Pub Date : 2023-01-01DOI: 10.12720/jait.14.1.66-76
S. Savita, Geeta Rani, Apeksha Mittal
Rising number of fatalities caused by Coronary Artery Disease is a major concern for the public as well as the health industry. Furthermore, diagnostic methods like angiography are expensive and unaffordable for those who are not well-off. Also, angiography is bothersome for the patient due to allergic responses, renal damage, and bleeding where the catheter is inserted. The researchers in literature proposed the machine learning-based approaches for the detection of Coronary Artery Disease. But, these techniques have low accuracy. Thus, there is a scope for optimization of these techniques. The objective of this paper is to develop a machine learning system for the early detection of Coronary Artery Disease early. Also, it employs optimization methods viz. Particle Swarm Optimization, and Firefly Algorithm with Principle Component Analysis based feature extraction and decision tree-based classification. The proposed technique reports an accuracy of 95.3%. Thus, the technological solution may be used as an automatic diagnostic aid.
{"title":"An Optimized Machine Learning Approach for Coronary Artery Disease Detection","authors":"S. Savita, Geeta Rani, Apeksha Mittal","doi":"10.12720/jait.14.1.66-76","DOIUrl":"https://doi.org/10.12720/jait.14.1.66-76","url":null,"abstract":"Rising number of fatalities caused by Coronary Artery Disease is a major concern for the public as well as the health industry. Furthermore, diagnostic methods like angiography are expensive and unaffordable for those who are not well-off. Also, angiography is bothersome for the patient due to allergic responses, renal damage, and bleeding where the catheter is inserted. The researchers in literature proposed the machine learning-based approaches for the detection of Coronary Artery Disease. But, these techniques have low accuracy. Thus, there is a scope for optimization of these techniques. The objective of this paper is to develop a machine learning system for the early detection of Coronary Artery Disease early. Also, it employs optimization methods viz. Particle Swarm Optimization, and Firefly Algorithm with Principle Component Analysis based feature extraction and decision tree-based classification. The proposed technique reports an accuracy of 95.3%. Thus, the technological solution may be used as an automatic diagnostic aid.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66329790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.2.342-349
D. Elangovan, V. Subedha
—Sentimental Analysis (SA) has recently received a lot of attention in decision-making because it can extract and analyze sentiments from web-based reviews made by customers. In this case, SA has been used as a Sentiment Classification (SC) problem, in which reviews are typically labeled as positive or negative depending upon online reviews. By combining FS (Feature Selection) and categorization, this work proposes an effective SA method for internet reviews. FireFly (FF) and Levy Flights (FFL) algorithms have been used for extracting features of web-based reviews, and also the Multilayer Perceptron (MLP) framework has been used to categorize the emotions. A standard DVD database displayed the efficacy of the FF-MLP model on the testing. The outcome shows that the suggested FF-MLP system accomplishes enhanced performance with maximum sensitivity of 98.97%, specificity of 93.67%, accuracy of 97.97%, F-score of 98.75, and kappa of 93.32%.
{"title":"Firefly with Levy Based Feature Selection with Multilayer Perceptron for Sentiment Analysis","authors":"D. Elangovan, V. Subedha","doi":"10.12720/jait.14.2.342-349","DOIUrl":"https://doi.org/10.12720/jait.14.2.342-349","url":null,"abstract":"—Sentimental Analysis (SA) has recently received a lot of attention in decision-making because it can extract and analyze sentiments from web-based reviews made by customers. In this case, SA has been used as a Sentiment Classification (SC) problem, in which reviews are typically labeled as positive or negative depending upon online reviews. By combining FS (Feature Selection) and categorization, this work proposes an effective SA method for internet reviews. FireFly (FF) and Levy Flights (FFL) algorithms have been used for extracting features of web-based reviews, and also the Multilayer Perceptron (MLP) framework has been used to categorize the emotions. A standard DVD database displayed the efficacy of the FF-MLP model on the testing. The outcome shows that the suggested FF-MLP system accomplishes enhanced performance with maximum sensitivity of 98.97%, specificity of 93.67%, accuracy of 97.97%, F-score of 98.75, and kappa of 93.32%.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66330459","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.3.523-531
Ashmeet Kaur Duggal, Meenu Dave
—Internet-based computing allows the sharing of on-demand resources. This computing technique includes data processing and storage to globally separated machines, known as Cloud Computing. Confidentiality and integrity of data on the cloud are vital. The key constraints include effective access control, accessibility, and transmission of files, in a dynamic cloud environment, seeking a Trusted Virtual Data Center (TVDC). So, to overcome challenges such as data security and integrity due to exponentially growing data size, this research paper aims to develop a prediction model using the machine learning approach, which identifies the type of incoming packet on the TVDC. Alternatively, in other words, this system predicts whether the incoming packets on the server in the cloud environment are malicious or not, using the machine learning approach. This research explored artificial intelligence verticals in building systems with learned data structures for efficient data access. This research describes the implementation of machine learning algorithms for an efficient model’s prediction of the type of incoming packet on the server. It has achieved 88% accuracy using the Gradient Boosted Tree classifier. Also, in this study, the author compares the results of two algorithms, Decision Tree and Gradient Boosted Tree, and finally selects the most optimal for this prediction.
{"title":"An Efficient Model to Predict Network Packets in TVDC Using Machine Learning","authors":"Ashmeet Kaur Duggal, Meenu Dave","doi":"10.12720/jait.14.3.523-531","DOIUrl":"https://doi.org/10.12720/jait.14.3.523-531","url":null,"abstract":"—Internet-based computing allows the sharing of on-demand resources. This computing technique includes data processing and storage to globally separated machines, known as Cloud Computing. Confidentiality and integrity of data on the cloud are vital. The key constraints include effective access control, accessibility, and transmission of files, in a dynamic cloud environment, seeking a Trusted Virtual Data Center (TVDC). So, to overcome challenges such as data security and integrity due to exponentially growing data size, this research paper aims to develop a prediction model using the machine learning approach, which identifies the type of incoming packet on the TVDC. Alternatively, in other words, this system predicts whether the incoming packets on the server in the cloud environment are malicious or not, using the machine learning approach. This research explored artificial intelligence verticals in building systems with learned data structures for efficient data access. This research describes the implementation of machine learning algorithms for an efficient model’s prediction of the type of incoming packet on the server. It has achieved 88% accuracy using the Gradient Boosted Tree classifier. Also, in this study, the author compares the results of two algorithms, Decision Tree and Gradient Boosted Tree, and finally selects the most optimal for this prediction.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66331631","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.4.694-700
Chunxi Wang, Maoshen Jia, Yanyan Zhang, Lu Li
—The goal of speech separation is to separate the target signal from the background interference. With the rapid development of artificial intelligence, speech separation technology combined with deep learning has received more attention as well as a lot of progress. However, in the “cocktail party problem”, it is still a challenge to achieve speech separation under reverberant conditions. In order to solve this problem, a model combining the Weighted Prediction Error (WPE) method and a fully-convolutional time-domain audio separation network (Conv-Tasnet) is proposed in this paper. The model target on separating multi-channel signals after dereverberation without prior knowledge of the second field environment. Subjective and objective evaluation results show that the proposed method outperforms existing methods in the speech separation tasks in reverberant and anechoic environments.
{"title":"Multi-speaker Speech Separation under Reverberation Conditions Using Conv-Tasnet","authors":"Chunxi Wang, Maoshen Jia, Yanyan Zhang, Lu Li","doi":"10.12720/jait.14.4.694-700","DOIUrl":"https://doi.org/10.12720/jait.14.4.694-700","url":null,"abstract":"—The goal of speech separation is to separate the target signal from the background interference. With the rapid development of artificial intelligence, speech separation technology combined with deep learning has received more attention as well as a lot of progress. However, in the “cocktail party problem”, it is still a challenge to achieve speech separation under reverberant conditions. In order to solve this problem, a model combining the Weighted Prediction Error (WPE) method and a fully-convolutional time-domain audio separation network (Conv-Tasnet) is proposed in this paper. The model target on separating multi-channel signals after dereverberation without prior knowledge of the second field environment. Subjective and objective evaluation results show that the proposed method outperforms existing methods in the speech separation tasks in reverberant and anechoic environments.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66333351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.5.1124-1131
Emre Oner Tartan, Cebrail Ciflikli
—In the last decade Reinforcement Learning (RL) has significantly changed the conventional control paradigm in many fields. RL approach is spreading with many applications such as autonomous driving and industry automation. Markov Decision Process (MDP) forms a mathematical idealized basis for RL if the explicit model is available. Dynamic programming allows to find an optimal policy for sequential decision making in a MDP. In this study we consider the elevator control as a sequential decision making problem, describe it as a MDP with finite state space and solve it using dynamic programming. At each decision making time step we aim to take the optimal action to minimize the total of hall call waiting times in the episodic task. We consider a sample 6-floor building and simulate the proposed method in comparison with the conventional Nearest Car Method (NCM).
{"title":"Sequential Decision Making for Elevator Control","authors":"Emre Oner Tartan, Cebrail Ciflikli","doi":"10.12720/jait.14.5.1124-1131","DOIUrl":"https://doi.org/10.12720/jait.14.5.1124-1131","url":null,"abstract":"—In the last decade Reinforcement Learning (RL) has significantly changed the conventional control paradigm in many fields. RL approach is spreading with many applications such as autonomous driving and industry automation. Markov Decision Process (MDP) forms a mathematical idealized basis for RL if the explicit model is available. Dynamic programming allows to find an optimal policy for sequential decision making in a MDP. In this study we consider the elevator control as a sequential decision making problem, describe it as a MDP with finite state space and solve it using dynamic programming. At each decision making time step we aim to take the optimal action to minimize the total of hall call waiting times in the episodic task. We consider a sample 6-floor building and simulate the proposed method in comparison with the conventional Nearest Car Method (NCM).","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135312019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Gregoryus Imannuel Perdana, M. Devanda, D. N. Utama
The previous study of the Butterfly Life Cycle Algorithm (BLCA) has been technically realized in two stages of BLCA in measuring a company's growth performance. It was based on a combined method of the Balanced Scorecard (BSC) and Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis. This paper aims to continue the BLCA implementation by performing five stages of BLCA and then improve the algorithm by implementing the Fuzzy Logic (FL) conception into BSC. The implementation of the FL method transforms the bias values in four BSC parameters into a precise value to make the model more precise. A complete BLCA algorithm combined with FL is used to accurately assess companies' growth performance. By doing some corrections to the preceding study’s data of contribution value, the simulation result shows the difference in the performance value of 0.0026 with the previous one.
蝴蝶生命周期算法(Butterfly Life Cycle Algorithm, BLCA)在过去的研究中,已经从技术上实现了蝴蝶生命周期算法在衡量公司成长绩效方面的两个阶段。它是基于平衡计分卡(BSC)和优势、劣势、机会和威胁(SWOT)分析相结合的方法。本文旨在通过执行BLCA的五个阶段来继续BLCA的实现,然后通过将模糊逻辑(FL)概念引入平衡计分卡来改进算法。FL方法的实现将四个BSC参数中的偏置值转换为精确值,使模型更加精确。采用完整的BLCA算法结合FL来准确评估公司的成长绩效。通过对前人研究的贡献值数据进行一些修正,仿真结果显示与前人的性能值相差0.0026。
{"title":"Fuzzy Based Butterfly Life Cycle Algorithm for Measuring Company's Growth Performance","authors":"Gregoryus Imannuel Perdana, M. Devanda, D. N. Utama","doi":"10.12720/jait.14.1.1-6","DOIUrl":"https://doi.org/10.12720/jait.14.1.1-6","url":null,"abstract":"The previous study of the Butterfly Life Cycle Algorithm (BLCA) has been technically realized in two stages of BLCA in measuring a company's growth performance. It was based on a combined method of the Balanced Scorecard (BSC) and Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis. This paper aims to continue the BLCA implementation by performing five stages of BLCA and then improve the algorithm by implementing the Fuzzy Logic (FL) conception into BSC. The implementation of the FL method transforms the bias values in four BSC parameters into a precise value to make the model more precise. A complete BLCA algorithm combined with FL is used to accurately assess companies' growth performance. By doing some corrections to the preceding study’s data of contribution value, the simulation result shows the difference in the performance value of 0.0026 with the previous one.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66329623","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.1.56-65
Anthony Anggrawan, Mayadi Mayadi, Christofer Satria, B. K. Triwijoyo, R. Rismayati
COVID-19 has become a global pandemic that causes many deaths, so medical treatment for COVID-19 patients gets special attention, whether hospitalized or self-isolated. However, the problem in medical action is not easy, and the most frequent mistakes are due to inaccuracies in medical decision-making. Meanwhile, machine learning can predict with high accuracy. For that, or that's why this study aims to propose a data mining classification method as a machine learning model to predict the treatment status of COVID-19 patients accurately, whether hospitalized or self-isolated. The data mining method used in this research is the Random Forest (RF) and Support Vector Machine (SVM) algorithm with Confusion Matrix and k-fold Cross Validation testing. The finding indicated that the machine learning model has an accuracy of up to 94% with the RF algorithm and up to 92% with the SVM algorithm in predicting the COVID-19 patient's treatment status. It means that the machine learning model using the RF algorithm has more accurate accuracy than the SVM algorithm in predicting or recommending the treatment status of COVID-19 patients. The implication is that RF machine learning can help/replace the role of medical experts in predicting the patient's care status.
{"title":"Comparative Analysis of Machine Learning in Predicting the Treatment Status of COVID-19 Patients","authors":"Anthony Anggrawan, Mayadi Mayadi, Christofer Satria, B. K. Triwijoyo, R. Rismayati","doi":"10.12720/jait.14.1.56-65","DOIUrl":"https://doi.org/10.12720/jait.14.1.56-65","url":null,"abstract":"COVID-19 has become a global pandemic that causes many deaths, so medical treatment for COVID-19 patients gets special attention, whether hospitalized or self-isolated. However, the problem in medical action is not easy, and the most frequent mistakes are due to inaccuracies in medical decision-making. Meanwhile, machine learning can predict with high accuracy. For that, or that's why this study aims to propose a data mining classification method as a machine learning model to predict the treatment status of COVID-19 patients accurately, whether hospitalized or self-isolated. The data mining method used in this research is the Random Forest (RF) and Support Vector Machine (SVM) algorithm with Confusion Matrix and k-fold Cross Validation testing. The finding indicated that the machine learning model has an accuracy of up to 94% with the RF algorithm and up to 92% with the SVM algorithm in predicting the COVID-19 patient's treatment status. It means that the machine learning model using the RF algorithm has more accurate accuracy than the SVM algorithm in predicting or recommending the treatment status of COVID-19 patients. The implication is that RF machine learning can help/replace the role of medical experts in predicting the patient's care status.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66329656","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.2.250-256
Kleddao Satcharoen, Pikulkaew Tangtisanon
—The objective of this research was to compare icon selection accuracy under varying icon entropy and concreteness conditions between different generational cohorts ( Millennial, Generation X, and Baby Boomers ). These generational cohorts have different levels of experience with technology, with younger generations often being framed as “digital natives” and holding stronger technological experience and competence in comparison to older groups. Generational groups also have variations in physiological factors including visual acuity and reaction time. Despite these differences between user groups, many user interaction systems and processes are designed for a single user, rather than considering differences in user processing between different groups. Therefore, this study compares generational cohorts in their icon selection accuracy under varying icon conditions, to help identify what generational differences can be observed in this task. The study selected a sample of 150 participants ( n = 50 for each generational cohort ). The experiment was a 2 2 3 design ( entropy ( high / low ) abstractness ( abstract / concrete ) time ( 9 / 6 / 3 seconds ) , with each participant completing 60 trials ( five questions per entropy / abstractness pair over three timed runs ). Results showed that there were significant differences in mean accuracy per trial under all of the time conditions and icon entropy and concreteness conditions . Mean differences showed that under most conditions, Millennial and Generation X participants did not have a significant mean difference, but Baby Boomers were significantly slower under almost all conditions . The implication of this finding is that Baby Boomers are more sensitive to icon abstractness and entropy conditions than other age groups tested .
{"title":"A Generational Cohort Comparison of Icon Selection Accuracy under Varying Conditions of Icon Entropy and Concreteness","authors":"Kleddao Satcharoen, Pikulkaew Tangtisanon","doi":"10.12720/jait.14.2.250-256","DOIUrl":"https://doi.org/10.12720/jait.14.2.250-256","url":null,"abstract":"—The objective of this research was to compare icon selection accuracy under varying icon entropy and concreteness conditions between different generational cohorts ( Millennial, Generation X, and Baby Boomers ). These generational cohorts have different levels of experience with technology, with younger generations often being framed as “digital natives” and holding stronger technological experience and competence in comparison to older groups. Generational groups also have variations in physiological factors including visual acuity and reaction time. Despite these differences between user groups, many user interaction systems and processes are designed for a single user, rather than considering differences in user processing between different groups. Therefore, this study compares generational cohorts in their icon selection accuracy under varying icon conditions, to help identify what generational differences can be observed in this task. The study selected a sample of 150 participants ( n = 50 for each generational cohort ). The experiment was a 2 2 3 design ( entropy ( high / low ) abstractness ( abstract / concrete ) time ( 9 / 6 / 3 seconds ) , with each participant completing 60 trials ( five questions per entropy / abstractness pair over three timed runs ). Results showed that there were significant differences in mean accuracy per trial under all of the time conditions and icon entropy and concreteness conditions . Mean differences showed that under most conditions, Millennial and Generation X participants did not have a significant mean difference, but Baby Boomers were significantly slower under almost all conditions . The implication of this finding is that Baby Boomers are more sensitive to icon abstractness and entropy conditions than other age groups tested .","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66330152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.3.510-517
Vasile Dan, I. Nascu, S. Folea
—Reading is an activity that leads to acquiring information and developing a person’s knowledge. Therefore, everyone should have equal access to the same sources of information. Unfortunately, blindness is a disease that restricts the affected people from reading books that are not converted into Braille. This paper describes a novel solution for the real-time conversion of any text into Braille. The system will rely on image processing and a camera to gather the raw text data from any book in physical format. Furthermore, e-books and documents in any digital format, Braille Ready Format (BRF), Portable Embosser Files (PEF), TXT, PDF, or PNG, can be provided for Braille conversion. Image enhancement algorithms, neural networks, and Optical Character Recognition (OCR) algorithms are used to extract accurate content. The process is controlled by a Raspberry Pi 4. A refreshable Braille mechanism, based on an Arduino Due microcontroller, is used to display the dots for each character. The algorithms are implemented to work with the mechanical structure design that was created to reduce the cost and give the user a complete reading experience of any book.
{"title":"Advanced Real Time Embedded Book Braille System","authors":"Vasile Dan, I. Nascu, S. Folea","doi":"10.12720/jait.14.3.510-517","DOIUrl":"https://doi.org/10.12720/jait.14.3.510-517","url":null,"abstract":"—Reading is an activity that leads to acquiring information and developing a person’s knowledge. Therefore, everyone should have equal access to the same sources of information. Unfortunately, blindness is a disease that restricts the affected people from reading books that are not converted into Braille. This paper describes a novel solution for the real-time conversion of any text into Braille. The system will rely on image processing and a camera to gather the raw text data from any book in physical format. Furthermore, e-books and documents in any digital format, Braille Ready Format (BRF), Portable Embosser Files (PEF), TXT, PDF, or PNG, can be provided for Braille conversion. Image enhancement algorithms, neural networks, and Optical Character Recognition (OCR) algorithms are used to extract accurate content. The process is controlled by a Raspberry Pi 4. A refreshable Braille mechanism, based on an Arduino Due microcontroller, is used to display the dots for each character. The algorithms are implemented to work with the mechanical structure design that was created to reduce the cost and give the user a complete reading experience of any book.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66331211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}