Volume 12 | Issue 2

 Abstract

According to the World Health Organization, around 40 million people are blind, while another 250 million have some form of visual impairment. They come across many troubles in their daily life, especially while navigating. They often depend on others for help to satisfy their day-to-day needs. So, it is quite a challenging task to implement a technological solution to assist them. Several technologies have been developed for the assistance of visually impaired people. One such attempt is that we would wish to make an Integrated Machine Learning System that allows blind victims to identify and classify real-time objects generating voice feedback and distance. Which also produces warnings whether they are very close or far away from the thing?

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blind assistance, machine learning, assistive devices, virtually-impaired people, obstacles detection, navigation and orientation systems, obstacles avoidance, neural architectecture search(NAS).

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This research proposes a novel approach to enhance the learning experience for slow learners by detecting their learning styles using fingerprint recognition technology. The proposed system integrates machine learning algorithms to analyze and classify learning styles based on the VAK (Visual, Auditory, Kinesthetic) model. By recognizing individual learning styles, the system aims to personalize educational strategies and resources to better suit the preferences and cognitive strengths of learners. This approach has the potential to optimize the learning experience for slow learners and improve their academic performance. The proposed system is expected to contribute to the development of personalized learning system that can cater to the diverse needs of learners.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Fingerprint, Learning Style, Convolutional Neural Network.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper is an attempt to critically examine the discipline of Economics. By introducing the concept of 'Popularized Economics', certain schools of thought are analyzed. While both classical and neo-classical schools are found to be patriarchal, it has been argued that Keynesianism can be partially viewed as feminist. Excessive patriarchy present in economic literature flows into classrooms through teaching which further hegemonizes male bodies. Such an approach creates a gender imbalance between women and men in a binary setup. Women's existence is undermined due to the application of popularized economics. Feminist Economics can be employed as a potential solution for producing sustainable masculinities and reimagining the discipline of Economics which will ultimately help to achieve the goal of gender inclusivity.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Popularized Economics, Feminist Economics, Hegemonic Male Bodies

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In this comprehensive study, we have advanced the field of agricultural technology by developing and comparing multiple deep learning models for the classification of rice varieties. Conducted in the agriculturally rich regions of Southern Bangladesh, our research utilized a diverse dataset comprising 20,000 high-resolution RGB images representing five principal rice varieties. The study primarily focused on a custom-engineered hybrid deep learning model, designed specifically for this agricultural application. This model's architecture encompasses an initial convolutional layer, zero-padding, batch normalization, and max pooling, followed by residual blocks that address the vanishing gradient problem, and concludes with Global Average Pooling leading into a Support Vector Machine (SVM) for final classification.Additionally, we incorporated and evaluated the performance of two renowned deep learning models: MobileNetV2 and VGG16. These models were adapted and fine-tuned to suit the specific requirements of our dataset and task. Across various metrics, including precision, recall, and F1-score, our hybrid model demonstrated superior performance, achieving an exceptional 99% accuracy. This was notably higher compared to the 95% and 93% accuracy achieved by VGG16 and MobileNetV2, respectively. Various optimizers, including SGD, RMSprop, Adam, and Nadam (all with a learning rate of 0.001), were employed to refine our models, with the Adam optimizer emerging as the most effective across all models.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Rice, Adam optimizer, MobileNetV2, VGG16, Accuracy,Max pooling, Zero-padding

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

ABSTRACT The study, design and finite element investigation of bolt and nut for fastening stress reduction were successfully investigated. researchers created a bolt of M6 with a hexagonal head of 12mm, and a hexagonal nut of M12 both having a right hand ANSI metric thread profile, with assigned material being Steel, High Strength, Low Alloy to reduce wear and tear using AutoDesk inventor. The created model was analyzed using FEA software. The bolt was first subjected to fastening moment of 240 N mm, with the nut being subjected to fixed constraints. Thereafter, the nut was subjected to fastening moment of 240 N mm with the bolt subjected to fixed constraints. Results showed that the ultimate tensile strength of assigned material reduced from 540 MPa to 448 MPa when fastening moment was applied at bolt head and Nut side respectively. This suggested that failure due to tensile stress would be predominant when fastening at the Nut side. Also, maximum fastening stresses were observed to be 7.109 MPa and 6.66822 MPa when fastening moment was applied at bolt head and Nut side respectively. This suggested that fastening at the Nut side would reduce fastening stress or thread wear. In addition, the maximum displacements were observed to be 0.00190463 mm and 0.00204884 mm when fastening moment was applied at bolt head and Nut side respectively. This indicated that fastening at the Nut side will give lesser revolution to achieve a tightened bolt and nut joint. Furthermore, the maximum contact pressures were observed to be 10.7241 MPa and 10.9924 MPa when fastening moment was applied at bolt head and Nut side respectively. This indicated that slippage is minimized at nut side fastening with high induce stress at the bolt shank, as the study revealed. The researchers made the following recommendations: Fastening at nut side should be adopted to reduce stress and maximize displacement/ advancement, bolt and nut design materials must have higher tensile strength rather than compressive strength or yield strength, since failure due to tensile stress is predominant, etc.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords ---- Stress, Bolt and nut, Fastening, Finite element analysis, Turning moment, Constraints

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Emotional Intelligence, Self-Efficacy, School Environment, Teacher Effectiveness and High School Teachers.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Several text-to-SQL systems have been created to bridge the gap between users and data, enabling individuals without SQL expertise to ask questions in natural language and interact with databases effectively. The progress observed in text-to-SQL tasks has contributed to advancements in deep learning methods. In order to truly advance the development of text-to-SQL systems, prior research needs to be deconstructed to comprehend the applicability and challenges of various strategies. The review paper's main goal is to give an overview of text-to-sql techniques that query data using natural language. This paper can help serve as a reference for researchers and practitioners interested in developing and applying natural language interfaces for data interaction in the era of large language models.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Text-to-SQL ,NLP, GPT-3.5,SQL ,LLM ,Database

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Cloud computing significantly alters the way we use computers and guarantees access and storage of our personal and business information. These new computing and communication models face new data security challenges. Existing data conservation procedures such as encryption fail to prevent data from the attacks of theft, especially in the cloud provider. So to overcome these problems we are proposing a new technology called Fog Computing. We propose a different approach in Fog computing to obtain data in the cloud using aggressive decoy technology and user behavior profiling. The users using the Cloud are trapped and their access patterns are recorded. Every User has a unique profile which is monitored and updated. We monitor data access in the cloud by the users and detect abnormal data entry patterns. When unauthorized access is suspected and challenged by challenge questions, we begin the wrong attack by returning the bulk of the information to the attacker. This protects users' real data from being misused. Experiments in a local file setting give evidence that this approach can provide an unprecedented level of user security in the cloud environment.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cloud Computing,Fog Computing,Data Security,Abnormal Data.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Securing information flow has emerged as a critical component of communication in the current digital era. It is crucial to have a strong encryption technique that can shield critical data from unwanted access given the rising number of cyber threats. Therefore, cryptography is essential for ensuring security. Symmetric Key and Asymmetric Key cryptography are the two fundamental types. Instead, then using many keys for encryption and decryption like Asymmetric Key does, Symmetric Key uses just one key.The most extensively used algorithms are those using symmetric keys. The difficulty of deciphering the original messages is what gives these algorithms their strength. Modulo69-based KED-A symmetric key is a cutting-edge encryption technique that offers secure data transfer. In this project data deduplication is a technique used to improve storage utilization by eliminating duplicate data. The technique of encoding plain information into an unintelligible format termed cipher text is known as encryption. Decryption is the procedure of turning encrypted text to plain text. Asymmetric key cryptography and symmetric key cryptography are the two forms of cryptography. Asymmetric key cryptography uses different keys, one for encryption and the other for decryption, as opposed to symmetric key cryptography, which uses the same key for both operations.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Symmetric Key , Asymmetric Key Cryptography,De-ciphering,Encryption,Decryption.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The amount of pollution caused by humans on the planet has increased dramatically since the industrial revolution. Many of the pollutants in the environment are visible, such as those in the air, water, and soil. Some people, particularly those who reside in big industrial cities, will be aware of air pollution. Since air quality is becoming one of the main factors affecting human health. Air pollution has become a major concern worldwide due to its detrimental effects on human health and the environment. Accurate prediction of air quality is crucial for implementing effective mitigation strategies and safeguarding public health. This study focuses on employing machine learning techniques, specifically the Long Short-Term Memory (LSTM) algorithm, for air quality prediction. The LSTM algorithm, a type of recurrent neural network, is known for its ability to capture temporal dependencies in sequential data. The methodology involves collecting historical air quality data, including pollutant concentrations, meteorological variables, and other relevant factors. These data are preprocessed and used to train the LSTM model, which learns the complex relationships between the input variables and the air quality outcomes. The trained model is then used to make predictions for future air quality conditions. The performance of the LSTM model is evaluated using various evaluation metrics, such as mean absolute error (MAE) and root mean square error (RMSE), to assess its accuracy in predicting air quality.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Long Short-Term Memory (LSTM) algorithm,Air Pollution,Quality Outcomes,Human Health,Environment

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In the contemporary world, the surge in the number of daily patients is evident, propelled by the swift evolution of lifestyles. The queues at hospitals and local doctor's residences are consequently experiencing a steep incline. For individuals with packed schedules, the significant waiting time to consult with a doctor becomes a considerable inconvenience. Some ailments demand prolonged periods for recovery, and heart disease, a widespread concern globally, claims lives on a daily basis, affecting both the young and the elderly. Addressing the escalating healthcare challenges of today and tomorrow necessitates a shift toward remote data collection by care providers, accurate diagnoses irrespective of distances, leveraging AI for data analysis to enhance both business and health outcomes, and more.In this transformative landscape, chatbots, also known as conversational interfaces, emerge as a novel means for individuals to engage with computer systems. The introduction of chatbots revolutionizes the user experience by allowing them to pose questions in a manner akin to conversing with a human. Notably, chatbots are rapidly gaining traction on computer chat platforms, harnessing artificial intelligence to comprehend human inputs effectively. This technological integration facilitates a more intuitive and user-friendly interaction, marking a pivotal advancement in healthcare and beyond. As the reliance on such innovative solutions grows, the intersection of AI, healthcare, and conversational interfaces holds the promise of reshaping how we approach and experience medical care in our increasingly dynamic world.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Hospitals,Doctors,Artificial Intelligence,Chatbots..

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The goal of the Improvisational Capability program is to introduce the fundamentals of improvisation mainly with Entrepreneur to promote innovative thinking and teamwork, enhance performance abilities, and boost team effectiveness. As an alternative to conventional coaching methods, the curriculum can be applied in a different field. The first part of each session is a review of the previous module, followed by a discussion that provides a more thorough explanation of how improvisation functions, its benefits, and risks, as well as how we may utilize it most efficiently. We will use the idea of freelancing in IT services in our business. There are many similarities between freelancers and businesses in the IT industry. When it comes to finding and keeping employees, creating a business culture, both groups have experience with managing projects and people similar difficulties. However, there are several important differences in how they approach business planning that might help you decide if your company will benefit more from a freelancer or corporation model. Some people find success working as a freelancer, while others find it difficult. Finding clients, keeping them, and getting the appropriate remuneration are the key obstacles to generating money as a freelancer. Additionally, self-employment calls for ongoing attempts to generate income through the investment in systems and infrastructure for ongoing success. Especially if you operate from home or other remote locations, being a freelancer frequently requires full-time dedication. To succeed as a freelancer in the IT business, you need to be persistent and patient when it comes to finding clients and making payment deadlines. There are numerous web services available if you're seeking for freelance work, and they can all help you quickly get your ideal position. The Naive Bayes machine learning algorithm, which is based on the Bayes theorem, is utilized for various classification functions. Gaussian Naive Bayes is the name given to the Naive Bayes generalization. Although there are numerous functions used to estimate data distribution, the Gaussian or normal distribution is the most straightforward to employ.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Naive Bayes, Self-Employement,IT Business, Improvisational Capability

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The realm of license plate detection methods, grounded in OpenCV, stands as a well-explored domain within computer vision, boasting applications in diverse fields such as traffic management, vehicle surveillance, and law enforcement. This project introduces an innovative license plate detection methodology rooted in OpenCV, leveraging a spectrum of computer vision techniques to adeptly extract and recognize characters embedded within license plates. The systematic approach of this proposed system unfolds through multiple phases. Initially, the input image of a vehicle undergoes meticulous preprocessing steps, encompassing grayscale conversion, contrast adjustment, and adaptive thresholding. Subsequently, contours emerge from the thresholded image, and potential license plate characters are sieved based on criteria like size and aspect ratio. Precision in grouping these potential characters is achieved through the implementation of a contour arrangement algorithm, ensuring the accurate formation of a license plate region. Post-extraction of this region, further preprocessing is applied to enhance character visibility. Individual character segmentation within the license plate region is accomplished using contour detection. Finally, the optical character recognition (OCR) prowess of Tesseract is harnessed to recognize the segmented characters and extract alphanumeric information from the license plate. The system's development unveils promising results in license plate recognition, affirming the efficacy of the applied computer vision techniques. Nonetheless, it is imperative to acknowledge that the system's performance is contingent on factors such as input image quality, character segmentation accuracy, and the OCR engine's performance, introducing a dimension of variability that necessitates attention and potential refinement.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Opencv, License Plate Recognition,Segmentation,OCR.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The relevance of security concerns has increased with the ongoing advancement of computer technology and the increasing reliance of humans on network technology. To prevent attacks and security flaws, user authentication is essential. There are several forms of authentication, including facial recognition, voice recognition, SMS one-time passcodes, and fingerprint scanning. One of the key uses for image processing in still photos is face recognition. Making an automated system that can recognize faces as well as a person is a real task. This paper's primary goals are to examine the value of CNN, describe the many datasets used in face recognition systems, and assess the various CNN models. The deep learning CNN may be applied to facial recognition to boost authentication security. Here we are collecting the dataset of different faces. Once after preprocessing it we train the data with the CNN algorithm. After training, we will test the results using the OpenCV and also can upload the image for recognition of faces.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

OpenCV, CNN Algorithm,Facial Recognition,Deep Learning

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The system is designed to work with live video feeds from cameras installed in strategic locations. It employs object detection algorithms to identify and track vehicles in real-time, allowing for accurate traffic analysis. The system incorporates speed violation detection by defining speed limit lines and calculating the speed of vehicles passing through those lines. Violation instances are flagged, and images or videos of the violations are captured for further analysis or evidence purposes. The project also includes a user-friendly interface that provides real-time traffic statistics, including the total number of vehicles, traffic congestion levels, and detected violations. Additionally, the system offers configurable settings for road-specific parameters, such as speed limits and the number of allowed vehicles. The proposed system aims to enhance traffic management and improve road safety by providing timely and accurate information to authorities. It can aid in monitoring traffic patterns, identifying congested areas, and enforcing speed limits. The system has the potential to reduce accidents, enhance traffic flow, and contribute to efficient transportation management.Overall, the project showcases the effective utilization of computer vision and deep learning algorithms to develop a comprehensive traffic monitoring and violation detection system that can significantly impact road safety and traffic management.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Traffic Management,Traffic Flow,Deep Learning Algorithms,Traffic Monitoring,Computer Vision.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

One of the most common problems faced by people suffering from common ailments or may be even major ones is the lack of immediate first aid consultation or a centralized service to a clinical database. Due to this lack of the knowledge of the standard operating procedure in such cases, these ailments might aggravate. This results in either physical or mental tension for the person suffering from such ailments. In some cases, the patient suffers from intense mental stress as they try to figure out the reason for their condition.The proposed system tries to eliminate their need to figure out their disease by giving them access to a centralized clinical repository in a much interactive way, just like in a virtual assistant, hence Virtual Health Assistant(VHA).The user gets asked several questions, each one contextually aware of the previous one. The user selects the ailments or their condition and thus a conclusion is reached.This project aims to develop a web service that can present information regarding the health issues and ailments & their history. At the end a precise prescription is generated. What this project can't ensure is the accuracy of the health condition that the service arrived at, and thus in such cases a physician must be contacted. These features thus eliminate the need to search for symptoms online.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Virtual Health Assistant,Centralized Service,Health Issues,Mental Stress.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The use of deepfake techniques in the area of converting images into live video has attracted a lot of attention recently. The term "deepfake," which combines the terms "deep learning" and "fake," describes the process of creating artificial content that is convincing and realistic, usually with the use of Generative Adversarial Networks (GANs). Based on a single input image, this method enables the synthesis of a video sequence that mimics the appearance of a target individual. This project shows how to make a movie of a person's facial driver using a first-order motion model.The algorithm can predict the movements of the head and face during driving after being trained on a dataset of driving videos and facial photos. The finished video is realistic and suitable for a range of objectives, including developing virtual reality experiences or instructing autonomous vehicle training programs. The imageio and matplotlib libraries are used in the project's Python implementation.The First Order Motion Model (FOMM) library is used to implement the first-order motion model. Using the first-order motion model, new techniques for face tracking and animation can be created. Video games and other applications could benefit from the increased realism provided by this technology.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

First Order Motion Model,Generative Adversarial Networks,Facial Driver,Vehicle Training Program.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

There is a vast increase in broadband access due to which this new generation netizens are spawned. In today's situation consumers mainly use the network as a interactive medium for multimedia entertainment and communication purpose. It includes interactive network applications such as teleconferencing, network gaming and online trading which are gaining popularity. We propose a latency equalization service (LEQ), which equalizes the perceived latency for all clients participating in an interactive network application. LEQ is used in variety of applications like gaming, video streaming and real-time communication systems. To effectively implement the proposed LEQ service, network support is essential. LEQ is a process used in data communication networks to ensure that all devices on the network experiences the same delay when transmitting and receiving the data. The LEQ architecture uses a few routers in the network as hubs to redirect packets of interactive applications along paths with similar end-to-end delay. We first formulate the hub selection problem, prove its NP-hardness, and provide a greedy algorithm to solve it.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Latency Equalization Service (LEQ),NP-Hardness,End-to-End Delay,Greedy Algorithm.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The problem of fraudulent mobile applications has grown significantly in importance as a result of the quick development of mobile technology and the rising popularity of mobile applications. These malicious apps not only endanger user's devices but also steal personal information. To safeguard users from potential harm, it is crucial to track down and identify fraudulent mobile applications. With the help of sentiment analysis and the Naive Bayes classifier, SVM etc.., this project is developed for identifying fraudulent applications based on user reviews. The goal is to create a framework that uses data mining and sentiment analysis to analyse user reviews and find review-based evidence of fraud. This projectseek to evaluate the authenticity and dependability of mobile applications before users download them by utilizing sentiment analysis. The suggested method involves gathering user reviews from the Google Play store and classifying them as positive or negative using sentiment analysis. Based on the opinions expressed in the reviews, the Naive Bayes classifier, SVM etc.., is used to categorize applications as either legitimate or fraudulent. By giving users a tool to make educated decisions about the applications they download, this strategy empowers users. Users will be able to recognize fraudulent applications and steer clear of any risks involved with downloading them by putting this framework into place. While giving users a trustworthy way to distinguish between fraudulent and legitimate applications, the system will help to ensure the security and integrity of the mobile application market.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The agricultural industry has been facing challenges in traditional and manual visual grading of fruits due to its laborious nature and inconsistent inspection and classification process. To accurately estimate yield and automate harvesting, it is crucial to classify the fruits based on their ripening stages. However, it can be difficult to differentiate between the ripening stages of the same fruit variety due to high similarity in their images during the ripening cycle. To address these challenges, we plan to develop an accurate, fast, and reliable fruit detection system using deep learning techniques. The modernization of crops offers opportunities for better quality harvests and significant cost savings. Our approach involves adapting the state-of-the art object detector faster R-CNN, using transfer learning, to detect fruits from images obtained through model colour (RGB). Spectroscopy analysis to predict the quality of fruit and categorization by using AS7265x Spectrophotometer. Our system's robustness will enable us to differentiate between fruit varieties and determine the ripening stage of a particular fruit with effectiveness and accuracy. The system will also efficiently segment multiple instances of fruits from an image and accurately grade individual objects

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

RGB,Spectrophotometer,Spectroscopy,Transfer Learning,R-CNN.

  License

Creative Commons Attribution 4.0 and The Open Definition