-
Heart Disease Prediction Using Machine Learning Techniques
Issue:
Volume 5, Issue 3, September 2022
Pages:
146-154
Received:
27 June 2022
Accepted:
13 July 2022
Published:
20 July 2022
DOI:
10.11648/j.ajcst.20220503.11
Downloads:
Views:
Abstract: Machine learning and artificial intelligence have been found useful in various disciplines during the course of their development, especially in the enormous increasing data in recent years. It can be more reliable for making better and faster decisions for disease predictions. So, machine learning algorithms are increasingly finding their application to predict various diseases. Constructing a model can also help us visualize and analyze diseases to improve reporting consistency and accuracy. This article has investigated how to detect heart disease by applying various machine learning algorithms. The study in this article has shown a two-step process. The heart disease dataset is first prepared into a required format for running through machine learning algorithms. Medical records and other information about patients are gathered from the UCI repository. The heart disease dataset is then used to determine whether or not the patients have heart disease. Secondly, Many valuable results are shown in this article. The accuracy rate of the machine learning algorithms, such as Logistic Regression, Support vector machine, K-Nearest-Neighbors, Random Forest, and Gradient Boosting Classifier, are validated through the confusion matrix. Current findings suggest that the Logistic Regression algorithm gives a high accuracy rate of 95% compared to other algorithms. It also shows high accuracy for f1-score, recall, and precision than the other four different algorithms. However, increasing the accuracy rates to approximately 97% to 100% of the machine learning algorithms is the future study and challenging part of this research.
Abstract: Machine learning and artificial intelligence have been found useful in various disciplines during the course of their development, especially in the enormous increasing data in recent years. It can be more reliable for making better and faster decisions for disease predictions. So, machine learning algorithms are increasingly finding their applicat...
Show More
-
iCARE-LUX: Instant Cloud Archive Repository Express for Leading-Edge User Experience
Sheldon Liang,
James Pogge,
Melanie Van Stry
Issue:
Volume 5, Issue 3, September 2022
Pages:
155-169
Received:
25 June 2022
Accepted:
18 July 2022
Published:
29 July 2022
DOI:
10.11648/j.ajcst.20220503.12
Downloads:
Views:
Abstract: iCARE-LUX or instant Cloud Archive Repository Express has emerged for leading-edge user experience through algorithmic machine learning that is involved in more and more aspects of daily life through cloud-based content management and delivery (CMD). iCARE acts like a “fastlane” to bridge the gap between DATA and wiseCIO where DATA stands for digital archiving (via transformed analytics), and wiseCIO for web-based intelligent service (engaged with cloud intelligence outlet). iCARE incorporates DATA and wiseCIO into a triad that best serves CMD for leading-edge user experience via algorithmic machine learning. This article presents the archival repository express as a “fastlane” to liaise with human-computer interfacing by providing mathematical and computational solutions to distributed and cloud-based problems. Leading-edge user experience or LUX is user-centric with “luxury” pleasure and inexpensive advantage offered to users while browsing cloud-based content and exploring usable intelligence in support of decision-making. iCARE in collaboration with DATA and wiseCIO establishes a triad of content management and delivery (CMD) as a whole that harnesses rapid prototyping for user interface design without explicitly cording required and propels leading-edge user experience by cohesive assembly from Anything orchestrated as a Service (XaaS). Basically leading-edge user experience makes end-users centered without often webpage swapping during browsing in hierarchical depth via “in-&-out” interactivity, and exploring in contextual breadth via self-paced spontaneity. Furthermore, iCARE or instant cloud archival repository express creatively incorporates express tokens for information interchange (eTokin) into the CMD triad for integral content under managed with trivial information eliminated. In particular, by exploiting eTokin, iCARE promotes seamless intercommunications in-between and empowers users to be QUINARY professionals cohesively, such as queryable agent, ubiquitous manager, interactive expert, novel designer, and available integrator. More importantly, iCARE uses algorithmic machine learning to coordinate instant publishing over DATA, assemble efficient presentations to end-users via wiseCIO, and aggregate diligent intelligence for business, education and entertainment (iBEE) through elastic process automation.
Abstract: iCARE-LUX or instant Cloud Archive Repository Express has emerged for leading-edge user experience through algorithmic machine learning that is involved in more and more aspects of daily life through cloud-based content management and delivery (CMD). iCARE acts like a “fastlane” to bridge the gap between DATA and wiseCIO where DATA stands for digit...
Show More
-
Risk Management Information Technology Based on ISO 31000:2018 at Institute of Philosophy and Creative Technology, Ledalero
Maria Florentina Rumba,
Robertus Mirsel,
Fransiskus Xaverius Sabu
Issue:
Volume 5, Issue 3, September 2022
Pages:
170-177
Received:
6 July 2022
Accepted:
29 July 2022
Published:
5 August 2022
DOI:
10.11648/j.ajcst.20220503.13
Downloads:
Views:
Abstract: Risk is defined as a state of uncertainty, where an undesirable situation occurs and causes a loss for an agency. Therefore, risks need to be managed properly. Risk management is all activities to manage risks or threats that can occur in an agency. One of the standard risk management tools is the ISO 31000:2018. There have been many studies that present how to analyze IT risk management in an agency using the ISO 31000:2018 framework with various methods. From the many articles on risk management in an institution or organization, this framework becomes a reference for analyzing IT risk management in higher education institutions. This research is a case study conducted at the Institute of Philosophy and Creative Technology (IPCT) at Ledalero. The IT risk management analysis work process used is ISO 31000:2018. The methods used in this study were interviews given to the head of the IT division, direct observation, and an open questionnaire given to all work units at IPCT. The purpose of this research is to identify IT assets, identify risks and their impacts, analyze, and treatment risks. The results of this study indicate that the risk impact of 28 elements which is the elaboration of 3 main factors, namely 2 elements are in the Low-Medium category with a scale (0.36-0.42), 5 elements are included in the Medium-Low category with a scale range (0.25-0.34), 7 elements are included in the Minimum-Low category with a scale range ((0.00-0.14), and the most are in the Low-Low category with a scale range (0.15-0.24) which is as many as 14 elements.
Abstract: Risk is defined as a state of uncertainty, where an undesirable situation occurs and causes a loss for an agency. Therefore, risks need to be managed properly. Risk management is all activities to manage risks or threats that can occur in an agency. One of the standard risk management tools is the ISO 31000:2018. There have been many studies that p...
Show More
-
Hybridized Cryptography and Cloud Folder Model (CFM) for Secure Cloud-Based Storage
Anietie Ekong,
Odikwa Henry,
Abasiama Silas,
Imou Douglas
Issue:
Volume 5, Issue 3, September 2022
Pages:
178-183
Received:
3 August 2022
Accepted:
27 August 2022
Published:
27 September 2022
DOI:
10.11648/j.ajcst.20220503.14
Downloads:
Views:
Abstract: Information security is the number one priority for any establishment concerned with its growth and privacy of its data. Attackers have devised a variety of strategies to get access to organizations' databases, both on cloud and offline systems. The incorporation of both cryptography and cloud folder model to secure cloud based storage presented in this paper, is a contemporary approach to securing cloud storage. The Cloud Folder Model adds to the storage strength. It serves as a significant deterrent against eavesdropping and injection assaults on the cloud storage. The model leverages RSA and AES data encryption, as well as a folder concept for storing the files in the cloud. It only enables authorized entities to have access to data and rejects suspicions and fraudulent attempts to access secured data. The system also created a mechanism that utilizes public and private keys. The system is be divided into two sections: online and offline. Data is encrypted using RSA and the Advanced Encryption Standard on the offline side (AES) while the online system adds to the encrypted data's security by guarding against injection attacks and data eavesdropping in transit. Our result shows that the new system provides better security for cloud storage than the existing system.
Abstract: Information security is the number one priority for any establishment concerned with its growth and privacy of its data. Attackers have devised a variety of strategies to get access to organizations' databases, both on cloud and offline systems. The incorporation of both cryptography and cloud folder model to secure cloud based storage presented in...
Show More
-
Damage Detection of Guqin in CT Images Based on Deep Neural Network
Issue:
Volume 5, Issue 3, September 2022
Pages:
184-189
Received:
25 August 2022
Accepted:
7 September 2022
Published:
27 September 2022
DOI:
10.11648/j.ajcst.20220503.15
Downloads:
Views:
Abstract: Guqin is the treasure of Chinese heritage culture and the top listed musical instrument. Among the Guqin collected in the Palace Museum, the most famous Guqin named Jiuxiao Huanpei was made by Lei Wei in the Tang Dynasty, which was regarded as an invaluable treasure and ranked first in the Palace Museum. There is a very rare situation in its internal structure where in the Dragon Pool, we saw a round ditch with a width of 2 cm and a depth of 1 cm, there is no such groove in the belly of an ordinary Guqin. This structure has led many well-known modern musical instrument makers to deliberately design their Guqin like this, because they believe that this is the exquisite design of the ancient master, which can make the sound performance better. Few people have questioned it for hundreds of years. Nowadays, the artificial intelligence technology in the world is developing very fast, among which the model used for object detection has gradually improved its accuracy. This paper applies the YOLO model in deep learning to train with 7803 CT slices of Guqin, and then tests the Jiuxiao Huanpei Guqin in the Forbidden City and several other Guqins. The conclusion is that this Guqin was not intended to be designed to that model, it was likely damaged, and like the Emperor’s New Clothes, no one except AI can tell the truth. The purpose of this paper is to arouse people's attention to the establishment of digital scientific analysis for cultural heritage.
Abstract: Guqin is the treasure of Chinese heritage culture and the top listed musical instrument. Among the Guqin collected in the Palace Museum, the most famous Guqin named Jiuxiao Huanpei was made by Lei Wei in the Tang Dynasty, which was regarded as an invaluable treasure and ranked first in the Palace Museum. There is a very rare situation in its intern...
Show More
-
Normal Versus Malignant Cell Classification in B-allwhite Blood Cancer Microscopic Images Using Deep Learning
Asad Ullah,
Muhammad Shoaib
Issue:
Volume 5, Issue 3, September 2022
Pages:
190-197
Received:
12 July 2022
Accepted:
4 August 2022
Published:
29 September 2022
DOI:
10.11648/j.ajcst.20220503.16
Downloads:
Views:
Abstract: In diagnosing cancer and determining its progress, an important aspect is the identification of malignant cells. Blood diseases such as leukemia are generally detected when cancer cells are much larger than normal cells in the late stages. Due to strong morphological similarities, the differentiation of cancer cells from normal blood cells is a challenge. Compared with normal cells, the precise classification of malignant cells in a microscopic image of blood cells depends on the early diagnosis of leukaemia. Transfer learning and fine-tuning of the VGG16 convolutional neural network through batch normalization can resolve the malignant and normal white blood cells classification problem with higher accuracy. Applying CLAHE to enhance image data quality is then passed as input to the network for training purposes. The results acquired by the fine- tuning of triple-loss and cross-entropy or cross- entropy loss with L2 normalization are compared. Furthermore, fine-tuning on a combined training validation dataset using simple cross-entropy loss can improve the model's performance. As an effective technique for diagnosing leukaemia, computer-aided cell classification has become popular. Fine-tuning VGG16 neural networks to classify normal and malignant cell images through batch standardization is part of our classification method. The proposed convolutional neural network detects cancer and normal cells with greater accuracy and time efficiency.
Abstract: In diagnosing cancer and determining its progress, an important aspect is the identification of malignant cells. Blood diseases such as leukemia are generally detected when cancer cells are much larger than normal cells in the late stages. Due to strong morphological similarities, the differentiation of cancer cells from normal blood cells is a cha...
Show More