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Expert Systems ◽

Technical System ◽

Bulletin Board ◽

Confidence Factor ◽

Complex Technical System ◽

Artificial Neural

In the proposed article, various methods of constructing an artificial neural network as one of the components of a hybrid expert system for diagnosis were investigated. A review of foreign literature in recent years was conducted, where hybrid expert systems were considered as an integral part of complex technical systems in the field of security. The advantages and disadvantages of artificial neural networks are listed, and the main problems in creating hybrid expert systems for diagnostics are indicated, proving the relevance of further development of artificial neural networks for hybrid expert systems. The approaches to the analysis of natural language sentences, which are used for the work of hybrid expert systems with artificial neural networks, are considered. A bulletin board is shown, its structure and principle of operation are described. The structure of the bulletin board is divided into levels and sublevels. At sublevels, a confidence factor is applied. The dependence of the values of the confidence factor on the fulfillment of a particular condition is shown. The links between the levels and sublevels of the bulletin board are also described. As an artificial neural network architecture, the «key-threshold» model is used, the rule of neuron operation is shown. In addition, an artificial neural network has the property of training, based on the application of the penalty property, which is able to calculate depending on the accident situation. The behavior of a complex technical system, as well as its faulty states, are modeled using a model that describes the structure and behavior of a given system. To optimize the data of a complex technical system, an evolutionary algorithm is used to minimize the objective function. Solutions to the optimization problem consist of Pareto solution vectors. Optimization and training tasks are solved by using the Hopfield network. In general, a hybrid expert system is described using semantic networks, which consist of vertices and edges. The reference model of a complex technical system is stored in the knowledge base and updated during the acquisition of new knowledge. In an emergency, or about its premise, with the help of neural networks, a search is made for the cause and the control action necessary to eliminate the accident. The considered approaches, interacting with each other, can improve the operation of diagnostic artificial neural networks in the case of emergency management, showing more accurate data in a short time. In addition, the use of such a network for analyzing the state of health, as well as forecasting based on diagnostic data using the example of a complex technical system, is presented.

Download Full-text

An artificial neural network identifies glyphosate-impacted brackish communities based on 16S rRNA amplicon MiSeq read counts

10.1101/711309 ◽

2019 ◽

Author(s):

René Janßen ◽

Jakob Zabel ◽

Uwe von Lukas ◽

Matthias Labrenz

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Environmental Monitoring ◽

Microbial Communities ◽

List Type ◽

Neural Net ◽

Next Generation ◽

Artificial Neural

AbstractArtificial neural networks can be trained on complex data sets to detect, predict, or model specific aspects. Aim of this study was to train an artificial neural network to support environmental monitoring efforts in case of a contamination event by detecting induced changes towards the microbial communities. The neural net was trained on taxonomic cluster count tables obtained via next-generation amplicon sequencing of water column samples originating from a lab microcosm incubation experiment conducted over 140 days to determine the effects of the herbicide glyphosate on succession within brackish-water microbial communities. Glyphosate-treated assemblages were classified correctly; a subsetting approach identified the clusters primarily responsible for this, permitting the reduction of input features. This study demonstrates the potential of artificial neural networks to predict indicator species in cases of glyphosate contamination. The results could empower the development of environmental monitoring strategies with applications limited to neither glyphosate nor amplicon sequence data.Highlight bullet pointsAn artificial neural net was able to identify glyphosate-affected microbial community assemblages based on next generation sequencing dataDecision-relevant taxonomic clusters can be identified by a stochastically subsetting approachJust a fraction of present clusters is needed for classificationFiltering of input data improves classification

Download Full-text

A single layer artificial neural network type architecture with molecular engineered bacteria for reversible and irreversible computing

Chemical Science ◽

10.1039/d1sc01505b ◽

2021 ◽

Author(s):

Kathakali Sarkar ◽

Deepro Bonnerjee ◽

Rajkamal Srivastava ◽

Sangram Bagh

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Basic Concept ◽

Single Layer ◽

Artificial Neural ◽

Engineered Bacteria ◽

Network Type

Here, we adapted the basic concept of artificial neural networks (ANN) and experimentally demonstrate a broadly applicable single layer ANN type architecture with molecular engineered bacteria to perform complex irreversible...

Download Full-text

Forecasting Stock Returns with Artificial Neural Networks

Neural Networks in Business Forecasting ◽

10.4018/978-1-59140-176-6.ch003 ◽

2011 ◽

pp. 47-79 ◽

Cited By ~ 5

Author(s):

Suraphan Thawornwong ◽

David Enke

Keyword(s):

Neural Network ◽

Neural Networks ◽

Artificial Neural Network ◽

Artificial Neural Networks ◽

Stock Returns ◽

Future Research ◽

Network Architectures ◽

Network Applications ◽

Artificial Neural ◽

Neural Network Applications

During the last few years there has been growing literature on applications of artificial neural networks to business and financial domains. In fact, a great deal of attention has been placed in the area of stock return forecasting. This is due to the fact that once artificial neural network applications are successful, monetary rewards will be substantial. Many studies have reported promising results in successfully applying various types of artificial neural network architectures for predicting stock returns. This chapter reviews and discusses various neural network research methodologies used in 45 journal articles that attempted to forecast stock returns. Modeling techniques and suggestions from the literature are also compiled and addressed. The results show that artificial neural networks are an emerging and promising computational technology that will continue to be a challenging tool for future research.

Download Full-text