Search results for: approximation algorithms

Authors: José A. Menezes, Giordano Cabral, Bruno T. Gomes

Abstract:

Choosing good features is an essential part of machine learning. Recent techniques aim to automate this process. For instance, feature learning intends to learn the transformation of raw data into a useful representation to machine learning tasks. In automatic audio classification tasks, this is interesting since the audio, usually complex information, needs to be transformed into a computationally convenient input to process. Another technique tries to generate features by searching a feature space. Genetic algorithms, for instance, have being used to generate audio features by combining or modifying them. We find this approach particularly interesting and, despite the undeniable advances of feature learning approaches, we wanted to take a step forward in the use of genetic algorithms to find audio features, combining them with more conventional methods, like PCA, and inserting search control mechanisms, such as constraints over a confusion matrix. This work presents the results obtained on particular audio classification problems.

Keywords: feature generation, feature learning, genetic algorithm, music information retrieval

Procedia PDF Downloads 435

Authors: Yunus Doğan, Ahmet Durap

Abstract:

Coastal regions are the one of the most commonly used places by the natural balance and the growing population. In coastal engineering, the most valuable data is wave behaviors. The amount of this data becomes very big because of observations that take place for periods of hours, days and months. In this study, some statistical methods such as the wave spectrum analysis methods and the standard statistical methods have been used. The goal of this study is the discovery profiles of the different coast areas by using these statistical methods, and thus, obtaining an instance based data set from the big data to analysis by using data mining algorithms. In the experimental studies, the six sample data sets about the wave behaviors obtained by 20 minutes of observations from Mersin Bay in Turkey and converted to an instance based form, while different clustering techniques in data mining algorithms were used to discover similar coastal places. Moreover, this study discusses that this summarization approach can be used in other branches collecting big data such as medicine.

Keywords: clustering algorithms, coastal engineering, data mining, data summarization, statistical methods

Procedia PDF Downloads 361

Authors: Wilayat Ali, Li Sheng, Waleed Ahmed

Abstract:

The ability of the robot to make simultaneously map of the environment and localize itself with respect to that environment is the most important element of mobile robots. To solve SLAM many algorithms could be utilized to build up the SLAM process and SLAM is a developing area in Robotics research. Robot Operating System (ROS) is one of the frameworks which provide multiple algorithm nodes to work with and provide a transmission layer to robots. Manyof these algorithms extensively in use are Hector SLAM, Gmapping and Cartographer SLAM. This paper describes a ROS-based Simultaneous localization and mapping (SLAM) library Google Cartographer mapping, which is open-source algorithm. The algorithm was applied to create a map using laser and pose data from 2d Lidar that was placed on a mobile robot. The model robot uses the gazebo package and simulated in Rviz. Our research work's primary goal is to obtain mapping through Cartographer SLAM algorithm in a static indoor environment. From our research, it is shown that for indoor environments cartographer is an applicable algorithm to generate 2d maps with LIDAR placed on mobile robot because it uses both odometry and poses estimation. The algorithm has been evaluated and maps are constructed against the SLAM algorithms presented by Turtlebot2 in the static indoor environment.

Keywords: SLAM, ROS, navigation, localization and mapping, gazebo, Rviz, Turtlebot2, slam algorithms, 2d indoor environment, cartographer

Procedia PDF Downloads 145

Authors: Md Habibur Rahman, Jaeho Kim

Abstract:

Efficient process scheduling is a crucial factor in ensuring optimal system performance and resource utilization in computer systems. While various algorithms have been proposed over the years, there are still limitations to their effectiveness. This paper introduces a new Wait-Optimized Scheduler (WOS) algorithm that aims to minimize process waiting time by dividing them into two layers and considering both process time and waiting time. The WOS algorithm is non-preemptive and prioritizes processes with the shortest WOS. In the first layer, each process runs for a predetermined duration, and any unfinished process is subsequently moved to the second layer, resulting in a decrease in response time. Whenever the first layer is free or the number of processes in the second layer is twice that of the first layer, the algorithm sorts all the processes in the second layer based on their remaining time minus waiting time and sends one process to the first layer to run. This ensures that all processes eventually run, optimizing waiting time. To evaluate the performance of the WOS algorithm, we conducted experiments comparing its performance with traditional scheduling algorithms such as First-Come-First-Serve (FCFS) and Shortest-Job-First (SJF). The results showed that the WOS algorithm outperformed the traditional algorithms in reducing the waiting time of processes, particularly in scenarios with a large number of short tasks with long wait times. Our study highlights the effectiveness of the WOS algorithm in improving process scheduling efficiency in computer systems. By reducing process waiting time, the WOS algorithm can improve system performance and resource utilization. The findings of this study provide valuable insights for researchers and practitioners in developing and implementing efficient process scheduling algorithms.

Keywords: process scheduling, wait-optimized scheduler, response time, non-preemptive, waiting time, traditional scheduling algorithms, first-come-first-serve, shortest-job-first, system performance, resource utilization

Procedia PDF Downloads 91

Authors: B. Chandra

Abstract:

In Association Rule Mining, much attention has been paid for developing algorithms for large (frequent/closed/maximal) itemsets but very little attention has been paid to improve the performance of rule generation algorithms. Rule generation is an important part of Association Rule Mining. In this paper, a novel approach named NARG (Association Rule using Antecedent Support) has been proposed for rule generation that uses memory resident data structure named FCET (Frequent Closed Enumeration Tree) to find frequent/closed itemsets. In addition, the computational speed of NARG is enhanced by giving importance to the rules that have lower antecedent support. Comparative performance evaluation of NARG with fast association rule mining algorithm for rule generation has been done on synthetic datasets and real life datasets (taken from UCI Machine Learning Repository). Performance analysis shows that NARG is computationally faster in comparison to the existing algorithms for rule generation.

Keywords: knowledge discovery, association rule mining, antecedent support, rule generation

Procedia PDF Downloads 324

Authors: Mabrouka Algherinai, Fatma Karkouri

Abstract:

Today, Short Message Service (SMS) is an important means of communication. SMS is not only used in informal environment for communication and transaction, but it is also used in formal environments such as institutions, organizations, companies, and business world as a tool for communication and transactions. Therefore, there is a need to secure the information that is being transmitted through this medium to ensure security of information both in transit and at rest. But, encryption has been identified as a means to provide security to SMS messages in transit and at rest. Several past researches have proposed and developed several encryption algorithms for SMS and Information Security. This research aims at comparing the performance of common Asymmetric encryption algorithms on SMS security. The research employs the use of three algorithms, namely RSA, McEliece, and RABIN. Several experiments were performed on SMS of various sizes on android mobile device. The experimental results show that each of the three techniques has different key generation, encryption, and decryption times. The efficiency of an algorithm is determined by the time that it takes for encryption, decryption, and key generation. The best algorithm can be chosen based on the least time required for encryption. The obtained results show the least time when McEliece size 4096 is used. RABIN size 4096 gives most time for encryption and so it is the least effective algorithm when considering encryption. Also, the research shows that McEliece size 2048 has the least time for key generation, and hence, it is the best algorithm as relating to key generation. The result of the algorithms also shows that RSA size 1024 is the most preferable algorithm in terms of decryption as it gives the least time for decryption.

Keywords: SMS, RSA, McEliece, RABIN

Procedia PDF Downloads 163

Authors: Reza Dinarvand, Mahdi Sadeghian, Somaye Sadeghian

Abstract:

Prediction of the ultimate bearing capacity of piles (Qu) is one of the basic issues in geotechnical engineering. So far, several methods have been used to estimate Qu, including the recently developed artificial intelligence methods. In recent years, optimization algorithms have been used to minimize artificial network errors, such as colony algorithms, genetic algorithms, imperialist competitive algorithms, and so on. In the present research, artificial neural networks based on colonial competition algorithm (ANN-ICA) were used, and their results were compared with other methods. The results of laboratory tests of short piles in clayey soils with parameters such as pile diameter, pile buried length, eccentricity of load and undrained shear resistance of soil were used for modeling and evaluation. The results showed that ICA-based artificial neural networks predicted lateral bearing capacity of short piles with a correlation coefficient of 0.9865 for training data and 0.975 for test data. Furthermore, the results of the model indicated the superiority of ICA-based artificial neural networks compared to back-propagation artificial neural networks as well as the Broms and Hansen methods.

Keywords: artificial neural network, clayey soil, imperialist competition algorithm, lateral bearing capacity, short pile

Procedia PDF Downloads 152

Authors: Najwa Mimouni, Omar Rahli, Rachid Bennacer, Salah Chikh

Abstract:

In the present work 2D and 3D numerical simulations of double diffusion natural convection in an elongated enclosure filled with a binary fluid saturating a porous medium are carried out. In the formulation of the problem, the Boussinesq approximation is considered and cross Neumann boundary conditions are specified for heat and mass walls conditions. The numerical method is based on the control volume approach with the third order QUICK scheme. Full approximation storage (FAS) with full multigrid (FMG) method is used to solve the problem. For the explored large range of the controlling parameters, we clearly evidenced that the increase in the depth of the cavity i.e. the lateral aspect ratio has an important effect on the flow patterns. The 2D perfect parallel flows obtained for a small lateral aspect ratio are drastically destabilized by increasing the cavity lateral dimension. This yields a 3D fluid motion with a much more complicated flow pattern and the classically studied 2D parallel flows are impossible.

Keywords: bifurcation, natural convection, heat and mass transfer, parallel flow, porous media

Procedia PDF Downloads 473

Authors: Lubabatu Sada Sodangi

Abstract:

The rapid growth in exchange and accessibility of information via the internet makes many organisations acquire data on their own operation. The aim of data mining is to analyse the different behaviour of a dataset using observation. Although, the subset of the dataset being analysed may not display all the behaviours and relationships of the entire data and, therefore, may not represent other parts that exist in the dataset. There is a range of techniques used in data mining to determine the hidden or unknown information in datasets. In this paper, the performance of two algorithms Chi-Square Automatic Interaction Detection (CHAID) and multilayer perceptron (MLP) would be matched using an Adult dataset to find out the percentage of an/the adults that earn > 50k and those that earn <= 50k per year. The two algorithms were studied and compared using IBM SPSS statistics software. The result for CHAID shows that the most important predictors are relationship and education. The algorithm shows that those are married (husband) and have qualification: Bachelor, Masters, Doctorate or Prof-school whose their age is > 41<57 earn > 50k. Also, multilayer perceptron displays marital status and capital gain as the most important predictors of the income. It also shows that individuals that their capital gain is less than 6,849 and are single, separated or widow, earn <= 50K, whereas individuals with their capital gain is > 6,849, work > 35 hrs/wk, and > 27yrs their income will be > 50k. By comparing the two algorithms, it is observed that both algorithms are reliable but there is strong reliability in CHAID which clearly shows that relation and education contribute to the prediction as displayed in the data visualisation.

Keywords: data mining, CHAID, multi-layer perceptron, SPSS, Adult dataset

Procedia PDF Downloads 378

Authors: Dmitriy Kuptsov

Abstract:

Tree data structures, such as binary or in general k-ary trees, are essential in computer science. The applications of these data structures can range from data search and retrieval to sorting and ranking algorithms. Naive implementations of these data structures can consume prohibitively large volumes of random access memory limiting their applicability in certain solutions. Thus, in these cases, more advanced representation of these data structures is essential. In this paper we present the design of the compact version of ternary tree data structure and demonstrate the results for the experimental evaluation using static dictionary problem. We compare these results with the results for binary and regular ternary trees. The conducted evaluation study shows that our design, in the best case, consumes up to 12 times less memory (for the dictionary used in our experimental evaluation) than a regular ternary tree and in certain configuration shows performance comparable to regular ternary trees. We have evaluated the performance of the algorithms using both 32 and 64 bit operating systems.

Keywords: algorithms, data structures, succinct ternary tree, per- formance evaluation

Procedia PDF Downloads 160

Authors: T. Praveenkumar, Kulpreet Singh, Divy Bhanpuriya, M. Saimurugan

Abstract:

This study analysed the classification accuracy for gearbox faults using Machine Learning Techniques. Gearboxes are widely used for mechanical power transmission in rotating machines. Its rotating components such as bearings, gears, and shafts tend to wear due to prolonged usage, causing fluctuating vibrations. Increasing the dependability of mechanical components like a gearbox is hampered by their sealed design, which makes visual inspection difficult. One way of detecting impending failure is to detect a change in the vibration signature. The current study proposes various machine learning algorithms, with aid of these vibration signals for obtaining the fault classification accuracy of an automotive 4-Speed synchromesh gearbox. Experimental data in the form of vibration signals were acquired from a 4-Speed synchromesh gearbox using Data Acquisition System (DAQs). Statistical features were extracted from the acquired vibration signal under various operating conditions. Then the extracted features were given as input to the algorithms for fault classification. Supervised Machine Learning algorithms such as Support Vector Machines (SVM) and unsupervised algorithms such as Deep Feed Forward Neural Network (DFFNN), Deep Belief Networks (DBN) algorithms are used for fault classification. The fusion of DBN & DFFNN classifiers were architected to further enhance the classification accuracy and to reduce the computational complexity. The fault classification accuracy for each algorithm was thoroughly studied, tabulated, and graphically analysed for fused and individual algorithms. In conclusion, the fusion of DBN and DFFNN algorithm yielded the better classification accuracy and was selected for fault detection due to its faster computational processing and greater efficiency.

Keywords: deep belief networks, DBN, deep feed forward neural network, DFFNN, fault diagnosis, fusion of algorithm, vibration signal

Procedia PDF Downloads 113

Authors: Samia Sadouki Chibani, Abdelkamel Tari

Abstract:

Web service composition combines available services to provide new functionality. Given the number of available services with similar functionalities and different non functional aspects (QoS), the problem of finding a QoS-optimal web service composition is considered as an optimization problem belonging to NP-hard class. Thus, an optimal solution cannot be found by exact algorithms within a reasonable time. In this paper, a meta-heuristic bio-inspired is presented to address the QoS aware web service composition; it is based on Elephant Herding Optimization (EHO) algorithm, which is inspired by the herding behavior of elephant group. EHO is characterized by a process of dividing and combining the population to sub populations (clan); this process allows the exchange of information between local searches to move toward a global optimum. However, with Applying others evolutionary algorithms the problem of early stagnancy in a local optimum cannot be avoided. Compared with PSO, the results of experimental evaluation show that our proposition significantly outperforms the existing algorithm with better performance of the fitness value and a fast convergence.

Keywords: bio-inspired algorithms, elephant herding optimization, QoS optimization, web service composition

Procedia PDF Downloads 327

Authors: Sung Duk Kim, Kwan U. Kim, Jin Sim, Ryong Jin Jang

Abstract:

The phenomenon of seasonal variations in the Earth’s rotation velocity was discovered in the 1930s when a crystal clock was developed and analyzed in a quantitative way for the first time between 1955 and 1968 when observation data of the seasonal variations was analyzed by an atomic clock. According to the previous investigation, atmospheric circulation is supposed to be a factor affecting the seasonal variations in the Earth’s rotation velocity in many cases, but the problem has not been solved yet. In order to solve the problem, it is necessary to apply dynamics to consider the Earth’s spatial motion, rotation, and change of shape of the Earth (movement of materials in and out of the Earth and change of the Earth’s figure) at the same time and in interrelation to the accuracy of post-Newtonian approximation regarding the Earth body as a system of mass points because the stability of the Earth’s rotation angular velocity is in the range of 10⁻⁸~10⁻⁹. For it, the equation was derived, which can consider the 3 kinds of motion above mentioned at the same time by taking the effect of the resultant external force on the Earth’s rotation into account in a relativistic way to the accuracy of post-Newtonian approximation. Therefore, the equation has been solved to obtain the theoretical values of periodic change in the Earth’s rotation velocity, and they have been compared with the astronomical observation data so to reveal the cause for the periodic change in the Earth’s rotation velocity.

Keywords: Earth rotation, moment function, periodic change, seasonal variation, relativistic change

Procedia PDF Downloads 73

Authors: Qusay Alghazali, Tibor Cinkler, Abdulhalim Fayad

Abstract:

In modern optical networks, the flex grid spectrum usage is most widespread, where higher bit rate streams get larger spectrum slices while lower bit rate traffic streams get smaller spectrum slices. To our practice, under the ITU-T recommendation, G.694.1, spectrum slices of 50, 75, and 100 GHz are being used with central frequency at 193.1 THz. However, when these spectrum slices are not sufficient, multiple spectrum slices can use either one next to another or anywhere in the optical wavelength. In this paper, we propose the analysis of the wavelength assignment problem. We compare different algorithms for this spectrum assignment with and without protection. As a reference for comparisons, we concluded that the Integer Linear Programming (ILP) provides the global optimum for all cases. The most scalable algorithm is the greedy one, which yields results in subsequent ranges even for more significant network instances. The algorithms’ benchmark implemented using the LEMON C++ optimization library and simulation runs based on a minimum number of spectrum slices assigned to lightpaths and their execution time.

Keywords: spectrum assignment, integer linear programming, greedy algorithm, international telecommunication union, library for efficient modeling and optimization in networks

Procedia PDF Downloads 169

Authors: Diego De Castro Rodrigues, Marcelo Lisboa Rocha, Daniela M. De Q. Trevisan, Marcos Dias Da Conceicao, Gabriel Rosa, Rommel M. Barbosa

Abstract:

The use of association rules algorithms within data mining is recognized as being of great value in the knowledge discovery in databases. Very often, the number of rules generated is high, sometimes even in databases with small volume, so the success in the analysis of results can be hampered by this quantity. The purpose of this research is to present a method for reducing the quantity of rules generated with association algorithms. Therefore, a computational algorithm was developed with the use of a Weka Application Programming Interface, which allows the execution of the method on different types of databases. After the development, tests were carried out on three types of databases: synthetic, model, and real. Efficient results were obtained in reducing the number of rules, where the worst case presented a gain of more than 50%, considering the concepts of support, confidence, and lift as measures. This study concluded that the proposed model is feasible and quite interesting, contributing to the analysis of the results of association rules generated from the use of algorithms.

Keywords: data mining, association rules, rules reduction, artificial intelligence

Procedia PDF Downloads 160

Authors: Li Long, Thomas Ortlepp

Abstract:

A theoretical model for the optimization of thermopile sensor performance is developed for thermoelectric-based infrared radiation detection. It is shown that the performance of polycrystalline silicon film thermopile sensor can be optimized according to the thermoelectric quality factor, sensor layer structure factor, and sensor layout geometrical form factor. Based on the properties of electrons, phonons, grain boundaries, and their interactions, the thermoelectric quality factor of polycrystalline silicon is analyzed with the relaxation time approximation of the Boltzmann transport equation. The model includes the effect of grain structure, grain boundary trap properties, and doping concentration. The layer structure factor is analyzed with respect to the infrared absorption coefficient. The optimization of layout design is characterized by the form factor, which is calculated for different sensor designs. A double-layer polycrystalline silicon thermopile infrared sensor on a suspended membrane has been designed and fabricated with a CMOS-compatible process. The theoretical approach is confirmed by measurement results.

Keywords: polycrystalline silicon, relaxation time approximation, specific detectivity, thermal conductivity, thermopile infrared sensor

Procedia PDF Downloads 139

Authors: Alper Akin, Ibrahim Aydogdu

Abstract:

This study presents a hybrid metaheuristic algorithm to obtain optimum designs for steel space buildings. The optimum design problem of three-dimensional steel frames is mathematically formulated according to provisions of LRFD-AISC (Load and Resistance factor design of American Institute of Steel Construction). Design constraints such as the strength requirements of structural members, the displacement limitations, the inter-story drift and the other structural constraints are derived from LRFD-AISC specification. In this study, a hybrid algorithm by using teaching-learning based optimization (TLBO) and harmony search (HS) algorithms is employed to solve the stated optimum design problem. These algorithms are two of the recent additions to metaheuristic techniques of numerical optimization and have been an efficient tool for solving discrete programming problems. Using these two algorithms in collaboration creates a more powerful tool and mitigates each other’s weaknesses. To demonstrate the powerful performance of presented hybrid algorithm, the optimum design of a large scale steel building is presented and the results are compared to the previously obtained results available in the literature.

Keywords: optimum structural design, hybrid techniques, teaching-learning based optimization, harmony search algorithm, minimum weight, steel space frame

Procedia PDF Downloads 545

Authors: Gautam Kumar Saharia, Sagardeep Talukdar, Riki Dutta, Sudipta Nandy

Abstract:

The physics-informed neural network (PINN) method opens up an approach for numerically solving nonlinear partial differential equations leveraging fast calculating speed and high precession of modern computing systems. We construct the PINN based on a strong universal approximation theorem and apply the initial-boundary value data and residual collocation points to weekly impose initial and boundary conditions to the neural network and choose the optimization algorithms adaptive moment estimation (ADAM) and Limited-memory Broyden-Fletcher-Golfard-Shanno (L-BFGS) algorithm to optimize learnable parameter of the neural network. Next, we improve the PINN with a weighted loss function to obtain both the bright and dark soliton solutions of the Fokas-Lenells equation (FLE). We find the proposed scheme of adjustable weight coefficients into PINN has a better convergence rate and generalizability than the basic PINN algorithm. We believe that the PINN approach to solve the partial differential equation appearing in nonlinear optics would be useful in studying various optical phenomena.

Keywords: deep learning, optical soliton, physics informed neural network, partial differential equation

Procedia PDF Downloads 70

Authors: Aydin Teymourifar, Gurkan Ozturk

Abstract:

In this paper, a new hybrid particle swarm optimization algorithm is proposed for the multi-objective flexible job shop scheduling problem that is very important and hard combinatorial problem. The Pareto approach is used for solving the multi-objective problem. Several new local search heuristics are integrated into an algorithm based on the critical block concept to enhance the performance of the algorithm. The algorithm is compared with the recently published multi-objective algorithms based on benchmarks selected from the literature. Several metrics are used for quantifying performance and comparison of the achieved solutions. The algorithms are also compared based on the Weighting summation of objectives approach. The proposed algorithm can find the Pareto solutions more efficiently than the compared algorithms in less computational time.

Keywords: swarm-based optimization, local search, Pareto optimality, flexible job shop scheduling, multi-objective optimization

Procedia PDF Downloads 368

Authors: Prisecaru Diana-Sorina

Abstract:

The paper will show that, while the EU legislator tackled a series of UX patterns used in e-commerce to induce the consumers take actions that they would not normally undertake, it leaves out many other aspects related to misuse or poor UX design that adversely affect EU consumers. Further, the paper proposes a reevaluation of the regulatory addressability of the issue and hand and focuses on explaining why a joint strategy, based on the interplay between provisions aiming consumer protection and personal data protection is the key approach to this matter.

Keywords: algorithms, consumer protection, European Union, user experience design

Procedia PDF Downloads 136

Authors: Syed Romat Ali Shah, Bilal Shoaib, Saleem Akhtar, Munib Ahmad, Shahan Sadiqui

Abstract:

Classification is a data mining technique, which is recognizedon Machine Learning (ML) algorithm. It is used to classifythe individual articlein a knownofinformation into a set of predefinemodules or group. Web mining is also a portion of that sympathetic of data mining methods. The main purpose of this paper to analysis and compare the performance of Naïve Bayse Algorithm, Decision Tree, K-Nearest Neighbor (KNN), Artificial Neural Network (ANN)and Support Vector Machine (SVM). This paper consists of different ML algorithm and their advantages and disadvantages and also define research issues.

Keywords: Data Mining, Web Mining, classification, ML Algorithms

Procedia PDF Downloads 303

Authors: Andre M. Carvalho, Pedro Sebastiao

Abstract:

This article is based on a dissertation that intends to analyze and make a model, intelligently, algorithms based on stochastic processes of a gamification application applied to marketing. Gamification is used in our daily lives to engage us to perform certain actions in order to achieve goals and gain rewards. This strategy is an increasingly adopted way to encourage and retain customers through game elements. The application of gamification aims to encourage children between 6 and 10 years of age to have healthy habits and the purpose of serving as a model for use in marketing. This application was developed in unity; we implemented intelligent algorithms based on stochastic processes, web services to respond to all requests of the application, a back-office website to manage the application and the database. The behavioral analysis of the use of game elements and stochastic processes in children’s motivation was done. The application of algorithms based on stochastic processes in-game elements is very important to promote cooperation and to ensure fair and friendly competition between users which consequently stimulates the user’s interest and their involvement in the application and organization.

Keywords: engage, games, gamification, randomness, stochastic processes

Procedia PDF Downloads 330

Authors: Apoorva Vinod, Archana Mathur, Snehanshu Saha

Abstract:

Though there exists a plethora of Activation Functions (AFs) used in single and multiple hidden layer Neural Networks (NN), their behavior always raised curiosity, whether used in combination or singly. The popular AFs –Sigmoid, ReLU, and Tanh–have performed prominently well for shallow and deep architectures. Most of the time, AFs are used singly in multi-layered NN, and, to the best of our knowledge, their performance is never studied and analyzed deeply when used in combination. In this manuscript, we experiment with multi-layered NN architecture (both on shallow and deep architectures; Convolutional NN and VGG16) and investigate how well the network responds to using two different AFs (Sigmoid-Tanh, Tanh-ReLU, ReLU-Sigmoid) used alternately against a traditional, single (Sigmoid-Sigmoid, Tanh-Tanh, ReLUReLU) combination. Our results show that using two different AFs, the network achieves better accuracy, substantially lower loss, and faster convergence on 4 computer vision (CV) and 15 Non-CV (NCV) datasets. When using different AFs, not only was the accuracy greater by 6-7%, but we also accomplished convergence twice as fast. We present a case study to investigate the probability of networks suffering vanishing and exploding gradients when using two different AFs. Additionally, we theoretically showed that a composition of two or more AFs satisfies Universal Approximation Theorem (UAT).

Keywords: activation function, universal approximation function, neural networks, convergence

Procedia PDF Downloads 158

Authors: Derick M. Rakgoale, Topside E. Mathonsi, Vusumuzi Malele

Abstract:

Cloud computing offers distributed online and on-demand computational services from anywhere in the world. Cloud computing services have grown immensely over the past years, especially in the past year due to the Coronavirus pandemic. Cloud computing has changed the working environment and introduced work from work phenomenon, which enabled the adoption of technologies to fulfill the new workings, including cloud services offerings. The increased cloud computing adoption has come with new challenges regarding data privacy and its integrity in the cloud environment. Previously advanced encryption algorithms failed to reduce the memory space required for cloud computing performance, thus increasing the computational cost. This paper reviews the existing encryption algorithms used in cloud computing. In the future, artificial neural networks (ANN) algorithm design will be presented as a security solution to ensure data integrity, confidentiality, privacy, and availability of user data in cloud computing. Moreover, MATLAB will be used to evaluate the proposed solution, and simulation results will be presented.

Keywords: cloud computing, data integrity, confidentiality, privacy, availability

Procedia PDF Downloads 132

Authors: Augustine Okolie, Johannes Müller, Mirjam Kretzchmar

Abstract:

We adopt a maximum-likelihood framework to estimate parameters of a stochastic susceptible-infected-recovered (SIR) model with contact tracing on a rooted random tree. Given the number of detectees per index case, our estimator allows to determine the degree distribution of the random tree as well as the tracing probability. Since we do not discover all infectees via contact tracing, this estimation is non-trivial. To keep things simple and stable, we develop an approximation suited for realistic situations (contract tracing probability small, or the probability for the detection of index cases small). In this approximation, the only epidemiological parameter entering the estimator is the basic reproduction number R0. The estimator is tested in a simulation study and applied to covid-19 contact tracing data from India. The simulation study underlines the efficiency of the method. For the empirical covid-19 data, we are able to compare different degree distributions and perform a sensitivity analysis. We find that particularly a power-law and a negative binomial degree distribution meet the data well and that the tracing probability is rather large. The sensitivity analysis shows no strong dependency on the reproduction number.

Keywords: stochastic SIR model on graph, contact tracing, branching process, parameter inference

Procedia PDF Downloads 77

Authors: Gabriel Quintero

Abstract:

This paper presents the result of the implementation of a series of algorithms intended to be used for representing in most of the 3D geographic software, even Google Earth, the subsurface formations properties combining 2D charts or 3D plots over a 3D background, allowing everyone to use them, no matter the economic size of the company for which they work. Besides the existence of complex and expensive specialized software for modeling subsurface formations based on the same information provided to this one, the use of this open source development shows a higher and easier usability and good results, limiting the rendered properties and polygons to a basic set of charts and tubes.

Keywords: chart, earth, formations, subsurface, visualization

Procedia PDF Downloads 442

Authors: Jeff Clarine, Chang-Shyh Peng, Daisy Sang

Abstract:

Bioassay is the measurement of the potency of a chemical substance by its effect on a living animal or plant tissue. Bioassay data and chemical structures from pharmacokinetic and drug metabolism screening are mined from and housed in multiple databases. Bioassay prediction is calculated accordingly to determine further advancement. This paper proposes a four-step preprocessing of datasets for improving the bioassay predictions. The first step is instance selection in which dataset is categorized into training, testing, and validation sets. The second step is discretization that partitions the data in consideration of accuracy vs. precision. The third step is normalization where data are normalized between 0 and 1 for subsequent machine learning processing. The fourth step is feature selection where key chemical properties and attributes are generated. The streamlined results are then analyzed for the prediction of effectiveness by various machine learning algorithms including Pipeline Pilot, R, Weka, and Excel. Experiments and evaluations reveal the effectiveness of various combination of preprocessing steps and machine learning algorithms in more consistent and accurate prediction.

Keywords: bioassay, machine learning, preprocessing, virtual screen

Procedia PDF Downloads 274

Authors: Pascal François Faye, Baye Mor Sall, Bineta Dembele, Jeanne Ana Awa Faye

Abstract:

In agriculture, estimating crop yields is key to improving productivity and decision-making processes such as financial market forecasting and addressing food security issues. The main objective of this paper is to have tools to predict and improve the accuracy of crop yield forecasts using machine learning (ML) algorithms such as CART , KNN and SVM . We developed a mobile app and a web app that uses these algorithms for practical use by farmers. The tests show that our system (collection and deployment architecture, web application and mobile application) is operational and validates empirical knowledge on agro-climatic parameters in addition to proactive decision-making support. The experimental results obtained on the agricultural data, the performance of the ML algorithms are compared using cross-validation in order to identify the most effective ones following the agricultural data. The proposed applications demonstrate that the proposed approach is effective in predicting crop yields and provides timely and accurate responses to farmers for decision support.

Keywords: prediction, machine learning, artificial intelligence, digital agriculture

Procedia PDF Downloads 80

Authors: Ahlem Mougaida, Hedi Bel Hadj Salah

Abstract:

Several methods are suggested to improve the numerical integration in Galerkin weak form for Meshless methods. In fact, integrating without taking into account the characteristics of the shape functions reproduced by Meshless methods (rational functions, with compact support etc.), causes a large integration error that influences the PDE’s approximate solution. Comparisons between different methods of numerical integration for rational functions are discussed and compared. The algorithms are implemented in Matlab. Finally, numerical results were presented to prove the efficiency of our algorithms in improving results.

Keywords: adaptive methods, meshless, numerical integration, rational quadrature

Procedia PDF Downloads 364

Authors: Vesile Evrim, Aliyu Awwal

Abstract:

By the evolvement in technology, the way of expressing opinions switched the direction to the digital world. The domain of politics as one of the hottest topics of opinion mining research merged together with the behavior analysis for affiliation determination in text which constitutes the subject of this paper. This study aims to classify the text in news/blogs either as Republican or Democrat with the minimum number of features. As an initial set, 68 features which 64 are constituted by Linguistic Inquiry and Word Count (LIWC) features are tested against 14 benchmark classification algorithms. In the later experiments, the dimensions of the feature vector reduced based on the 7 feature selection algorithms. The results show that Decision Tree, Rule Induction and M5 Rule classifiers when used with SVM and IGR feature selection algorithms performed the best up to 82.5% accuracy on a given dataset. Further tests on a single feature and the linguistic based feature sets showed the similar results. The feature “function” as an aggregate feature of the linguistic category, is obtained as the most differentiating feature among the 68 features with 81% accuracy by itself in classifying articles either as Republican or Democrat.

Keywords: feature selection, LIWC, machine learning, politics

Procedia PDF Downloads 382