Recently, artificial intelligence technology is taking roots in various sectors as a core technology that will lead the knowledge and information society. Interest in Artificial Intelligence education is growing as it is time to prepare for Artificial Intelligence education in addition to existing information literacy education. However, students who are not computer majors tend to perceive Artificial Intelligence technology as a difficult field, which is why a more efficient Artificial Intelligence education is needed. Therefore, this study reviewed educational programming languages and Artificial Intelligence education methods and proposed an Artificial Intelligence education program that offers Artificial Intelligence learning based on the supervised learning theory of machine learning. As for learning suitability test of the program developed in this study, the program was found to be valid as a result of expert validity analysis and it produced positive results when applied to learners. Based on these findings, the program is expected to be the basis of research on Artificial Intelligence education implementable for non-computer majors who are beginning to learn Artificial Intelligence.
The efficient message routing is highly challenging in terms of low power and lossy networks (loT) for transmission of data with overhead and delay. The protocols used for routing need to be designed such that they should be working efficiently. Efficiency in calculated in terms of energy and delivery of packets. RPL protocol is also designed with the aim of making these two parameters efficient. Even then it contains drawbacks. Trickle algorithm is designed with a goal to reduce the drawbacks in RPL. Trickle algorithm is used in RPL protocols for creation of routes between nodes in the network with different intervals. Unfortunately, there exists some more downsides for the trickle algorithm, which made design of several algorithms inorder to analyse different drawbacks. In this paper, on analysing different types of trickle algorithms and locating the drawback in every algorithm, a novel algorithm is designed which helps in reduction of the drawbacks that are found. The description of this algorithm along with the simulation results done using Cooja 3.0 simulator is also discussed in this paper. The Simulation of the algorithm that is newly designed is done by assuming a network with different count of nodes and comparing the results with the previously introduced Trickle algorithms.
Big data defines the state where the size, speed and kind of data go beyond a memory or executing capabilities for precise and timely decision-making. Big data analytics is integrated with ML and statistical methods for processing big data and recognizes the important data. At present times, the generation of online product reviews has exponentially increased at each and every second. These applications have resulted in developing the volumes of data which can be used for prediction and classification for decision making process. Compared with other models, various techniques are applied in solving the big data problem, feature selection (FS) is known to be an efficient method. FS operations could be exploring with the application of a subset of features which is related to the topic of précised definition of the existing datasets. Deplorably, search using this type of sub-sets results in the problems of combinatorial as well as maximum time consuming. The meta-heuristic approaches are typically employed to facilitate the choice of features. This paper presents an optimal extreme learning machine (ELM) based binary particle swarm optimization to precede the FS process. The proposed method develops a Fitness Function (FF) by applying ELM. And the best solution of the FF has been explored under the application of BPSO technique. For instance, the dataset of product review which are derived from Amazon including synthetic data, which is comprised with total of 235,000 positive and 147,000 negative review records is used. The experimental result implied that the ELM-BPSO technique is comparably best
In this paper, it is described a new design of a digital Lock-In amplifier applied to 4-Kelvin- probe electrodes for the measurement of complex electrical variables. The proposed design is based on the operation of a Phase Sensitive Detection (PSD) circuit and on signal acquisition by the Nyquist principle. The hardware basically consists of a programmable embedded system and an analog interfacing circuit. The microcontroller within the circuit was programmed using standard C language for portability and performs the acquisition of the resulting signal along with mathematical operations. Experimental tests on the prototype have shown that it performs as theoretically predicted.
We propose a new generated family of distributions that is known as the generalized alpha-power transformation-G (GAPT-G) family of distributions. The new class of family can be more flexible since the density shapes are left skewed, symmetrical and reversed-J. Some special models derived and discussed. Several of its important properties are derived. The maximum likelihood equations are derived for GAPT-G family parameters. The importance and flexibility of the derived models is assessed using one real dataset.
In the manufacturing industry, there is a transition to multi-item production for reinforcement of competitiveness. Therefore, the hybrid manufacturing technology is increasing. Especially, many efforts in production quality improvement are made through the adoption of the Manufacturing Execution System and ERP, so it is necessary to operate MES for prompt and effective management. Therefore, the adoption of MES is necessary. MES should improve ineffective parts in production activities while managing all stages related to production of products. If there is change in the process, the changed items should be reflected to the system. However, most manufacturing execution systems are operated passively and repetitively by system administrators. This study presents a model that system administrators can comprehensively apply reference information about production related requirements on specific line’s equipment to the same equipment of other lines. To practice the model, demands of MES administrators were organized, the current system’s screen is analyzed, data for applicable reference information are classified, the screens for blanket and selective applications are designed and realized, the reference information is registered to same equipment in all the lines in case of blanket application. The reference information is registered to selected lines in case of selective application. In the results of the presented model’s test, at first it wasn’t 100% successful due to exceptional situation that the wrong data exist in the database. The validity check function for the original data that causes the exceptional situation was added, which was notified to the system administrators so that they can correct the wrong original data. The equipment reference information blanket application was implemented, and it was 100% data process accuracy. In addition, the flexible response for application to production lines is possible thanks to the division of blanket application and selective application of reference information.
In this article we examine the weighted average finite difference methods to approximate the fractional variable-order wave equation, where, the order of the differentiation can be a function of time. The fractional differentiation of the of variable-order are described in terms of the Riemann-Liouville concept. The stability of the utilized method are proved by using a kind of Von Neumann analysis. To reveals that the method is effective, two examples are offered. and the obtained solutions were compared with the exact solutions.
In the digital world, it’s somewhat difficult to manage the expectations of the present customers especially E-Commerce Shoppers who are often challenging and pushed to keep up with what is anticipated to them. Numerous E-shoppers are under pressure to locate and approach to convey an offline in-store ride to the online store ride of higher level interaction with their customers. Experts have cited that all kinds of powers are clearly in the fingers of customers and that their never ever change their behaviors, this leads to force them to approach various online shops portals to adapt. Online shopping comes to be greatly empowered, the E-shoppers expectations from retailers to vendors and are developing a rapid shift in their behavior with an emphasis on pace and convenience. Shopper’s interest spans have become shorter, with 50 percent of shoppers leaping from one retailer website to another, earlier they making an enduring purchase. In addition, greater than eighty percent of shoppers are taking their shopping experience in both online as well as offline. A retailer’s internet site is one of the key places where customers go to search about their dream product’s records and pricing. The demand for the quick shipping of products purchased and handy services, such as click on online and collect in store, as properly as seamless returns have led to groups developing easy, no-fuss fee options. Research indicates that primary shops will continue to offer higher approaches to gather products purchased online from their stores, as a way of appealing to customers and supplying larger convenience. Based on these prevailing conditions a study was performed on E-Commerce components impacting consumer’s online purchasing conducted on the United Arab Emirates (UAE). Major portions of primary data collection were done via the questionnaires method and by means of emails in two predominant cities of UAE (Sharjah, Abu Dhabi). Cost, time saving (shipping), maintaining customers private details in a secured manner and comfort while do shopping were recognized as vital elements and this leads to assured buying conduct of online shopping. The www is to remodel in the region of populace where communal networked group impact and leads to online shopping.
Recently, research on methods to improve upon the security vulnerability of password based authentication is being actively conducted. Research on how to use real-time biometric recognition is also actively underway. The system development of Public Key Infrastructure (PKI) key pairs and uses a method that safely stores the produced keys in secure zones, and encryption key data is composed so that access can only be made in Fast Identity Online (FIDO) Authenticator zones. Face certification data extracted from the face recognition engine is created, data can be accessed and saved through encryption. With the Fast Identity Online (FIDO) Authenticator, a method was developed that compares and face certification data. The system in this thesis developed a Fast Identity Online (FIDO) based face certification system that uses biometric authentication to supplement the weaknesses of password based systems which were used for user certification in the past because they cost little and are convenient. System development in this thesis was conducted by applying a Fast Identity Online (FIDO) based certification system to an Active Shape Model (ASM) style face recognition algorithm. After taking a photo of the face with the user’s cellular phone and saving it, certification is carried out through comparisons made of the photo and real-time input images. Once certification is made using Fast Identity Online (FIDO) protocols, tests of face detection speed and matching speed are conducted to measure accuracy and speed. A total of 30 tests were conducted and test results showed that detection speed was an average of 30.35 f/s and matching speed was 14.16 ms. In this paper, biometric authentication security and convenience are provided by using security functions provided by user devices and supplementing them with a server operation method.
VR 360 Cam is an emerging device. By combining this with the rising webtoon industry, we want to show people an immersive webtoon. Based on the python language, face detection was performed from images received in real time from VR 360 Cam through dlib, a machine learning library that supports python. The VR 360 Cam performs trekking on the detected face to receive each detected position value, and is converted into a natural face through rectification to be shown to the audience. The exhibition, which performed face detection from the VR 360 Cam, and showed the image of the person’s face mapped to the audience, drew meaningful results. Unlike cameras such as webcams, VR 360 Cam has a wider viewing angle, allowing more people to interact. Existing webcams can only interact with one person at a time because it is impossible to interact with more people due to a narrow angle when one person enters. On the other hand, interaction with multiple people is possible through VR 360 Cam. Various exhibitions were possible.
Background/Objectives: An auxiliary device was developed using 3D printing and its usefulness and quality performance were evaluated to improve the speed, stability, and accuracy of plain X-ray exams that require changing the central axis of the human body according to anatomical positions. Methods/Statistical analysis: The device was fabricated by assembling commercially available parts after printing the design model with a 3D printer. In terms of performance, the stability of the device was evaluated by tensile tests with a Universal Testing Machine, and the speed and accuracy were evaluated by simulations. Findings: The developed auxiliary device enabled patients to maintain accurate positions at any location on the examination table and provided features such as 360° rotation and fast and easy attachment and detachment. The tensile strength of the auxiliary device was 45.5 kgf at a 67 mm displacement in the horizontal direction and 73.3 kgf at a 47.6 mm displacement in the vertical direction, indicating the stability of the device through quantitative figures. Improvements/Applications: The developed auxiliary device showed high clinical usefulness by enabling patients to change positions more accurately and stably than conventional fixed devices during plain X-ray exams according to changes in the central axis.
Cybercrime characteristics in cyberattack environment is malicious tendency. Therefore, in this paper, we describe the malicious nature of the attacker who caused the cyberattack. This paper focuses on the cultural values or prejudices that individuals and society give to the malicious tendencies of attackers who cause cyberattacks. First of all, in this paper, we research about the related issues, including cybercrime and cyberattack spread. This study interested in maliciousness which affecting cyber security incident and approached the issue of how integrity, rationality and benevolence will affect cyberattacks. So these attributes have found relevance to money gains or extortion cyberattacks, accidental cyberattacks and persistent cyberattacks. Also, we wanted to find a point of discussion about how the character traits of integrity, rationality, and benevolence affect the cyberattack. How these properties are linked to these cyberattacks and what trends these characteristics would have if they were assumed to occur with an arbitrary probability distribution were studied. Human maliciousness affects cyberattacks. What effect will human good character have on cyberattacks? If a cyber attacker had a sense of integrity, how would this attacker be affected by financial gains or extortion attacks? Usually, people with a high sense of integrity tend to have less or less desire for money or profit taking than those who do not have a sense of integrity. Therefore, it can be judged that a person with a high consciousness has a relatively low probability of a takeover attack targeting financial interests, etc. In this paper, There are used these factors such as sense of charity, impulsive rationality, integrity for probabilistically analysis. We can have the following insights: If the integrity is low, it is more likely to pursue money interests and attack them. Attackers with high impulsive emotions have a high chance of accidental attack, while those with high self-interest have a low chance of sustained attack. In this paper, we considered from a probabilistic point of view, focusing on how the propensity to cause cyberattacks relates to benevolence, integrity and impulse consciousness. This is because cyberattacks are determined by right value system as a moral judgment of human beings. If right value system determines that attack is right, it will cause an attack, and if it is wrong, it is highly likely that it will not cause an attack. This research result could be used as a basis for the probabilistic study of human characteristics in the relevance of cyberattacks and human maliciousness in the future.
The robust segmentation of color images in a natural environment without specific constraints such as lighting or background is very important in the field of image processing and computer vision. In this paper, an environmentally adaptive image segmentation method using color invariant is proposed. The proposed method introduces a number of color invariant, such as W, C, U, N, and H, and automatically detects factors in the surrounding environment in which images such as lighting, shading, and highlights are taken. The image is then effectively split based on the edge by selecting the color invariant optimal for the detected environmental factors. In the experiment, we implemented the proposed edge-based image segmentation algorithm. Various image data taken in general environments without specific constraints were utilized as input images of the suggested system. In this study, various kinds of color images taken in different environments were tested, and each color invariant was extracted from the experiments that best expressed the environmental changes around them. As a result, a largest number of images were determined to have a change in the intensity of lighting, followed by highlights and shadows. In addition, there were a few images that determined that no special state environmental changes existed. As the results of the experiment show visually, the existing method did not correctly remove shadows and did not detect some areas of the circular shape. In addition, the existing method can also be found to be partially inaccurate in edge detection in many areas. On the other hand, the proposed method confirmed stable segmentation of images. The proposed color invariant-based image segmentation algorithm is expected to be useful in various pattern recognition areas such as face tracking, mobile object detection, gesture recognition, motion understanding, etc.
This study shows O2O (offline to online) based studying abroad platform, we called it as DRM (DoDream), that provides foreign students with various information for studying in Korea and matching platform with Korean universities (Language Institute) (Lim, J.H., et al., 2019. The architecture design of students’ career information management system using blockchain technology. International Journal of Innovative Technology and Exploring Engineering, 8(8S2), pp.188–193). Empirical results reveal studying abroad provides a new environment for foreign students who dream of studying in Korea by innovating structural problems of existing studying abroad market and sharing open information about Korea n universities, and to provide systematic and efficient study abroad support services to Korean universities (Language institutes) and local study abroad. DRM (DoDream) Tokens will be otherwise intended to be generally obtainable by users of the DRM (DoDream) Platform through the methods externally from third party cryptocurrency exchanges or internally on the DRM (DoDream) platform through conversion of “step” points that are awarded to users who have participated in events, promotions and/or who have made positive contributions to the DoDream Platform community.
Since the visible light communication (VLC) has to perform the functions of communication and illumination at the same time, a method for communication as well as lighting is needed. In this paper, when text data is transmitted through visible light communication, the level of illumination dimming according to the frequency of occurrence of alphabets in sentences is analyzed, and a method of improving the dimming level and Bit Error Rate (BER) performance through error correction codes generated during data transmission is studied. Visible light communication systems must perform both communication and lighting functions, so not only communication but also a method for the lighting role is needed. When transferring data (transfer text), the frequency of occurrence of alphabets in sentences is different. Depending on the frequency of occurrence of these alphabetic characters, if there are many ‘0’s in the code to be transmitted, the dimming level will be lowered and flicker will occur. Also, when a 1-bit error occurs, the alphabet code itself is changed. To solve this problem, an error correction code using parity bits has been added. Through this, it was confirmed that the overall dimming level and Bit Error Rate (BER) performance were improved. Also, in visible light communication, the function of lighting is closely related to the performance of the overall system. As we have seen above, when there is a continuous zero period, the function of the lighting is severely degraded. This reduces the performance of the entire system, not just the lighting. Therefore, the dimming level and BER performance were improved by improving the performance through the algorithm and error correction code to improve the overall dimming level.
Background/Objectives: This paper deals with the handling method of incomplete training data that is partially lost. It is unavoidable in a ubiquitous environment that deals with data collected from multiple devices or from a long distance, but it is difficult to expect good results due to loss of information if the lost part is discarded. Methods/Statistical analysis: Various algorithms have been proposed to solve this problem. Among them, the algorithm for learning by changing the format of the training data to fit incomplete data has been applied to various problems and has shown good results. This data format conversion method is called a data expansion technique, and has two characteristics: the importance can be adjusted differently for each event, and a probability value can be assigned to each cardinality of each variable. Findings: The second feature, the ability to assign a probability value, was used to assign a compensation value for loss data. The first attempt to do this was to assign equal probability values to the loss values. In the classification algorithm using the entropy function, the variable containing more loss values is not selected in the upper node. However, the method of allocating the equal value had the point that the original information is ignored. Therefore, a method of obtaining the entropy probability with complete information excluding the loss value and filling it in the loss value was also proposed. This paper starts with the basic idea that this method is further developed and the lost information can be found in the area classified by the classification algorithm. In terms of implementation, the training data is divided into two parts: lost and non-lost events, and then classified using the C4.5 classification algorithm with the data that is not lost, and each classification area is obtained. Then, by using the information remaining in the event of the loss data, the classification area is sequentially traversed according to a predetermined algorithm to find the area closest to the loss event. And after expressing the value of this area as a probability, the loss value is replaced with this probability value, which means compensation for the loss value. Improvements/Applications: After compensating for the total loss values, the recovered training data is learned with one of the SVM algorithms in order to evaluate the performance of how much information leakage has been preserved, and its performance is compared. As a result of the experiment with different degrees of loss for each variable, it was confirmed that the loss of information can be minimized and used as a method of meaningful compensation.
Currently, many researchers are working on stock price prediction system by using deep learning algorithms. Stock market is completely random, and there is no pattern. Even though, a pattern in stock market could be found, it will not be last for a long time because the stock market will adopt a new situation and the strategy is no longer available on already changed stock market. There are many auto trading programs such as a trading bot on stock market. However, they are literally trade stocks based on human’s direction or rules. It will not affect any changes, and it keeps working as what rules are set up from the initial status on the stock market. Stock price depends on volume of total sales, stock news, revenue, total asset, big buyer’s position and so on. There are many aspects for affecting stock price, and it changes all the time. Therefore, it keeps monitoring stock market and makes a decision whether buy or sell at the right time for earning profits. This research uses Bidirectional Long Short-Term Memory (BLSTM) to predict stock price in the near future. BLSTM is more accurate than LSTM which is one directional. In addition, stock market is like a living creature. Data to manipulate stock price must be inputted and analyzed consistently. Therefore, stock price can be predicted by consistent monitoring with BLSTM.
Background/Objectives: Today, the deep neural network has more intermediate layers along with learning data and has gradually become larger to draw a lower result due to the bound conventional neural network. Methods/Statistical analysis: Therefore, in this paper, we proposed the line-segment feature extraction for input dimensionality reduction in pattern mining. The proposed algorithm extracts the line segment information, constituting the image of input data, and assigns a unique value to each segment using predefined filters. Using such unique values to identify the number of line segments, create a one-dimensional vector, in the size of 256 or 512. Findings: This vector is used as the input data for a multi-layer perceptron (MLP). For performance evaluation, LFA was compared with principal component analysis (PCA). As a result, LFA-256 (at 96.81% accuracy) had −0.1% lower performance but a faster speed than PCA (96.91% accuracy), and LFA-512 (at 97.25% accuracy) had 0.35% higher performance than PCA. Improvements/Applications: We will study the recognition service possible to use in industry (mobile devices, PC, edge computing, etc.) and real-life through this algorithm.
The fading phenomenon in mobile communications is mainly a stochastic process and plays an essential role in the reliability and stability of the systems. The frequency bands currently in use it is reasonably well characterized and described. With the use of mmW, only experimental, theoretical models are being proposed; there is no unanimity regarding a model that effectively contemplates the phenomenon in this spectrum range. Some studies suggest the traditional models applied to the frequencies in use; Rician and Rayleigh are the main ones. This study aims to advance and compare, with accepted models, a broader statistical model that overcomes the limitations of the current models. It will be analyzed and compared a model using the Nakagami-m distribution covering a greater range of possibilities, given the low spreading capacity of mmW and the need for links almost always LOS situation in which the Rician model fails.
The 5G mobile wireless network systems faces a lot of security issues due to the opening of network and its insecurity. The insecure network prone to various attacks and it disrupts secure data communications between legitimate users. Many works have addressed the security problems in 3G and 4G networks in efficient way through authentication and cryptographic techniques. But, the security in 5G networks during data communication was not improved. Subtractive Gradient Boost Clustered Node Authentication (SGBCNA) Method is introduced to perform secure data communication. The subtractive gradient boost clustering technique is applied to authenticate the mobile node as normal nodes and malicious nodes based on the selected features. The designed ensemble clustering model combines the weak learners to make final strong clustering results with minimum loss. Finally, the malicious nodes are eliminated and normal mobile nodes are taken for performing the secured communication in 5G networks. Simulation is carried out on factors such as authentication accuracy, computation overhead and security level with respect to a number of mobile nodes and data packets. The observed outcomes clearly illustrate that the SGBCNA Method efficiently improves node authentication accuracy, security level with minimum overhead than the state-of-the-art-methods.
