Browse Items (2150 total)
Sort by:
-
Exploring BERT and Bi-LSTM for Toxic Comment Classification: A Comparative Analysis
This study analyzes on the classification of toxic comments in online conversations using advanced natural language processing (NLP) techniques. Leveraging advanced natural language processing (NLP) techniques and classification models, including BERT and Bi-LSTM models to classify comments into 6 types of toxicity: toxic, obscene, threat, insult, severe toxic and identity hate. The study achieves competitive performance. Specifically, fine-tuning BERT using TensorFlow and Hugging Face Transformers resulted in an AUC ROC rate of 98.23%, while LSTM yielded a binary accuracy of 96.07%. The results demonstrate the effectiveness of using transformer-based models like BERT for toxicity classification in text data. The study discusses the methodology, model architectures, and evaluation metrics, highlighting the effectiveness of each approach in identifying and classifying toxic language. Additionally, the paper discusses the implementation of a userfriendly interface for real-time toxic comment detection, leveraging the trained models for efficient moderation of online content. 2024 IEEE. -
Convergent replicated data structures that tolerate eventual consistency in NoSQL databases
The Eventual consistency is a new type of database approach that has emerged in the field of NoSQL, which provides tremendous benefits over the traditional databases as it allows one to scale an application to new levels. The consistency models used by NoSQL database is explained in this paper. It elaborates how the eventually consistent data structure ensures consistency on storage system with multiple independent components which replicate data with loose coordination. The paper also discusses the eventually consistent model that ensures that all updates are applied in a consistent manner to all replicas. 2013 IEEE. -
Smart Facial Emotion Recognition with Gender and Age Factor Estimation
Human-Computer Interaction (HCI) in an intelligent way, which aims at creating scalable and flexible solutions. Big tech firms and businesses believe in the success of HCI as it allows them to profit from on-demand technology and infrastructure for information-centric applications without having to use public clouds. Because of its capacity to imitate human coding abilities, facial expression recognition and software-based facial expression identification systems are crucial. This paper proposes a system of recognizing the emotional condition of humans, given a facial expression, and conveys two methods of predicting the age and gender factors from human faces. This research also aims in understanding the influences posed by gender and age of humans on their facial expressions. The model can currently detect 7 emotions based on the facial data of a person - (Anger, Disgust, Happy, Fear, Sad, Surprise, and Neutral state). The proposed system is divided into three segments: a.) Gender Detection b.) Age Detection c.) Emotion Recognition. The initial model is created using 2 algorithms - KNN, and SVM. We have also utilized the architectures of some of the deep learning models such as CNN and VGG - 16 pre-trained models (Transfer Learning). The evaluation metrics show the model performance regarding the accuracy of the Recognition system. Future enhancements of this work can include the deployment of the DL and ML model onto an android or a wearable device such as a smartphone or a watch for a real-time use case. 2022 Elsevier B.V.. All rights reserved. -
Deep Learning for Arrhythmia Classification: A Comparative Study on Different Deep Learning Models
Arrhythmias, or irregular heart rhythms, are a major global health concern. Since arrhythmias can cause fatal conditions like cardiac failure and strokes, they must be rapidly identified and treated. Traditional arrhythmia diagnostic techniques include manual electrocardiogram (ECG) image interpretation, which is time consuming and frequently required for expertise. This research automates and improves the identification of heart problems, with a focus on arrhythmias, by utilizing the capabilities of deep learning, an advanced machine learning technique that performs well at recognizing patterns in data. Specifically, we implement and compare Custom CNN, VGG19, and Inception V3 deep learning models, which classify ECG images into six categories, including normal heart rhythms and various types of arrhythmias. The VGG19 model excelled, achieving a training accuracy of 95.7% and a testing accuracy of 93.8%, showing the effectiveness of deep learning in the comprehensive diagnosis of heart diseases. 2023 IEEE. -
Synthesis and Studies on Partially Stabilized Zirconia and Rare-Earth Zirconate Pyrochlore Structured Multilayered Coatings
This work is focused on the thermal fatigue behaviour studies of ceramic coatings, as TBC (Thermal Barrier Coating) system, its importance in determining the thermo-mechanical properties and service-life estimation of the coatings when exposed to elevated operating temperatures. Commercial 6-8%Yttria stabilized zirconia (YSZ) top coat (TC) and NiCrAlY bond coat (BC) in (a) conventional YSZ (BC and TC), (b) multi-layered functionally graded materials (FGM) i.e., BC-blend (50BC+50TC)-(TC) configuration and (c) lab synthesized Zirconia based pyrochlore (Lanthanum Zirconate-LZ) were the coating materials involved. Nickel based super alloy Inconel 718 substrates were coated by using Atmosphere Plasma Spray (APS) system with three different (varying power) plasma spray parameters. All the sides of the 25mm x 10mm x 5mm thick substrates were completely covered with the bond coat and ceramic coating. FGM configuration was spray coated only on one side of the Inconel flat plates. Thermal shock cycle tests were performed on the coated specimen by following the ASTM B214-07 guidelines which comprised of introducing the coated specimen in a muffle furnace at 1150C, held in it for 2 minutes before removing from furnace followed by forced fan air cooling (one shock cycle). The specimen were periodically subjected to visual inspection for faults, before continuing the shock cycles, until the coating flaked off or cracked or detached from substrate. Cross section metallographic samples were prepared and analysed under SEM (Scanning Electron Microscope) and Energy Dispersive spectroscope (EDS) to study the as-sprayed coating morphology and interface quality, measure coating thickness, study defects characteristics and the chemical composition. Crystal structural phases were analysed using X-Ray Diffraction (XRD). 2019 Elsevier Ltd. -
Advancing Gold Market Predictions: Integrating Machine Learning and Economic Indicators in the Gold Nexus Predictor (GNP)
This study employs advanced machine learning algorithms to predict gold prices, using a comprehensive dataset from Bloomberg. The Gold Nexus Predictor (GNP), a key innovation, integrates historical data and economic indicators through advanced feature engineering. Methodologies include exploratory data analysis, model training with various algorithms like Linear regression, Random Forest, Ada Boost, SVM, and ARIMA, and evaluation using metrics like MSC, MAPE, and RMSE. The study's philosophical foundation emphasizes rationalism in economic forecasting and ethical model use. This research offers significant insights for investors and policymakers, enhancing understanding and decision-making in the gold market. 2024 IEEE. -
Enhancing Network Security with Comparative Study of Machine Learning Algorithms for Intrusion Detection
With the ever-increasing network systems and dependency on digital technologies, ensuring the security and integrity of these systems is of paramount importance. Intrusion detection systems (IDS) play a major role in sheltering such systems. Intrusion detection systems are technologies that are designed to monitor network and system activities and detect suspicious, unauthorized, malicious behavior. This research paper conducts a comprehensive comparative analysis of three popular machine learning algorithmsK-Nearest Neighbors (KNN), Random Forest (RF), and Logistic Regression (LR)in the context of intrusion detection using the renowned NSL-KDD dataset. Preprocessing techniques are applied, and the dataset is split for rigorous evaluation. The findings of this research highlight the effectiveness of Random Forest in detecting intrusions, showcasing its potential for real-world network security applications. This study contributes to the field of intrusion detection and offers valuable insights for network administrators and cybersecurity professionals to enhance network protection. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Smart Air Pollution Monitoring System Using Arduino Based on Wireless Sensor Networks
Impurity levels in air have risen throughout time as a result of several reasons, such as population expansion, increased automobile use, industry, and urbanization. All of these elements harm the health of individuals who are exposed to them, which has a detrimental effect on human well-being. We will create an air pollution monitoring system based on an IoT that uses a Internet server to track the air quality online in order to keep track of everything. An alert will sound when the level of harmful gases such CO2, smoking, alcohol, benzene, and NH3 is high enough or when the air quality drops below a specified threshold. The air quality will be displayed on the LCD in PPM. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Applications of Classification and Recommendation Techniques to Analyze Soil Data and Water Using IOT
As we are moving to a computerized and scientific world, data becomes an intrinsic part of our life. Agriculture sector is still unorganized with regard to automation and data analytics. This task is accomplished through sensors, data mining and analysis. In this paper, we propose real-time sensors to detect the soil features and predict the suitable crop cultivation using trained dataset. This would help the farmers to predict the type of cultivation to be done depending on the soil features. Today, the farmer can understand what type of cultivation will be prepared in the soil. Also, people of the upcoming generation will be using that sensor, different plant can be make. The cost of cultivation can be improved. Water level of the soil can be easily predicted. Which type of plant will be produced in the different soil can be predicted. So, this new type of cultivation followed by the next generation also. This paper has presented an improved by the pH sensor, water level sensor. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Simulation study of droplet formation in inkjet printing using ANSYS FLUENT
Flow simulations of jetting of inkjet drops are presented for water and ethylene glycol. In the inkjet printing process, droplet jetting behaviour is the deciding parameter for print quality. The multiphase volume of fluid (VOF) method is used because the interaction between two phases (air and liquid) is involved in the drop formation process. The commercial inkjet printer has a nozzle diameter of ~73.2?m. In this work, a simulation model of inkjet printer nozzles with different diameters 40?m, 60?m, and 80?m are developed using ANSYS FLUENT software. It is observed that when water is taken as solvent then the stable droplets are generated at 60?m nozzle diameter till 9?s because of its low viscosity. For higher diameter, the stamen formation is observed. Ethylene glycol stable droplets are achieved at 80?m nozzle diameter till 9?s because of their high viscosity (~10 times that of water). Along with the droplet formation, the sustainability of the droplet in the air before reaching the substrate is also important. The simulation model is an inexpensive, fast, and flexible alternative to study the ink characteristics of the real-world system without wasting resources. 2022 Institute of Physics Publishing. All rights reserved. -
Design and Implementation of an Optimized Mask RCNN Model for Liver Tumour Prediction and Segmentation
Segmentation of liver tumour is a tedious job due to their large variation in location and closeness to nearby organs. In this research, a novel Mask RCNN prototype is developed which uses ResNet-50 model. The architecture utilizes the masked location of convolution neural network to precisely detect liver tumours by recognizing liver sites to deal with changes in liver and CT snaps with distinct metrics. The preprocessed CT scans are subjected to ResNet-50 model. The data samples used here comprises 130 instances recorded from several clinical sites that are publicly available on the LiTS weblink. The designed model upon deployment generates a promising outcome thereby obtaining a DSC of 0.97%. Thus, we can conclude that the developed model is capable enough to accurately assess liver tumours and thus help patients in early diagnosis. 2023 IEEE. -
Enhancing Digital Citizenship Through Secure Identification Technologies in the Global Unified Digital Passport
Passports play a vital role in enabling international movement and security, as well as confirming one's identity. However, the existing passport system has many problems and limitations, such as identity fraud, passport falsification, human smuggling, terrorism, and border control. Despite the fast growth and adoption of digital technologies in various fields, the passport system has not been able to adapt to the changing demands and expectations of the global community. Therefore, there is a pressing need to investigate and develop a digital passport and verification system that can address the shortcomings of the conventional passport system and provide a more safe, convenient, and effective way of managing and verifying the identity and travel history of individuals across the world. This paper presents the solution and requirement for the development of a digital passport system that can be applied globally and universally. The paper proposes a conceptual framework and a technical architecture for the digital passport system, based on the principles of blockchain, biometrics, and cryptography. The paper also discusses the possible benefits, challenges, and implications of the digital passport system for various stakeholders, such as travelers, governments, airlines, and immigration authorities. The paper aims to contribute to the research and innovation of digital identity and citizenship, as well as to the progress of the sustainable development goals (SDGs) related to peace, justice, and strong institutions. 2024 IEEE. -
Integrated Automated Attendance System with RFID, Wi-Fi, and Visual Recognition Technology for Enhanced Classroom Security and Precise Monitoring
The integrated automated smart attendance system utilizes RFID, Wi-Fi, and visual recognition technologies to elevate classroom security and ensure precise monitoring of attendance records. It consolidates cutting-edge components such as RFID tags, ESP8266 Wi-Fi modules, ESP-32 CAM modules, solenoid locks, servo motors, and PIR sensors to devise a strong remedy. RFID technology enables accurate attendance tracking by assigning tags to students and faculty members. The Wi-Fi and visual recognition components enhance the system's functionalities, facilitating wireless connectivity, instantaneous data transfer, and validation of identities. Solenoid locks and servo motors ensure controlled access, responding to validated attendance records. PIR sensors detect motion, contrasting between genuine presence and proximity. The paper's methodology delineates the necessary hardware and software requirements, procedures for system initialization, testing phases, establishment of server connectivity, implementation of access control mechanisms, and formulation of end-of-session protocols. It highlights the successful integration and validation of hardware components, backend connectivity, identity confirmation, attendance recording, data encryption, and session termination procedures. The research aims to modernize attendance tracking in educational settings, improving efficiency, accuracy, and security while appreciating the need for further adaptation to suit diverse educational environments for broader adoption and sustained advancement. 2024 IEEE. -
Identification of the Functional Limitation of Marine Loading or Unloading Arm; A Case Study
Marine loading or unloading arms are used to transfer product from tanker vessels that often carries products like petroleum or chemicals from or to the tankers. Cochin Port has dedicated Tanker jetties for handling petroleum with Marine Loading Arms installed for safe handling of cargo. However, my studies in Cochin Port Trust have shown that it has a potential threat to tackle while it is taken for the maintenance process. The case study aids in understanding of the working of marine unloading arm installed in the port and to identify the functional or safety limitations of the existing model installed. This case study also proves that a small change in the design can bring about a big change in the safety of the people working with the equipment. The identified parameters have been studied for providing the necessary alterations of the design which could be implemented on the upcoming project of constructing the marine unloading arm in Cochin Port Trust. To support faster and safety loading/unloading requirement these hydraulically operated marine loading arms are fitted with emergency release couplings and emergency release system. Marine Loading Arms are operated by using the hydraulic system. During maintenance procedure while checking the Emergency Release System (ERS) functionality, accidental release of Emergency release coupling can cause fatality. Hence a fool proof design is suggested with an extra locking arrangement. The studies conducted till now and the reviews conducted contributed in the analysis of the development and validation of the design. A design of a locking machanism for preventing the fatality is created and analysed for suggesting it to the industry so that it could be incorperated in the upcoming project of constructing the marine loading and unloading arm. 2023 American Institute of Physics Inc.. All rights reserved. -
EFMD-DCNN: Efficient Face Mask Detection Model in Street Camera Using Double CNN
The COVID-19 pandemic has necessitated the widespread use of masks, and in India, mask-wearing in public gatherings has become mandatory, with violators being fined. In densely populated nations like India, strict regulations must be established and enforced to mitigate the pandemics impact. Authorities and cameras conduct real-time monitoring of individuals leaving their homes, but 24/7 surveillance by humans is not feasible. A suggested approach to resolve this problem is to connect human intelligence and Artificial Intelligence (AI) by employing two Machine Learning (ML) models to recognize people who arent wearing masks in live-stream feeds from surveillance, street, and new IP mask recognition cameras. The effectiveness of this method has been demonstrated through its high accuracy compared to other algorithms. The first ML model uses the YOLO (You Only Look Once) model to recognize human faces in real-time video streams. The second ML model is a pre-trained classifier using 180,000 photos to categorize photos of humans into two groups: masked and unmasked. Double is a model that combines face recognition and mask classification into a single model. CNN provides a potential solution that may be utilized with image or video-capturing equipment such as CCTV cameras to monitor security breaches, encourage mask usage, and promote a secure workplace. This studys proposed mask detection technology utilized pre-trained datasets, face detection, and various classifiers to classify faces as having a proper mask, an improper mask, or no mask. The Double CNN-based model incorporated dual convolutional neural networks and a technology-based warning system to provide real-time facial identification detection. The ML model achieved high performance and accuracy of 98.15%, with the highest precision and recall, and can be used worldwide due to its cost-effectiveness. Overall, the proposed mask detection approach can potentially be a valuable instrument for preventing the spread of infectious diseases. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024. -
Analysis of Social Media Marketing Impact on Customer Behaviour using AI & Machine Learning
The study of client behaviour has been revolutionized by the combination of social media marketing with cutting-edge technology like Artificial Intelligence (AI) and Machine Learning (ML) in today's age of digital transformation. This study delves into the complex interplay between AI/ML, consumer involvement, and social media marketing methods. Our research exposes crucial insights via careful data collecting, sentiment analysis, and the construction of prediction models. By stressing the importance of catering content to individual interests, AI-driven customization emerges as a potent tool, increasing user engagement by 18%. Analysis of online sentiment shows how important it is to keep people feeling good about a business; postings with positive feelings get 30% more likes and comments on average. Accurate and time-saving insights from machine learning models provide up new avenues for optimizing marketing's use of available resources. As a result of the study's conclusions, companies will be able to better connect with their customers, use their resources more efficiently, and behave ethically moving forward. Promising new developments in the subject include the next steps, which include sophisticated AI models, temporal dynamics analysis, and investigation of long-term consequences, ethical issues, and multichannel techniques. This study helps companies, marketers, and policymakers better understand the convergence of technology and marketing in today's ever-changing digital world so that they may better serve their customers and build a successful brand over time. 2024 IEEE. -
Application of Spray Drying process to convert Beneficial Compounds extracted from Plants into free-flowing powder
The use of herbal tablets has been rapidly growing and significant research work is being carried out worldwide with the goal to reap the benefits of the many useful plants that are available with medicinal values. Many of these plants go largely underutilized either due to lack of information on not only just the medicinal properties but simple and effective extraction methodologies as well, without sacrificing the properties of the extracts. Once extracted, the concentrates also must be converted into a suitable form that can be loaded in a capsule etc., ready to be consumed. While there many process methodologies being used worldwide to extract the useful resources from the plant, focus also must be on the process methodology that is being practiced to convert the extract (liquid or semi solid) into a solid free flowing powder form. Thus, in an herbal tablet, there many factors concerned with the manufacturing. They are (i) Identifying the most suitable plant for a particular immunity boosting purpose (ii) extraction of the useful contents, mostly in a liquid or slurry form (iii) transform the extract into a user-friendly product such as powder and finally (iv) encapsulation of the powder for ease of human consumption. This paper brings in a review of the several useful plants available around us across the world. In addition, the paper also highlights the suitable experimental results of the usefulness of spray-drying technology, which is a highly versatile process methodology to transform the extracts into free-flowing powder. Published under licence by IOP Publishing Ltd. -
Machine learning approach for automatic solar panel direction by using nae bayes algorithm
The upsurge in fuel prices are pointing out the fact that, the deficiency of conventional form of natural resources and building dams can never fulfill the demand of the growing population and it is exponentially increasing the electricity demand. Electricity is a day-to-day component, which is utilized for lighting, running appliances, machines. Moreover a large number of people are now switching to electric cars. Henceforth, it is equally important to achieve self-sustainability in energy needs and also it is necessary to have an infinite energy source. Sustainable power is the solitary solution to resolve this issue. On the other hand, the Indian government is promoting solar technology a lot in the year 2021 by providing subsidies to a maximum limit of 65% for the installation of home solar projects and this encourages people to switch to electric vehicles to reduce the pollution. This article presents a machine learning based dual-axis solar tracker to enhance the energy harnessing efficiency. Furthermore, the proposed method utilizes Nae Bayes algorithm to develop a better solution for producing higher energy from the solar panel. The Nae Bayes algorithm is a type of machine learning algorithm, which has been used to predict the reliable direction. This proposed method generates higher electricity, when compared with the traditional method. The experimental results aim to fix the north east direction of solar panel that produces 17.4 watts per hour, wherein the proposed method produces 24.8 watts. It is indicated that, more than 25% additional power generation is obtained by using Nae Bayes algorithm method. 2021 IEEE. -
Gender as a Predictor in the Perception of Sexual Harassment Definition
Sexual harassment is a pervasive problem across the globe and it is generally viewed subjectively. The review of the literature suggests that individual perception, history of past sexual harassment and other personal factors influence beliefs concerning the seriousness of the problem. The present study aims to explore the role of socio demographic variables in the definition of sexual harassment. One hundred and sixty-one college students volunteered for this study. Personal profile sheet and Sexual Harassment Definition Questionnaire were used to collect the data. The results of the chi-square test suggested that girls and students who already experienced sexual harassment found larger social incidents as harassment. However, the results of logistic regression found gender as a strong predictor of sexual harassment definition and the history of past harassment was failed to provide a statistical significance. Educating men on male privilege, violence against women and identifying behaviours in them that are not acceptable by women will be helpful. The Electrochemical Society -
Data Encryption and Decryption Techniques Using Line Graphs
Secure data transfer has become a critical aspect of research in cryptography. Highly effective encryption techniques can be designed using graphs in order to ensure secure transmission of data. The proposed algorithm in this paper uses line graphs along with adjacency matrix and matrix properties to encrypt and decrypt data securely in order to arrive at a ciphertext using a shared-key. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.