Browse Items (9795 total)
Sort by:
-
Natural Language Processing in Medical Applications
Medical applications of machine learning are very new, and there are still several obstacles that limit their widespread use. There is still a need to address issues like high dimensionality data and a lack of a standard data schema. An intriguing way to explore the possibilities of machine learning in healthcare is to apply it to the difficult problem of cardiovascular disease diagnosis. At the present day, cardiovascular disorders account for the majority of deaths worldwide. It is often too late to adopt appropriate treatment for many of them because they progress for a long time without showing any symptoms. Because of this, its crucial to get checked up on routinely so that any developing diseases can be caught early. If the sickness is caught early enough, effective therapy can be put into place to stop the progression of the illness. This is done with the intention of analysing data from many sources and making use of NLP to overcome data heterogeneity. This paper evaluates the usefulness of several machine learning methods (such as the Naive Bayes (NB), Transductive Neuro-Fuzzy Inference, and Terminated Ramp-Support Vector Machine (TR-SVM)) for healthcare applications and suggests using Natural Language Processing (NLP) to address issues of data heterogeneity and the transformation of plain text. The implementation, testing, comparison of performance and analysis of the parameters of the algorithms used for diagnosis have simplified the process of selecting an algorithm better suited to a certain instance. TWNFI is particularly effective on larger datasets, while Terminated Ramp-Support Vector Machine is well suited to lesser datasets with a lower number of magnitudes due to performance difficulties. 2024 Scrivener Publishing LLC. -
Revolutionizing Biodegradable and Sustainable Materials: Exploring the Synergy of Polylactic Acid Blends with Sea Shells
This study explores the mechanical properties of a novel composite material, blending polylactic acid (PLA) with sea shells, through a comprehensive tensile test analysis. The tensile test results offer valuable insights into the materials behavior under axial loading, shedding light on its strength, stiffness, and deformation characteristics. The results suggest that the incorporation of sea shells decrease the tensile strength of 14.55% and increase the modulus of 27.44% for 15 wt% SSP (sea shell powder) into PLA, emphasizing the reinforcing potential of the mineral-rich sea shell particles. However, a potential trade-off between decreased strength and reduced ductility is noted, highlighting the need for a delicate balance in material composition. The study underscores the importance of uniform sea shell particle distribution within the PLA matrix for consistent mechanical performance. These results offer a basis for additional PLA-sea shell blend optimization, directing future efforts to balance strength, flexibility, and other critical attributes for a range of applications, including biomedical devices and sustainable packaging. This investigation opens the door to more sustainable and mechanically strong materials in the field of additive manufacturing by demonstrating the positive synergy between nature-inspired materials and cutting-edge testing techniques. 2024 The Authors. -
Enhanced technique for detection and prevention of phishing on websites
Phishing is a kind of assault where cyber criminals trap individuals to gain access to someone's private data like credit card details, passwords, account details, etc. The false e-mails look shockingly genuine and even the Web pages where clients are requested to enter their data may look legitimate. Forgery of a website is a sort of online assault where the phishing person builds a duplicate of a true authorized site, with the objective of misguiding a client by fishing out data that could be utilized to dupe or instigate different assaults upon the victim. In this paper, a new technique is developed using the combination of CORS, Public Repository technique and Heuristic functions. This technique allows only authorized Domain to replicate the original website. Copyright 2019 American Scientific Publishers All rights reserved. -
A literature review on friction stir welding of dissimilar materials
Friction stir welding (FSW) employs a tool that does not require any filler materials; frictional heat is produced and performs a solid-state joining method. Severe plastic deformation causes to join similar and dissimilar materials without melting the workpiece at the welding line. Friction stir welding is the most recent friction welded joining processes with the most surprising features when welding various metal alloys, including magnesium, aluminium, copper, and steel. FSW is victorious of all the other conventional welding methods implied in many industrial applications like automobile, aerospace, fabrication, shipping, marines and robotics. It gives high-quality welds, energy input, and distortion are lower, better retention of mechanical properties; it is eco-friendly and can be performed less operating cost. This research work aims at the FSW process in Al-Cu alloys, highlighting:(a) Optimizing the welding process parameters, welding feed rate, tool rotation speed, (b) Evaluation of Electrical Conductance properties of joints, (c) Mechanical properties and metallography characteristics of joints. 2021 Elsevier Ltd. All rights reserved. -
Experimental analysis and RSM-based optimization of friction stir welding joints made of the alloys AA6101 and C11000
In the present study, the evaluation of FSW input parameters on output response ultimate tensile strength (UTS) of the friction stir welded AA6101-C11000 joint is in agreement. The response surface methodology (RSM) was adapted for generating the mathematical regression equation to predict the UTS and to develop the FSW parameters to attain the highest UTS of the FSW joints. The central composite design (CCD) method from RSM with five levels and three factors, i.e., tool rotational speed, feed rate, and tool offset used to conduct and minimize the number of tests. During FSW, base sheet cu (hard metal) was stationed on the advancing side (+1 mm, +1.68 mm tool offset) and the base sheet Al (soft metal) on the retreating side (?1 mm, ?1.68 mm tool offset). The radiography studies were accomplished to inspect the internal flaws of the FSW joints (Al-Cu).The XRD and SEM investigation of the ruptured specimens during the tensile test to evaluate the IMCs phase anatomy and fracture analysis. The maximum UTS value measured during the experimental work was 142.69 MPa at 1000 rpm, 40 mm min?1, and ?1.68 mm tool offset. The highest joint efficiency obtained was 82% compared with the AA6101 UTS value. RSM adapted for this work was 92% accurate and satisfactory. 2023 The Author(s). Published by IOP Publishing Ltd. -
Design and Optimization of Friction Stir Welding of Al-Cu BUTT Joint Configuration using Taguchi Method
Friction stir welding (FSW) is a solid-state welding technique in which the joint quality was predominantly subjected to heat formation throughout the metal welding process. The weld joint produced from FSW was better than the other fusion welding process. In this research, the base plates AA6101 and C11000 of 5 mm thickness were joined using the hardened oil-hardened non-shrinkable steel(OHNS) tool by the FSW method. The design of experiment (DOE) was used to optimize the input parameters such as tool rotational speed (rpm), feed rate (mm/min), and tool pin offset (mm) on output parameter ultimate tensile strength (UTS). The design of experiment (DOE) was carried out by employing a Taguchi L9 orthogonal array, three factors, and three levels for obtaining a quality joint with good strength. The results of nine trial runs from the Taguchi experimental approach were formulated and analyzed using the statistical tool analysis of variance (ANOVA) using MINITAB 19 software. ANOVA analysis was employed to find the contribution of the input parameters toward the output. The optimized input process parameters will help to create effective weld joints. This study revealed that tool pin offset towards softer metal at medium tool rotational speed would create joints with the highest UTS. Scanning Electron Microscope (SEM) was applied to investigate the structural changes in the FSW of Al-Cu joints. 2022, Books and Journals Private Ltd.. All rights reserved. -
Geo-spatial crime density attribution using optimized machine learning algorithms
Law enforcement agencies use various crime analysis tools. A large amount of crime data has enabled crime analysis. In this paper, the proposed research methodology uses Kernel Density Estimation (KDE) in a Geographical Information System (GIS) to analyze crime-type data. Bangalore and India newsfeeds are considered for experimental purposes. The paper introduces an optimized KDE machine learning algorithm that detects hotspots, estimates a locations crime rate, and identifies point pattern lows and highs. As a result of the experiment, the proposed methodology identified that the bandwidth of the Geographical information system influences the visualization of crime density. The paper also aids in visually determining the appropriate bandwidth for the problem using an optimized KDE algorithm. We had identified a significant correlation between Newsfeed data and Official Government data, both overall Crime and by crime type. The proposed KDE model achieved a predictive performance of 77.49%. 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management. -
Geospatial crime analysis and forecasting with machine learning techniques
People use social media to engage, connect, and exchange ideas, for professional interests, and for sharing images, videos, and other contents. According to the investigation, social media allows researchers to examine individual behavior features and geographic and temporal interactions. According to studies, criminology has become a prominent subject of study globally, using data gathered from online social media sites such as Facebook, News feed articles, Twitter, and other sources. It is possible to obtain useful information for the analysis of criminal activity by using spatiotemporal linkages in user-generated content. The study refers to the application of text-based data science by gathering data from several news sources and visualizing it. This research is motivated by the abovementioned work from various social media crimes and government crime statistics. This chapter looks at 68 various crime keywords to help you figure out what kind of crime you are dealing with concerning geographical and temporal data. For categorizing crime into subgroups of categories with geographical and time aspects using news feeds, the Naive Bayes classification algorithm is used. For retrieving keywords from news feeds, the Mallet package is used. The hotspots in crime hotspots are identified using the K-means method. The KDE approach is utilized to address crime density and this methodology has solved the difficulties that the current KDE algorithm has. The study results demonstrated equivalence between the suggested crimes forecasting model as well as the ARIMA model. 2022 Elsevier Inc. All rights reserved. -
Crime analysis and forecasting on spatio temporal news feed dataan indian context
Social media is a platform where people communicate, interact, share ideas, interest in careers, photos, videos, etc. The study says that social media provides an opportunity to observe human behavioral traits, spatial and temporal relationships. Based on study Crime analysis using social media data such as Facebook, Newsfeed articles, Twitter, etc. is becoming one of the emerging areas of research across the world. Using spatial and temporal relationships of social media data, it is possible to extract useful data to analyse criminal activities. The research focuses on implementing textual data analytics by collecting the data from different news feeds and provides visualization. This researchs motivation was identified based on relevant work from different social media crime and Indian government crime statistics. This article focuses on 68 types of different crime keywords for identifying the type of crime. Nae Bayes classification algorithm is used to classify the crime into subcategories of classes with geographical factors, and temporal factors from RSS feeds. Mallet package is used for extracting the keywords from the news-feeds. K-means algorithm is used to identify the hotspots in the crime locations. KDE algorithm is used to identify the density of crime, and also our approach has overcome the challenges in the existing KDE algorithm. The outcome of research validated the proposed crime prediction model with that of the ARIMA model and found equivalent prediction performance. The Author(s), under exclusive license to Springer Nature Switzerland AG 2021. -
AI-Based Yolo V4 Intelligent Traffic Light Control System
With the growing number of city vehicles, traffic management is becoming a persistent challenge. Traffic bottlenecks cause significant disturbances in our everyday lives and raise stress levels, negatively impacting the environment by increasing carbon emissions. Due to the population increase, megacities are experiencing severe challenges and significant delays in their day-to-day activities related to transportation. An intelligent traffic management system is required to assess traffic density regularly and take appropriate action. Even though separate lanes are available for various vehicle types, wait times for commuters at traffic signal points are not reduced. The proposed methodology employs artificial intelligence to collect live images from signals to address this issue in the current system. This approach calculates traffic density, utilizing the image processing technique YOLOv4 for effective traffic congestion management. The YOLOv4 algorithm produces better accuracy in the detection of multiple vehicles. Intelligent monitoring technology uses a signal-switching algorithm at signal intersections to coordinate time distribution and alleviate traffic congestion, resulting in shorter vehicle waiting times. 2022 Boppuru Rudra Prathap et al., published by Sciendo. -
Machine Learning and Artificial Intelligence Techniques for Detecting Driver Drowsiness
The number of automobiles on the road grows in lockstep with the advancement of vehicle manufacturing. Road accidents appear to be on the rise, owing to this growing proliferation of vehicles. Accidents frequently occur in our daily lives, and are the top ten causes of mortality from injuries globally. It is now an important component of the worldwide public health burden. Every year, an estimated 1.2 million people are killed in car accidents. Driver drowsiness and weariness are major contributors to traffic accidents this study relies on computer software and photographs, as well as a Convolutional Neural Network (CNN), to assess whether a motorist is tired. The Driver Drowsiness System is built on the Multi-Layer Feed-Forward Network concept CNN was created using around 7,000 photos of eyes in both sleepiness and non-drowsiness phases with various face layouts. These photos were divided into two datasets: training (80% of the images) and testing (20% of the images). For training purposes, the pictures in the training dataset are fed into the network. To decrease information loss as much as feasible, backpropagation techniques and optimizers are applied. We developed an algorithm to calculate ROI as well as track and evaluate motor and visual impacts. 2022 Boppuru Rudra Prathap et al., published by Sciendo. -
Machine Learning and Artificial Intelligence Techniques for Detecting Driver Drowsiness
The number of automobiles on the road grows in lock-step with the advancement of vehicle manufacturing. Road accidents appear to be on the rise, owing to this growing proliferation of vehicles. Accidents frequently occur in our daily lives, and are the top ten causes of mortality from injuries globally. It is now an important component of the worldwide public health burden. Every year, an estimated 1.2 million people are killed in car ac-cidents. Driver drowsiness and weariness are major con-tributors to traffic accidents this study relies on computer software and photographs, as well as a Convolutional Neural Network (CNN), to assess whether a motorist is tired. The Driver Drowsiness System is built on the Multi-Layer Feed-Forward Network concept CNN was created using around 7,000 photos of eyes in both sleepiness and non-drowsiness phases with various face layouts. These photos were divided into two datasets: training (80% of the images) and testing (20% of the images). For training purposes, the pictures in the training dataset are fed into the network. To decrease information loss as much as feasible, backpropagation techniques and optimizers are applied. We developed an algorithm to calculate ROI as well as track and evaluate motor and visual impacts. 2022, Industrial Research Institute for Automation and Measurements. All rights reserved. -
A pragmatic study on heuristic algorithms for prediction and analysis of crime using social media data
Advancement in technology and Social media has grown to become one amongst the foremost powerful communication channels in human history and this is where individuals are sharing their perspectives, thoughts, suppositions, and feelings. Law enforcement units are having hard time fighting crime with evergrowing population, regional issues and political con-sequences. The adoption of social media data for crime analysis is increasing day by day. Crime analysis can help use the resources wisely. A crime prediction alerts the department at the right time to focus their staff with better equipment in suspected areas. Crime analysis prevents threats to life and money loss in terms of damage. In recent days, the collection of crime data from different heterogeneous sources becomes a primary step for the crime analysis and prediction. In this paper Overview of Heuristic Based Crime Prediction and Analysis algorithms identified by different authors. Also, various sources of social media used for analysis and prediction are also reviewed in detail. This information can be considered for one of the prominent asset for crime investigation through social media data procedure and also, we had identified the different algorithms and research gaps of that algorithms with related to crime analysis and prediction. 2019, Institute of Advanced Scientific Research, Inc. All rights reserved. -
Geospatial crime analysis to determine crime density using kernel density estimation for the indian context
Crime is the most common social problem faced in a developing country. Crime affects the reputation of a nation and the quality of life of its citizens. Crime also affects the economy of the country, increasing the financial burden of the government due to the need for expenditure in the police force and judicial system. Various initiatives are taken by law enforcement to reduce the crime rate. One such initiative, real-time accurate crime predictions can help reduce the occurrence of crime. In this paper, a crime analytics platform is developed, which processes newsfeed data analysis for different types of crimes and identify crime hotspots using Kernel Density Estimation method. This system enables criminologists to understand the hidden relationships between crime and geographical locations. Interactive visualization features are available that enable law enforcement agencies to predict crime. 2020 American Scientific Publishers. -
Twitter sentiment for analysing different types of crimes
Online social media like a twitter play a vital role as it helps to track the Spatialoral on social media data with respect crime rate. With the very fast evolving of users in social media, sentimental analysis has become an excellent source of information in decision making. Twitter is one of the most popular social networking site for communication and a primary source of information. More than 150 million users publish above 500 million 140 character TWEETS each day. Tweets have become a basis for product recommendation using sentimental analysis. This paper explains the approach for analyzing the sentiments of the users about a particular crime event tweets posted by the active users. The results so obtained will let you know about the change in the public opinion about the crime events whether it's positive or negative and to find out emotions on different types of crimes. 2018 IEEE. -
Polarity detection on real-time news data using opinion mining
Sentimental Analysis or Opinion Mining plays a vital role in the experimentation field that determines the users opinions, emotions and sentiments concealing a text. News on the Internet is becoming vast, and it is drawing attention and has reached the point of adequately affecting political and social realities. The popular way of checking online content, i.e. manual knowledge-based on the facts, is practically impossible because of the enormous amount of data that has now generated online. The issue can address by using Machine Learning Algorithms and Artificial Intelligence. One of the Machine Learning techniques used in this is Naive Bayes classifier. In this paper, the polarity of the news article determined whether the given news article is a positive, negative or neutral Naive Bayes Classifier, which works well with NLP (Natural Language problems) used for many purposes. It is a family of probabilistic algorithms that used to identify a word from a given text. In this, we calculate the probability of each word in a given text. Using Bayes theorem, they are getting the probabilities based on the given conditions. Topic Modeling is analytical modelling for finding the abstract of topics from a cluster of documents. Latent Dirichlet Allocation (LDA) is a topic model is used to classify the text in a given document to a specified topic. The news article is classified as positive or negative or neutral using Naive Bayes classifier by calculating the probabilities of each word from a given news article. By using topic modelling (LDA), topics of articles are detected and record data separately. The calculation of the overall sentiment of a chosen topic from different newspapers from previously recorded data done. 2020 The authors and IOS Press. -
Prediction of Answer Keywords using Char-RNN
Generating sequences of characters using a Recurrent Neural Network (RNN) is a tried and tested method for creating unique and context aware words, and is fundamental in Natural Language Processing tasks. These type of Neural Networks can also be used a question-answering system. The main drawback of most of these systems is that they work from a factoid database of information, and when queried about new and current information, the responses are usually bleak. In this paper, the author proposes a novel approach to finding answer keywords from a given body of news text or headline, based on the query that was asked, where the query would be of the nature of current affairs or recent news, with the use of Gated Recurrent Unit (GRU) variant of RNNs. Thus, this ensures that the answers provided are relevant to the content of query that was put forth. Copyright 2019 Institute of Advanced Engineering and Science. All rights reserved. -
The Influence of Cartoons Soundscape Irrelevant Sound Effects on young children's Auditory Processing and Working memory skills
Background: Irrelevant sound or speech effect (ISE) affects an individual's serial recall task of visual and auditory presentations. Cartoon soundscape mimics irrelevant sound effect hypothesis. A constant and repeated exposure to cartoons in early childhood should influence children's auditory learning or recall performance. Purpose: To investigate the effects of cartoons' soundscape irrelevant sound effects on young children's auditory processing and working memory skills. Research Design: A cross-sectional study was used to observe the influence of the cartoon soundscape irrelevant sound effects on children. Study sample: Sixty young children with normal hearing in the age range 5-6years were exposed to cartoons (Indian plus Non-Indian) considered for the study. Data Collection and analysis: Pitch Pattern Test (PPT), Duration Pattern Test (DPT), and Corsi-Block working memory apparatus were applied to the participants exposed to cartoons. The data obtained were compared statistically in terms of the groups' performances. Results: There was a significant difference in PPT (p=.023) and DPT (p=.001) between the cartoon exposed and non-exposed groups. In contrast, there was no significant difference between the two groups in Corsi-Block working memory(p>0.05). Conclusion: Cartoon soundscape irrelevant sound or speech affects young children's auditory processing skills. The visual-spatial recall follows a different developmental pattern in young children without recoding to phonological aspects. It is predicted that our study findings might help determine the ill effects of cartoons on the auditory and language development process. 2022 American Academy of Audiology. All rights reserved. -
Non-Fungible Token (NFT): Bubble or Future in the World of Block Chain Technology
The introduction of blockchain technology entering into human existence, which is a reinforcement of the cryptocurrency space, is both a concern and an opportunity. The main motivation underlying such an invention is conditional transparency and the unmatched ability to protect people against data destruction. The collecting drive of NFTs is profitable and also has sparked curiosity, with everyone vying for the first piece of the package, increasing the future Value of an NFT, as it is a very new topic about NFT using block-chain technology. It is something quite about a flurry of blockchain technological stories that leave us wondering. In this research paper, we explained the new emerging Non-Fungible Token (NFT), its uses, and implications. 2023 American Institute of Physics Inc.. All rights reserved. -
Leveraging Ensemble Methods for Accurate Prediction of Customer Spending Scores in Retail
This study primarily aims to estimate consumer spending trends in a retail context. The goal is to identify the best model for predicting Purchasing Scores, which indicate customer loyalty and potential income, using demographic and financial data. The dataset included information about customers' age, gender, and annual income, and the objective was to analyze their Spending Scores. Several regression models were tested, including Linear Regression, Random Forest, Gradient Boosting, K-Nearest Neighbors (KNN), and Lasso Regression. To improve the models, we engineered features like Age Squared, Income per Age, and Spending Score per Income. Each model was trained and tested using 3fold cross-validation. We evaluated their performance with Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R-squared (R2) metrics. The results showed significant differences in model performance. The Random Forest model stood out, with the lowest Mean Absolute Error (MAE) of 0.33, Root Mean Square Error (RMSE) of 0.52, and the highest R-squared (R22) score of 0.9997. Gradient Boosting also performed well, achieving a Mean Absolute Error (MAE) of 1.77, Root Mean Square Error (RMSE) of 2.41, and an Rsquared (R2) score of 0.9930. While Linear Regression showed moderate accuracy, KNN and Lasso Regression had higher errors and lower R2 values, indicating less reliable predictions. The findings suggest that ensemble methods, particularly Random Forest, excel at predicting customer Spending Scores. The high accuracy and reliability of this model point to its potential for customer segmentation and targeted marketing strategies, ultimately enhancing customer relationship management and boosting business value. Further refinement and exploration of additional features could further improve these prediction capabilities. 2024 IEEE.