Yükleniyor…
Yükleniyor…
18
Arşivlenen Tez
0
DOI Atanmış
0
Araştırmacı
0%
DOI Oranı
Understanding how to forecast a product's sales and demand is crucial for businesses that sell goods. Knowing how much demand will be in a given time gives them many benefits and gains. Many methods have been developed and used for demand forecasting from past to present. If we divide the methods used into two, traditional and machine learning methods are used for demand forecasting. We can say that traditional methods have left their place to machine learning due to less and slow data processing. Machine learning methods have the ability to process a lot of data faster and analyze the data it uses and provide a more accurate prediction by identifying hidden patterns in the data. The problem here is that there is no one "onesize-fits-all" prediction algorithm. Typically, demand forecasting features consist of several machine learning approaches. Therefore, the choice of machine learning models depends on many factors such as business goal, data type, data quantity and quality, forecast time. Therefore, the main problem here is to determine which algorithm will be used with which parameters. In this study, different machine learning methods and parameters was used and compared to select the most suitable machine learning algorithm and parameters according to the selected data set and provide more accurate predictions. Algorithms such as time series, linear regression, random forest was studied and external factors such as seasonal, regional and economic factors was used as parameters. The algorithm with the best results will be chosen from models with or without external factors. Keywords: Machine Learning, Demand Forecasting, Regression, Time Series
This thesis aims to explore the influence and advancements of Artificial Intelligence (AI) in education, particularly focusing on Intelligent Tutoring Systems (ITS) for programming education. A comprehensive review of existing ITS models will be carried out, examining their implications on academic performance, student engagement, and motivation. Building upon this analysis, the research will conceptualize and design an innovative ITS, prioritizing interactivity, enhanced code evaluation, and accessibility. While empirical testing of this tool will not be within the scope of this study, a theoretical comparison against existing non-programming chatbot tutors and other ITS will be made. This work is expected to provide substantial contributions to educators, policymakers, and researchers interested in the intersection of AI and programming education. Keywords: Artificial Intelligence, Intelligent Tutoring Systems, Programming Education, Interactive Learning, Educational Technology, Student Outcomes
Agile approaches have become a significant force, transforming several industries, including software development and Information Technology (IT). Agile is a set of values and principles that guide the delivery of a quality and satisfactory software project for the customer, using an adaptive, incremental, and iterative way of working by cross-functional and self-organized teams. This study delves into the complex world of Agile methodologies, clarifying the responsibilities, lifecycles, benefits, and drawbacks. The design and development of a new software solution as a design case called an Employment System (EMS), using the Agile Scrum software development paradigm. The Software Usability Measurement Inventory (SUMI) is used in this study to gauge users' opinions of the usability and quality of the EMS. The SUMI survey data are subjected to meticulous statistical analysis to produce verifiable insights. The EMS outperforms the global average usability criteria, earning a score of 53.67 in the SUMI evaluation of the usability of the EMS produced within the Agile Scrum Software Development paradigm. These findings confirm the EMS's efficiency in meeting fundamental functional needs and highlight its high level of user satisfaction. This study has significant ramifications, especially for software development teams striving to improve system usability and user experiences. Keywords: Agile, Scrum, Agile Scrum Software development, Employment System Design, Software usability measurement, SUMI (Software Usability Measurement Inventory), Agile project management.
Telemedicine has witnessed rapid growth, especially in the context of medical tourism. This systematic literature review explores the security and privacy challenges associated with the integration of cloud-based telemedicine in the realm of medical tourism. The study employs the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) method to ensure a comprehensive and transparent review process. This review extensively scrutinizes scholarly articles, reports, and case studies published in reputable journals and conference proceedings from 2010 to 2024. With an in-depth assessment of over 50 papers, research highlights the advancements in telemedicine for medical tourism while shedding light on the critical issues related to data security and patient privacy. The identified challenges encompass regulatory compliance, information vulnerability, and ethical considerations. In-depth analysis of various scholarly works provides insights into the existing gaps and areas requiring attention. To address these challenges, this thesis proposes IT-based solutions aimed at enhancing the security posture and privacy measures in cloud-based telemedicine systems catering to medical tourism. These solutions encompass robust encryption protocols, secure data storage practices, and compliance frameworks tailored to the unique nature of medical tourism. In summary, this systematic literature review not only unveils the complexities surrounding the intersection of cloud-based telemedicine and medical tourism but also offers IT-based solutions to fortify the security and privacy aspects, ensuring a safer and more reliable healthcare delivery system.
The rapid improvement of technology among technical devices such as smart phone, provide more learning opportunities beyond the classroom in educational environment. However, beside these advantages and effects of smartphone use on educational setting, there are disadvantages, which have negative effect on learners. One of which is a phenomenon known as nomophobia. This is a psychological condition regarded as fear of being without mobile phone or not being able to access the Internet on mobile phone. The aim of this study is to investigate an assessment of nomophobia situation among IT students in Eastern Mediterranean University (EMU). The research method of the study was quantitative survey approach using Nomohpobia Questionnaire (NMP-Q). To reach this aim, 205 questionnaires were gathering among Information and Technology (IT) students from four different age group and two diverse genders. The gathered data of the study is analyzed with descriptive statistics, mean, frequency standard deviation, percentage, t-Test and Anova. The Findings revealed that participants have almost high-level nomophobic behavior and female participants have more predisposed to nomophobic situations in comparison with male participants and the students between ages 21-25 struggled more with the effects of nomophobia when compared to the other age groups. Additionally, according to t-Test analysis, nomophobia is significant on gender of IT students. Moreover, consistent with Anova analysis, nomophobia has significance on age of participants.
The spread of hate speech on social media platforms is a problem that is constantly becoming more imminent as the access to related technologies gets easier. This study focuses on detecting hate speech on an imbalanced multiclass twitter dataset using Machine Learning (ML) algorithms. The most commonly used ML algorithms namely, Logistic Regression, Support Vector Machines (SVM) and deep learning systems Gated Recurrent Unit (GRU), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Bi-directional Long Short-Term Memory (BiLSTM) and a hybrid model CNNBiLSTM have been used for hate speech detection. In order to overcome the problems that arise from using an imbalanced dataset several techniques are used to balance the dataset, Synthetic Minority Oversampling Technique (SMOTE), SMOTETomek, SMOTEENN, Adaptive Synthetic (ADASYN), class weights and the proposed method. Each classifier was trained with all data balancing techniques and their performances were compared in order to find the best classifier for classifying hate speech in the dataset. The best classifier was CNN using the proposed method and it had an F1-score of 0.96 with a Cohen Kappa score of 0.94 and an overall Recall and Precision score of 0.96. For the best system, the recall and precision scores for the hate class was 1.00 and 0.94 respectively.
Communication has been a big part of how our species has changed since the beginning of time. So much that today, it is hard to imagine running a business or living a normal life without a language that everyone speaks (Freitag et al., 2021). The use of mechanical dictionaries to bridge the barriers between languages was first suggested in the 17th century (Hutchins & Somers 1992), and these technologies have had a transformative impact on communication, the mode of information sharing and access globally especially with frequent cross-cultural communication among people from countries and regions. Machine translation (MT) is simply automatic translation. Systran (2004) describes machine translation as a process that uses computer software to convert text from one language to another. It is the process of translating words from one natural language such as English to another such as Turkish using computer software. Translation software have gained popularity given that they provide a useful environment to facilitate and manage translation projects. This research includes a review of existing translation approaches, it sheds light on the current state of machine translation technology and its impact on the translation industry. The study further explores the nature of the translation process assisted by software and implements a model which will be tested from the end user`s perspective for effectiveness using Software Usability Measurement Inventory.
Mobile Ad Hoc Networks (MANETs) have been applied in many different fields in recent years. Although MANETs are highly vulnerable to malicious behavior, complete security is complicated to achieve. Due to the insufficiency of prevention techniques, the Intrusion Detection System (IDS), which monitors system activity and detects intrusions, is generally used with other security measures. Denial of Service (DoS) type attacks such as flooding, blackhole, and grayhole attacks are acute types of network intrusion that aim to make computer/network resources unavailable to legitimate users. Intrusion Detection (ID) is a security management system that serves as an alarm mechanism for any computer network such as MANET. It detects the incoming security threats to a network and then issues an alarm message to an entity to take needed actions against the intrusion. An IDS gathers and examines information from numerous areas within a computer or a network to identify possible security breaches, including intrusions (attacks from outside the organization) and misuse (attacks from within the organization). The goal of this study is to develop a multistage ID technique for detecting flooding, blackhole, and gray-hole intrusions using Support Vector Machines (SVM). The SVM mechanism supports binary classification and separating data points into two classes. Hence, in this research SVM approach is used for classifying and detecting multiple attacks after breaking down the multiclassification problem into numerous binary classification problems. Keywords: Mobile Ad-hoc network, Support Vector Machine, On-demand Distance Vector, black hole, grayhole, flooding
The future of this world, the technology which is going to be built will evolve around ideas which will make the world a better place for us humans and for the generations coming to accept the basics of the nature and enhance our relationship to this world. Online (eCommerce) and offline retail are becoming less and less distinct (Bauer & Garaus, 2018). However, one benefit of eCommerce is that it allows for a quicker transaction because many buyers publish reviews on websites like amazon.com or ebay.com for the majority of the items. In this study, we have aimed to expand the world of AI and data recognition. The way retail establishments work is changing as a result of artificial intelligence (AI). In my vision, the future of physical retail will be AI-Powered Automated Stores. Consumers at these retail establishments deal with totally automated technologies. As a result, it's crucial to carefully consider the factors that led to consumers' decision to visit AI Powered Automated Retail Stores. This study explores this field in an effort to identify the factors that indicate consumers' inclination to visit AI-Powered Automated Retail Stores. Using the help of artificial intelligence, we can improve in much higher pace. One of the greatest major that AI is playing a significant role is ecommerce and digital marketing (Berman, 2019). The article's goal is to examine various technologies that may be used to enhance customers' indoor shopping experiences. As a result, we want to investigate several strategies that, when combined with consumer sentiment analysis, can shorten the time customers take to pay for goods and personalize their shopping experience. Also, it will be explaining the potential of integrating AI and blockchain and providing the world with a better technology and more secure and trustable fields to operate upon. Blockchain will ensure the security of every network and optimize each transaction to its maximum and reduce every possible obstacle and wastes in its path (Bauer & Garaus, 2018). Based on the experiences of companies that accept cryptocurrency payments, this study will also seek to identify the process and affective metrics in adopting cryptocurrencies for purchasing. Consumers and retailers prefer cryptocurrencies since they are a quick, secure, and affordable form of international payment. However, it has been shown that some consumers are completely ignorant of the possibility of making purchases using cryptocurrency. Some consumers prefer not to utilize cryptocurrency as payment, either due to limited technology utilization or a lack of faith in the system (Liljander, Gillberg, Gummerus, & Riel, 2006).
The usefulness of neural techniques in computer-assisted translation (CAT) is thoroughly examined. The development of neural networks has significantly improved machine translation and other aspects of natural language processing (NLP). By using computer aid, CAT is an NLP task that seeks to increase the effectiveness and caliber of human translation. However, because standard statistical-based approaches lack semantic comprehension, CAT's usefulness has been constrained. With their capacity to recognize intricate linguistic structures, neural networks have demonstrated significant promise in overcoming this restriction. Comparison research between neural and conventional statistical-based methods was done to ascertain the effectiveness of neural approaches in CAT. Two datasets were employed in the investigation, one with technical and scientific documents and the other with legal words. The outcomes demonstrated that in terms of accuracy, fluency, and overall translation quality, neural-based models beat the conventional statisticalbased methods. The neural models were very good at translating colloquial language and handling complicated sentence patterns. The study also examined how many variables, including corpus size, language pairs, and training methods, affected the effectiveness of neural-based models. The results showed that the performance of neural-based models can be greatly enhanced by using larger corpus sizes and properly chosen training approaches. The results of this work show the potential of neural approaches in enhancing computer-assisted translation efficacy and emphasize the significance of taking numerous elements into account when creating and training neural-based CAT systems. Keywords: neural machine translation, computer-assisted translation, CAT system, out-of-vocabulary words, rare words, translation effectiveness, machine learning, natural language processing, deep learning, neural networks, language modeling.