World Academy of Science, Engineering and Technology
[Electrical and Information Engineering]
Online ISSN : 1307-6892
162 Robot Movement Using the Trust Region Policy Optimization
Authors: Romisaa Ali
Abstract:
The Policy Gradient approach is one of the deep reinforcement learning families that combines deep neural networks (DNN) with reinforcement learning RL to discover the optimum of the control problem through experience gained from the interaction between the robot and its surroundings. In contrast to earlier policy gradient algorithms, which were unable to handle these two types of error because of over-or under-estimation introduced by the deep neural network model, this article will discuss the state-of-the-art SOTA policy gradient technique, trust region policy optimization (TRPO), by applying this method in various environments compared to another policy gradient method, the Proximal Policy Optimization (PPO), to explain their robust optimization, using this SOTA to gather experience data during various training phases after observing the impact of hyper-parameters on neural network performance.Keywords: deep neural networks, deep reinforcement learning, proximal policy optimization, state-of-the-art, trust region policy optimization
Procedia PDF Downloads 173161 Artificial Neural Network Approach for Modeling Very Short-Term Wind Speed Prediction
Authors: Joselito Medina-Marin, Maria G. Serna-Diaz, Juan C. Seck-Tuoh-Mora, Norberto Hernandez-Romero, Irving Barragán-Vite
Abstract:
Wind speed forecasting is an important issue for planning wind power generation facilities. The accuracy in the wind speed prediction allows a good performance of wind turbines for electricity generation. A model based on artificial neural networks is presented in this work. A dataset with atmospheric information about air temperature, atmospheric pressure, wind direction, and wind speed in Pachuca, Hidalgo, México, was used to train the artificial neural network. The data was downloaded from the web page of the National Meteorological Service of the Mexican government. The records were gathered for three months, with time intervals of ten minutes. This dataset was used to develop an iterative algorithm to create 1,110 ANNs, with different configurations, starting from one to three hidden layers and every hidden layer with a number of neurons from 1 to 10. Each ANN was trained with the Levenberg-Marquardt backpropagation algorithm, which is used to learn the relationship between input and output values. The model with the best performance contains three hidden layers and 9, 6, and 5 neurons, respectively; and the coefficient of determination obtained was r²=0.9414, and the Root Mean Squared Error is 1.0559. In summary, the ANN approach is suitable to predict the wind speed in Pachuca City because the r² value denotes a good fitting of gathered records, and the obtained ANN model can be used in the planning of wind power generation grids.Keywords: wind power generation, artificial neural networks, wind speed, coefficient of determination
Procedia PDF Downloads 128160 Proportional and Integral Controller-Based Direct Current Servo Motor Speed Characterization
Authors: Adel Salem Bahakeem, Ahmad Jamal, Mir Md. Maruf Morshed, Elwaleed Awad Khidir
Abstract:
Direct Current (DC) servo motors, or simply DC motors, play an important role in many industrial applications such as manufacturing of plastics, precise positioning of the equipment, and operating computer-controlled systems where speed of feed control, maintaining the position, and ensuring to have a constantly desired output is very critical. These parameters can be controlled with the help of control systems such as the Proportional Integral Derivative (PID) controller. The aim of the current work is to investigate the effects of Proportional (P) and Integral (I) controllers on the steady state and transient response of the DC motor. The controller gains are varied to observe their effects on the error, damping, and stability of the steady and transient motor response. The current investigation is conducted experimentally on a servo trainer CE 110 using analog PI controller CE 120 and theoretically using Simulink in MATLAB. Both experimental and theoretical work involves varying integral controller gain to obtain the response to a steady-state input, varying, individually, the proportional and integral controller gains to obtain the response to a step input function at a certain frequency, and theoretically obtaining the proportional and integral controller gains for desired values of damping ratio and response frequency. Results reveal that a proportional controller helps reduce the steady-state and transient error between the input signal and output response and makes the system more stable. In addition, it also speeds up the response of the system. On the other hand, the integral controller eliminates the error but tends to make the system unstable with induced oscillations and slow response to eliminate the error. From the current work, it is desired to achieve a stable response of the servo motor in terms of its angular velocity subjected to steady-state and transient input signals by utilizing the strengths of both P and I controllers.Keywords: DC servo motor, proportional controller, integral controller, controller gain optimization, Simulink
Procedia PDF Downloads 113159 A Grey-Box Text Attack Framework Using Explainable AI
Authors: Esther Chiramal, Kelvin Soh Boon Kai
Abstract:
Explainable AI is a strong strategy implemented to understand complex black-box model predictions in a human-interpretable language. It provides the evidence required to execute the use of trustworthy and reliable AI systems. On the other hand, however, it also opens the door to locating possible vulnerabilities in an AI model. Traditional adversarial text attack uses word substitution, data augmentation techniques, and gradient-based attacks on powerful pre-trained Bidirectional Encoder Representations from Transformers (BERT) variants to generate adversarial sentences. These attacks are generally white-box in nature and not practical as they can be easily detected by humans e.g., Changing the word from “Poor” to “Rich”. We proposed a simple yet effective Grey-box cum Black-box approach that does not require the knowledge of the model while using a set of surrogate Transformer/BERT models to perform the attack using Explainable AI techniques. As Transformers are the current state-of-the-art models for almost all Natural Language Processing (NLP) tasks, an attack generated from BERT1 is transferable to BERT2. This transferability is made possible due to the attention mechanism in the transformer that allows the model to capture long-range dependencies in a sequence. Using the power of BERT generalisation via attention, we attempt to exploit how transformers learn by attacking a few surrogate transformer variants which are all based on a different architecture. We demonstrate that this approach is highly effective to generate semantically good sentences by changing as little as one word that is not detectable by humans while still fooling other BERT models.Keywords: BERT, explainable AI, Grey-box text attack, transformer
Procedia PDF Downloads 141158 Simulation, Optimization, and Analysis Approach of Microgrid Systems
Authors: Saqib Ali
Abstract:
Sources are classified into two depending upon the factor of reviving. These sources, which cannot be revived into their original shape once they are consumed, are considered as nonrenewable energy resources, i.e., (coal, fuel) Moreover, those energy resources which are revivable to the original condition even after being consumed are known as renewable energy resources, i.e., (wind, solar, hydel) Renewable energy is a cost-effective way to generate clean and green electrical energy Now a day’s majority of the countries are paying heed to energy generation from RES Pakistan is mostly relying on conventional energy resources which are mostly nonrenewable in nature coal, fuel is one of the major resources, and with the advent of time their prices are increasing on the other hand RES have great potential in the country with the deployment of RES greater reliability and an effective power system can be obtained In this thesis, a similar concept is being used and a hybrid power system is proposed which is composed of intermixing of renewable and nonrenewable sources The Source side is composed of solar, wind, fuel cells which will be used in an optimal manner to serve load The goal is to provide an economical, reliable, uninterruptable power supply. This is achieved by optimal controller (PI, PD, PID, FOPID) Optimization techniques are applied to the controllers to achieve the desired results. Advanced algorithms (Particle swarm optimization, Flower Pollination Algorithm) will be used to extract the desired output from the controller Detailed comparison in the form of tables and results will be provided, which will highlight the efficiency of the proposed system.Keywords: distributed generation, demand-side management, hybrid power system, micro grid, renewable energy resources, supply-side management
Procedia PDF Downloads 101157 Model Predictive Controller for Pasteurization Process
Authors: Tesfaye Alamirew Dessie
Abstract:
Our study focuses on developing a Model Predictive Controller (MPC) and evaluating it against a traditional PID for a pasteurization process. Utilizing system identification from the experimental data, the dynamics of the pasteurization process were calculated. Using best fit with data validation, residual, and stability analysis, the quality of several model architectures was evaluated. The validation data fit the auto-regressive with exogenous input (ARX322) model of the pasteurization process by roughly 80.37 percent. The ARX322 model structure was used to create MPC and PID control techniques. After comparing controller performance based on settling time, overshoot percentage, and stability analysis, it was found that MPC controllers outperform PID for those parameters.Keywords: MPC, PID, ARX, pasteurization
Procedia PDF Downloads 167156 Between Reality and Fiction: Self-Representation as an Avatar and Its Effects on Self-Presence
Authors: Leonie Laskowitz
Abstract:
A self-confident appearance is a basic prerequisite for success in the world of work 4.0. Within a few seconds, people convey a first impression that usually lasts. Artificial intelligence is making it increasingly important how our virtual selves appear and communicate (nonverbally) in digital worlds such as the metaverse. In addition to the modified creation of an avatar, the field of photogrammetry is developing fast, creating exact likenesses of ourselves in virtual environments. Given the importance of self-representation in virtual space for future collaborations, it is important to investigate the impact of phenotype in virtual worlds and how an avatar type can profitably be used situationally. We analyzed the effect of self-similar versus desirable self-presentation as an avatar on one's self-awareness, considering various theoretical constructs in the area of self-awareness and stress stimuli. The avatars were arbitrarily created on the one hand and scanned on the other hand with the help of a lidar sensor, the state-of-the-art photogrammetry method. All subjects were exposed to the established Trier Social Stress Test. The results showed that especially insecure people prefer to create rather than be scanned when confronted with a stressful work situation. (1) If they are in a casual work environment and a relaxed situation, they prefer a 3D photorealistic avatar that reflects them in detail. (2) Confident people will give their avatar their true appearance in any situation, while insecure people would only do so for honesty and authenticity. (3) Thus, the choice of avatar type has considerable impact on self-confidence in different situations.Keywords: avatar, virtual identity, self-presentation, metaverse, virtual reality, self-awareness
Procedia PDF Downloads 151155 Cross-border Data Transfers to and from South Africa
Authors: Amy Gooden, Meshandren Naidoo
Abstract:
Genetic research and transfers of big data are not confined to a particular jurisdiction, but there is a lack of clarity regarding the legal requirements for importing and exporting such data. Using direct-to-consumer genetic testing (DTC-GT) as an example, this research assesses the status of data sharing into and out of South Africa (SA). While SA laws cover the sending of genetic data out of SA, prohibiting such transfer unless a legal ground exists, the position where genetic data comes into the country depends on the laws of the country from where it is sent – making the legal position less clear.Keywords: cross-border, data, genetic testing, law, regulation, research, sharing, South Africa
Procedia PDF Downloads 128154 3D Guided Image Filtering to Improve Quality of Short-Time Binned Dynamic PET Images Using MRI Images
Authors: Tabassum Husain, Shen Peng Li, Zhaolin Chen
Abstract:
This paper evaluates the usability of 3D Guided Image Filtering to enhance the quality of short-time binned dynamic PET images by using MRI images. Guided image filtering is an edge-preserving filter proposed to enhance 2D images. The 3D filter is applied on 1 and 5-minute binned images. The results are compared with 15-minute binned images and the Gaussian filtering. The guided image filter enhances the quality of dynamic PET images while also preserving important information of the voxels.Keywords: dynamic PET images, guided image filter, image enhancement, information preservation filtering
Procedia PDF Downloads 137153 Solving a Micromouse Maze Using an Ant-Inspired Algorithm
Authors: Rolando Barradas, Salviano Soares, António Valente, José Alberto Lencastre, Paulo Oliveira
Abstract:
This article reviews the Ant Colony Optimization, a nature-inspired algorithm, and its implementation in the Scratch/m-Block programming environment. The Ant Colony Optimization is a part of Swarm Intelligence-based algorithms and is a subset of biological-inspired algorithms. Starting with a problem in which one has a maze and needs to find its path to the center and return to the starting position. This is similar to an ant looking for a path to a food source and returning to its nest. Starting with the implementation of a simple wall follower simulator, the proposed solution uses a dynamic graphical interface that allows young students to observe the ants’ movement while the algorithm optimizes the routes to the maze’s center. Things like interface usability, Data structures, and the conversion of algorithmic language to Scratch syntax were some of the details addressed during this implementation. This gives young students an easier way to understand the computational concepts of sequences, loops, parallelism, data, events, and conditionals, as they are used through all the implemented algorithms. Future work includes the simulation results with real contest mazes and two different pheromone update methods and the comparison with the optimized results of the winners of each one of the editions of the contest. It will also include the creation of a Digital Twin relating the virtual simulator with a real micromouse in a full-size maze. The first test results show that the algorithm found the same optimized solutions that were found by the winners of each one of the editions of the Micromouse contest making this a good solution for maze pathfinding.Keywords: nature inspired algorithms, scratch, micromouse, problem-solving, computational thinking
Procedia PDF Downloads 128152 Price Effect Estimation of Tobacco on Low-wage Male Smokers: A Causal Mediation Analysis
Authors: Kawsar Ahmed, Hong Wang
Abstract:
The study's goal was to estimate the causal mediation impact of tobacco tax before and after price hikes among low-income male smokers, with a particular emphasis on the effect estimating pathways framework for continuous and dichotomous variables. From July to December 2021, a cross-sectional investigation of observational data (n=739) was collected from Bangladeshi low-wage smokers. The Quasi-Bayesian technique, binomial probit model, and sensitivity analysis using a simulation of the computational tools R mediation package had been used to estimate the effect. After a price rise for tobacco products, the average number of cigarettes or bidis sticks taken decreased from 6.7 to 4.56. Tobacco product rising prices have a direct effect on low-income people's decisions to quit or lessen their daily smoking habits of Average Causal Mediation Effect (ACME) [effect=2.31, 95 % confidence interval (C.I.) = (4.71-0.00), p<0.01], Average Direct Effect (ADE) [effect=8.6, 95 percent (C.I.) = (6.8-0.11), p<0.001], and overall significant effects (p<0.001). Tobacco smoking choice is described by the mediated proportion of income effect, which is 26.1% less of following price rise. The curve of ACME and ADE is based on observational figures of the coefficients of determination that asses the model of hypothesis as the substantial consequence after price rises in the sensitivity analysis. To reduce smoking product behaviors, price increases through taxation have a positive causal mediation with income that affects the decision to limit tobacco use and promote low-income men's healthcare policy.Keywords: causal mediation analysis, directed acyclic graphs, tobacco price policy, sensitivity analysis, pathway estimation
Procedia PDF Downloads 117151 Human Activities Recognition Based on Expert System
Authors: Malika Yaici, Soraya Aloui, Sara Semchaoui
Abstract:
Recognition of human activities from sensor data is an active research area, and the main objective is to obtain a high recognition rate. In this work, we propose a recognition system based on expert systems. The proposed system makes the recognition based on the objects, object states, and gestures, taking into account the context (the location of the objects and of the person performing the activity, the duration of the elementary actions, and the activity). This work focuses on complex activities which are decomposed into simple easy to recognize activities. The proposed method can be applied to any type of activity. The simulation results show the robustness of our system and its speed of decision.Keywords: human activity recognition, ubiquitous computing, context-awareness, expert system
Procedia PDF Downloads 144150 Challenges of Blockchain Applications in the Supply Chain Industry: A Regulatory Perspective
Authors: Pardis Moslemzadeh Tehrani
Abstract:
Due to the emergence of blockchain technology and the benefits of cryptocurrencies, intelligent or smart contracts are gaining traction. Artificial intelligence (AI) is transforming our lives, and it is being embraced by a wide range of sectors. Smart contracts, which are at the heart of blockchains, incorporate AI characteristics. Such contracts are referred to as "smart" contracts because of the underlying technology that allows contracting parties to agree on terms expressed in computer code that defines machine-readable instructions for computers to follow under specific situations. The transmission happens automatically if the conditions are met. Initially utilised for financial transactions, blockchain applications have since expanded to include the financial, insurance, and medical sectors, as well as supply networks. Raw material acquisition by suppliers, design, and fabrication by manufacturers, delivery of final products to consumers, and even post-sales logistics assistance are all part of supply chains. Many issues are linked with managing supply chains from the planning and coordination stages, which can be implemented in a smart contract in a blockchain due to their complexity. Manufacturing delays and limited third-party amounts of product components have raised concerns about the integrity and accountability of supply chains for food and pharmaceutical items. Other concerns include regulatory compliance in multiple jurisdictions and transportation circumstances (for instance, many products must be kept in temperature-controlled environments to ensure their effectiveness). Products are handled by several providers before reaching customers in modern economic systems. Information is sent between suppliers, shippers, distributors, and retailers at every stage of the production and distribution process. Information travels more effectively when individuals are eliminated from the equation. The usage of blockchain technology could be a viable solution to these coordination issues. In blockchains, smart contracts allow for the rapid transmission of production data, logistical data, inventory levels, and sales data. This research investigates the legal and technical advantages and disadvantages of AI-blockchain technology in the supply chain business. It aims to uncover the applicable legal problems and barriers to the use of AI-blockchain technology to supply chains, particularly in the food industry. It also discusses the essential legal and technological issues and impediments to supply chain implementation for stakeholders, as well as methods for overcoming them before releasing the technology to clients. Because there has been little research done on this topic, it is difficult for industrial stakeholders to grasp how blockchain technology could be used in their respective operations. As a result, the focus of this research will be on building advanced and complex contractual terms in supply chain smart contracts on blockchains to cover all unforeseen supply chain challenges.Keywords: blockchain, supply chain, IoT, smart contract
Procedia PDF Downloads 132149 Collision Avoidance Based on Model Predictive Control for Nonlinear Octocopter Model
Authors: Doğan Yıldız, Aydan Müşerref Erkmen
Abstract:
The controller of the octocopter is mostly based on the PID controller. For complex maneuvers, PID controllers have limited performance capability like in collision avoidance. When an octocopter needs avoidance from an obstacle, it must instantly show an agile maneuver. Also, this kind of maneuver is affected severely by the nonlinear characteristic of octocopter. When these kinds of limitations are considered, the situation is highly challenging for the PID controller. In the proposed study, these challenges are tried to minimize by using the model predictive controller (MPC) for collision avoidance with a nonlinear octocopter model. The aim is to show that MPC-based collision avoidance has the capability to deal with fast varying conditions in case of obstacle detection and diminish the nonlinear effects of octocopter with varying disturbances.Keywords: model predictive control, nonlinear octocopter model, collision avoidance, obstacle detection
Procedia PDF Downloads 194148 The Bernstein Expansion for Exponentials in Taylor Functions: Approximation of Fixed Points
Authors: Tareq Hamadneh, Jochen Merker, Hassan Al-Zoubi
Abstract:
Bernstein's expansion for exponentials in Taylor functions provides lower and upper optimization values for the range of its original function. these values converge to the original functions if the degree is elevated or the domain subdivided. Taylor polynomial can be applied so that the exponential is a polynomial of finite degree over a given domain. Bernstein's basis has two main properties: its sum equals 1, and positive for all x 2 (0; 1). In this work, we prove the existence of fixed points for exponential functions in a given domain using the optimization values of Bernstein. The Bernstein basis of finite degree T over a domain D is defined non-negatively. Any polynomial p of degree t can be expanded into the Bernstein form of maximum degree t ≤ T, where we only need to compute the coefficients of Bernstein in order to optimize the original polynomial. The main property is that p(x) is approximated by the minimum and maximum Bernstein coefficients (Bernstein bound). If the bound is contained in the given domain, then we say that p(x) has fixed points in the same domain.Keywords: Bernstein polynomials, Stability of control functions, numerical optimization, Taylor function
Procedia PDF Downloads 138147 Application to Monitor the Citizens for Corona and Get Medical Aids or Assistance from Hospitals
Authors: Vathsala Kaluarachchi, Oshani Wimalarathna, Charith Vandebona, Gayani Chandrarathna, Lakmal Rupasinghe, Windhya Rankothge
Abstract:
It is the fundamental function of a monitoring system to allow users to collect and process data. A worldwide threat, the corona outbreak has wreaked havoc in Sri Lanka, and the situation has gotten out of hand. Since the epidemic, the Sri Lankan government has been unable to establish a systematic system for monitoring corona patients and providing emergency care in the event of an outbreak. Most patients have been held at home because of the high number of patients reported in the nation, but they do not yet have access to a functioning medical system. It has resulted in an increase in the number of patients who have been left untreated because of a lack of medical care. The absence of competent medical monitoring is the biggest cause of mortality for many people nowadays, according to our survey. As a result, a smartphone app for analyzing the patient's state and determining whether they should be hospitalized will be developed. Using the data supplied, we are aiming to send an alarm letter or SMS to the hospital once the system recognizes them. Since we know what those patients need and when they need it, we will put up a desktop program at the hospital to monitor their progress. Deep learning, image processing and application development, natural language processing, and blockchain management are some of the components of the research solution. The purpose of this research paper is to introduce a mechanism to connect hospitals and patients even when they are physically apart. Further data security and user-friendliness are enhanced through blockchain and NLP.Keywords: blockchain, deep learning, NLP, monitoring system
Procedia PDF Downloads 136146 Application of Federated Learning in the Health Care Sector for Malware Detection and Mitigation Using Software-Defined Networking Approach
Authors: A. Dinelka Panagoda, Bathiya Bandara, Chamod Wijetunga, Chathura Malinda, Lakmal Rupasinghe, Chethana Liyanapathirana
Abstract:
This research takes us forward with the concepts of Federated Learning and Software-Defined Networking (SDN) to introduce an efficient malware detection technique and provide a mitigation mechanism to give birth to a resilient and automated healthcare sector network system by also adding the feature of extended privacy preservation. Due to the daily transformation of new malware attacks on hospital Integrated Clinical Environment (ICEs), the healthcare industry is at an undefinable peak of never knowing its continuity direction. The state of blindness by the array of indispensable opportunities that new medical device inventions and their connected coordination offer daily, a factor that should be focused driven is not yet entirely understood by most healthcare operators and patients. This solution has the involvement of four clients in the form of hospital networks to build up the federated learning experimentation architectural structure with different geographical participation to reach the most reasonable accuracy rate with privacy preservation. While the logistic regression with cross-entropy conveys the detection, SDN comes in handy in the second half of the research to stack up the initial development phases of the system with malware mitigation based on policy implementation. The overall evaluation sums up with a system that proves the accuracy with the added privacy. It is no longer needed to continue with traditional centralized systems that offer almost everything but not privacy.Keywords: software-defined network, federated learning, privacy, integrated clinical environment, decentralized learning, malware detection, malware mitigation
Procedia PDF Downloads 191145 Analyze and Visualize Eye-Tracking Data
Authors: Aymen Sekhri, Emmanuel Kwabena Frimpong, Bolaji Mubarak Ayeyemi, Aleksi Hirvonen, Matias Hirvonen, Tedros Tesfay Andemichael
Abstract:
Fixation identification, which involves isolating and identifying fixations and saccades in eye-tracking protocols, is an important aspect of eye-movement data processing that can have a big impact on higher-level analyses. However, fixation identification techniques are frequently discussed informally and rarely compared in any meaningful way. With two state-of-the-art algorithms, we will implement fixation detection and analysis in this work. The velocity threshold fixation algorithm is the first algorithm, and it identifies fixation based on a threshold value. For eye movement detection, the second approach is U'n' Eye, a deep neural network algorithm. The goal of this project is to analyze and visualize eye-tracking data from an eye gaze dataset that has been provided. The data was collected in a scenario in which individuals were shown photos and asked whether or not they recognized them. The results of the two-fixation detection approach are contrasted and visualized in this paper.Keywords: human-computer interaction, eye-tracking, CNN, fixations, saccades
Procedia PDF Downloads 140144 Using Q-Learning to Auto-Tune PID Controller Gains for Online Quadcopter Altitude Stabilization
Authors: Y. Alrubyli
Abstract:
Unmanned Arial Vehicles (UAVs), and more specifically, quadcopters need to be stable during their flights. Altitude stability is usually achieved by using a PID controller that is built into the flight controller software. Furthermore, the PID controller has gains that need to be tuned to reach optimal altitude stabilization during the quadcopter’s flight. For that, control system engineers need to tune those gains by using extensive modeling of the environment, which might change from one environment and condition to another. As quadcopters penetrate more sectors, from the military to the consumer sectors, they have been put into complex and challenging environments more than ever before. Hence, intelligent self-stabilizing quadcopters are needed to maneuver through those complex environments and situations. Here we show that by using online reinforcement learning with minimal background knowledge, the altitude stability of the quadcopter can be achieved using a model-free approach. We found that by using background knowledge instead of letting the online reinforcement learning algorithm wander for a while to tune the PID gains, altitude stabilization can be achieved faster. In addition, using this approach will accelerate development by avoiding extensive simulations before applying the PID gains to the real-world quadcopter. Our results demonstrate the possibility of using the trial and error approach of reinforcement learning combined with background knowledge to achieve faster quadcopter altitude stabilization in different environments and conditions.Keywords: reinforcement learning, Q-leanring, online learning, PID tuning, unmanned aerial vehicle, quadcopter
Procedia PDF Downloads 183143 Using Social Network Analysis for Cyber Threat Intelligence
Authors: Vasileios Anastopoulos
Abstract:
Cyber threat intelligence assists organizations in understanding the threats they face and helps them make educated decisions on preparing their defenses. Sharing of threat intelligence and threat information is increasingly leveraged by organizations and enterprises, and various software solutions are already available, with the open-source malware information sharing platform (MISP) being a popular one. In this work, a methodology for the production of cyber threat intelligence using the threat information stored in MISP is proposed. The methodology leverages the discipline of social network analysis and the diamond model, a model used for intrusion analysis, to produce cyber threat intelligence. The workings are demonstrated with a case study on a production MISP instance of a real organization. The paper concluded with a discussion on the proposed methodology and possible directions for further research.Keywords: cyber threat intelligence, diamond model, malware information sharing platform, social network analysis
Procedia PDF Downloads 184142 Algorithms Inspired from Human Behavior Applied to Optimization of a Complex Process
Authors: S. Curteanu, F. Leon, M. Gavrilescu, S. A. Floria
Abstract:
Optimization algorithms inspired from human behavior were applied in this approach, associated with neural networks models. The algorithms belong to human behaviors of learning and cooperation and human competitive behavior classes. For the first class, the main strategies include: random learning, individual learning, and social learning, and the selected algorithms are: simplified human learning optimization (SHLO), social learning optimization (SLO), and teaching-learning based optimization (TLBO). For the second class, the concept of learning is associated with competitiveness, and the selected algorithms are sports-inspired algorithms (with Football Game Algorithm, FGA and Volleyball Premier League, VPL) and Imperialist Competitive Algorithm (ICA). A real process, the synthesis of polyacrylamide-based multicomponent hydrogels, where some parameters are difficult to obtain experimentally, is considered as a case study. Reaction yield and swelling degree are predicted as a function of reaction conditions (acrylamide concentration, initiator concentration, crosslinking agent concentration, temperature, reaction time, and amount of inclusion polymer, which could be starch, poly(vinyl alcohol) or gelatin). The experimental results contain 175 data. Artificial neural networks are obtained in optimal form with biologically inspired algorithm; the optimization being perform at two level: structural and parametric. Feedforward neural networks with one or two hidden layers and no more than 25 neurons in intermediate layers were obtained with values of correlation coefficient in the validation phase over 0.90. The best results were obtained with TLBO algorithm, correlation coefficient being 0.94 for an MLP(6:9:20:2) – a feedforward neural network with two hidden layers and 9 and 20, respectively, intermediate neurons. Good results obtained prove the efficiency of the optimization algorithms. More than the good results, what is important in this approach is the simulation methodology, including neural networks and optimization biologically inspired algorithms, which provide satisfactory results. In addition, the methodology developed in this approach is general and has flexibility so that it can be easily adapted to other processes in association with different types of models.Keywords: artificial neural networks, human behaviors of learning and cooperation, human competitive behavior, optimization algorithms
Procedia PDF Downloads 110141 Use of Machine Learning in Data Quality Assessment
Authors: Bruno Pinto Vieira, Marco Antonio Calijorne Soares, Armando Sérgio de Aguiar Filho
Abstract:
Nowadays, a massive amount of information has been produced by different data sources, including mobile devices and transactional systems. In this scenario, concerns arise on how to maintain or establish data quality, which is now treated as a product to be defined, measured, analyzed, and improved to meet consumers' needs, which is the one who uses these data in decision making and companies strategies. Information that reaches low levels of quality can lead to issues that can consume time and money, such as missed business opportunities, inadequate decisions, and bad risk management actions. The step of selecting, identifying, evaluating, and selecting data sources with significant quality according to the need has become a costly task for users since the sources do not provide information about their quality. Traditional data quality control methods are based on user experience or business rules limiting performance and slowing down the process with less than desirable accuracy. Using advanced machine learning algorithms, it is possible to take advantage of computational resources to overcome challenges and add value to companies and users. In this study, machine learning is applied to data quality analysis on different datasets, seeking to compare the performance of the techniques according to the dimensions of quality assessment. As a result, we could create a ranking of approaches used, besides a system that is able to carry out automatically, data quality assessment.Keywords: machine learning, data quality, quality dimension, quality assessment
Procedia PDF Downloads 154140 Blockchain for IoT Security and Privacy in Healthcare Sector
Authors: Umair Shafique, Hafiz Usman Zia, Fiaz Majeed, Samina Naz, Javeria Ahmed, Maleeha Zainab
Abstract:
The Internet of Things (IoT) has become a hot topic for the last couple of years. This innovative technology has shown promising progress in various areas, and the world has witnessed exponential growth in multiple application domains. Researchers are working to investigate its aptitudes to get the best from it by harnessing its true potential. But at the same time, IoT networks open up a new aspect of vulnerability and physical threats to data integrity, privacy, and confidentiality. It's is due to centralized control, data silos approach for handling information, and a lack of standardization in the IoT networks. As we know, blockchain is a new technology that involves creating secure distributed ledgers to store and communicate data. Some of the benefits include resiliency, integrity, anonymity, decentralization, and autonomous control. The potential for blockchain technology to provide the key to managing and controlling IoT has created a new wave of excitement around the idea of putting that data back into the hands of the end-users. In this manuscript, we have proposed a model that combines blockchain and IoT networks to address potential security and privacy issues in the healthcare domain. Then we try to describe various application areas, challenges, and future directions in the healthcare sector where blockchain platforms merge with IoT networks.Keywords: IoT, blockchain, cryptocurrency, healthcare, consensus, data
Procedia PDF Downloads 186139 Design of an Instrumentation Setup and Data Acquisition System for a GAS Turbine Engine Using Suitable DAQ Software
Authors: Syed Nauman Bin Asghar Bukhari, Mohtashim Mansoor, Mohammad Nouman
Abstract:
Engine test-Bed system is a fundamental tool to measure dynamic parameters, economic performance, and reliability of an aircraft Engine, and its automation and accuracy directly influences the precision of acquired and analysed data. In this paper, we present the design of digital Data Acquisition (DAQ) system for a vintage aircraft engine test bed that lacks the capability of displaying all the analyzed parameters at one convenient location (one panel-one screen). Recording such measurements in the vintage test bed is not only time consuming but also prone to human errors. Digitizing such measurement system requires a Data Acquisition (DAQ) system capable of recording these parameters and displaying them on one screen-one panel monitor. The challenge in designing upgrade to the vintage systems arises with a need to build and integrate digital measurement system from scratch with a minimal budget and modifications to the existing vintage system. The proposed design not only displays all the key performance / maintenance parameters of the gas turbine engines for operator as well as quality inspector on separate screens but also records the data for further processing / archiving.Keywords: Gas turbine engine, engine test cell, data acquisition, instrumentation
Procedia PDF Downloads 128138 Skills Needed Amongst Secondary School Students for Artificial Intelligence Development in Southeast Nigeria
Authors: Chukwuma Mgboji
Abstract:
Since the advent of Artificial Intelligence, robots have become a major stay in developing societies. Robots are deployed in Education, Health, Food and in other spheres of life. Nigeria a country in West Africa has a very low profile in the advancement of Artificial Intelligence especially in the grass roots. The benefits of Artificial intelligence are not fully maximised and harnessed. Advances in artificial intelligence are perceived as impossible or observed as irrelevant. This study seeks to ascertain the needed skills for the development of artificialintelligence amongst secondary schools in Nigeria. The study focused on South East Nigeria with Five states namely Imo, Abia, Ebonyi, Anambra and Enugu. The sample size is 1000 students drawn from Five Government owned Universities offering Computer Science, Computer Education, Electronics Engineering across the Five South East states. Survey method was used to solicit responses from respondents. The findings from the study identified mathematical skills, analytical skills, problem solving skills, computing skills, programming skills, algorithm skills amongst others. The result of this study to the best of the author’s knowledge will be highly beneficial to all stakeholders involved in the advancements and development of artificial intelligence.Keywords: artificial intelligence, secondary school, robotics, skills
Procedia PDF Downloads 159137 Surprise Fraudsters Before They Surprise You: A South African Telecommunications Case Study
Authors: Ansoné Human, Nantes Kirsten, Tanja Verster, Willem D. Schutte
Abstract:
Every year the telecommunications industry suffers huge losses due to fraud. Mobile fraud, or generally, telecommunications fraud is the utilisation of telecommunication products or services to acquire money illegally from or failing to pay a telecommunication company. A South African telecommunication operator developed two internal fraud scorecards to mitigate future risks of application fraud events. The scorecards aim to predict the likelihood of an application being fraudulent and surprise fraudsters before they surprise the telecommunication operator by identifying fraud at the time of application. The scorecards are utilised in the vetting process to evaluate the applicant in terms of the fraud risk the applicant would present to the telecommunication operator. Telecommunication providers can utilise these scorecards to profile customers, as well as isolate fraudulent and/or high-risk applicants. We provide the complete methodology utilised in the development of the scorecards. Furthermore, a Determination and Discrimination (DD) ratio is provided in the methodology to select the most influential variables from a group of related variables. Throughout the development of these scorecards, the following was revealed regarding fraudulent cases and fraudster behaviour within the telecommunications industry: Fraudsters typically target high-value handsets. Furthermore, debit order dates scheduled for the end of the month have the highest fraud probability. The fraudsters target specific stores. Applicants who acquire an expensive package and receive a medium-income, as well as applicants who obtain an expensive package and receive a high income, have higher fraud percentages. If one month prior to application, the status of an account is already in arrears (two months or more), the applicant has a high probability of fraud. The applicants with the highest average spend on calls have a higher probability of fraud. If the amount collected changes from month to month, the likelihood of fraud is higher. Lastly, young and middle-aged applicants have an increased probability of being targeted by fraudsters than other ages.Keywords: application fraud scorecard, predictive modeling, regression, telecommunications
Procedia PDF Downloads 124136 Grey Wolf Optimization Technique for Predictive Analysis of Products in E-Commerce: An Adaptive Approach
Authors: Shital Suresh Borse, Vijayalaxmi Kadroli
Abstract:
E-commerce industries nowadays implement the latest AI, ML Techniques to improve their own performance and prediction accuracy. This helps to gain a huge profit from the online market. Ant Colony Optimization, Genetic algorithm, Particle Swarm Optimization, Neural Network & GWO help many e-commerce industries for up-gradation of their predictive performance. These algorithms are providing optimum results in various applications, such as stock price prediction, prediction of drug-target interaction & user ratings of similar products in e-commerce sites, etc. In this study, customer reviews will play an important role in prediction analysis. People showing much interest in buying a lot of services& products suggested by other customers. This ultimately increases net profit. In this work, a convolution neural network (CNN) is proposed which further is useful to optimize the prediction accuracy of an e-commerce website. This method shows that CNN is used to optimize hyperparameters of GWO algorithm using an appropriate coding scheme. Accurate model results are verified by comparing them to PSO results whose hyperparameters have been optimized by CNN in Amazon's customer review dataset. Here, experimental outcome proves that this proposed system using the GWO algorithm achieves superior execution in terms of accuracy, precision, recovery, etc. in prediction analysis compared to the existing systems.Keywords: prediction analysis, e-commerce, machine learning, grey wolf optimization, particle swarm optimization, CNN
Procedia PDF Downloads 118135 Derivatives Balance Method for Linear and Nonlinear Control Systems
Authors: Musaab Mohammed Ahmed Ali, Vladimir Vodichev
Abstract:
work deals with an universal control technique or single controller for linear and nonlinear stabilization and tracing control systems. These systems may be structured as SISO and MIMO. Parameters of controlled plants can vary over a wide range. Introduced a novel control systems design method, construction of stable platform orbits using derivative balance, solved transfer function stability preservation problem of linear system under partial substitution of a rational function. Universal controller is proposed as a polar system with the multiple orbits to simplify design procedure, where each orbit represent single order of controller transfer function. Designed controller consist of proportional, integral, derivative terms and multiple feedback and feedforward loops. The controller parameters synthesis method is presented. In generally, controller parameters depend on new polynomial equation where all parameters have a relationship with each other and have fixed values without requirements of retuning. The simulation results show that the proposed universal controller can stabilize infinity number of linear and nonlinear plants and shaping desired previously ordered performance. It has been proven that sensor errors and poor performance will be completely compensated and cannot affect system performance. Disturbances and noises effect on the controller loop will be fully rejected. Technical and economic effect of using proposed controller has been investigated and compared to adaptive, predictive, and robust controllers. The economic analysis shows the advantage of single controller with fixed parameters to drive infinity numbers of plants compared to above mentioned control techniques.Keywords: derivative balance, fixed parameters, stable platform, universal control
Procedia PDF Downloads 139134 Clustering Performance Analysis using New Correlation-Based Cluster Validity Indices
Authors: Nathakhun Wiroonsri
Abstract:
There are various cluster validity measures used for evaluating clustering results. One of the main objectives of using these measures is to seek the optimal unknown number of clusters. Some measures work well for clusters with different densities, sizes and shapes. Yet, one of the weaknesses that those validity measures share is that they sometimes provide only one clear optimal number of clusters. That number is actually unknown and there might be more than one potential sub-optimal option that a user may wish to choose based on different applications. We develop two new cluster validity indices based on a correlation between an actual distance between a pair of data points and a centroid distance of clusters that the two points are located in. Our proposed indices constantly yield several peaks at different numbers of clusters which overcome the weakness previously stated. Furthermore, the introduced correlation can also be used for evaluating the quality of a selected clustering result. Several experiments in different scenarios, including the well-known iris data set and a real-world marketing application, have been conducted to compare the proposed validity indices with several well-known ones.Keywords: clustering algorithm, cluster validity measure, correlation, data partitions, iris data set, marketing, pattern recognition
Procedia PDF Downloads 105133 Exploring Data Leakage in EEG Based Brain-Computer Interfaces: Overfitting Challenges
Authors: Khalida Douibi, Rodrigo Balp, Solène Le Bars
Abstract:
In the medical field, applications related to human experiments are frequently linked to reduced samples size, which makes the training of machine learning models quite sensitive and therefore not very robust nor generalizable. This is notably the case in Brain-Computer Interface (BCI) studies, where the sample size rarely exceeds 20 subjects or a few number of trials. To address this problem, several resampling approaches are often used during the data preparation phase, which is an overly critical step in a data science analysis process. One of the naive approaches that is usually applied by data scientists consists in the transformation of the entire database before the resampling phase. However, this can cause model’ s performance to be incorrectly estimated when making predictions on unseen data. In this paper, we explored the effect of data leakage observed during our BCI experiments for device control through the real-time classification of SSVEPs (Steady State Visually Evoked Potentials). We also studied potential ways to ensure optimal validation of the classifiers during the calibration phase to avoid overfitting. The results show that the scaling step is crucial for some algorithms, and it should be applied after the resampling phase to avoid data leackage and improve results.Keywords: data leackage, data science, machine learning, SSVEP, BCI, overfitting
Procedia PDF Downloads 156