Browse Items (2150 total)
Sort by:
-
Cloud-Based Cataract Recognition System Using Hybrid Classifier Model
One of the key challenges of ophthalmologists is to diagnose the various ranges of ophthalmological illnesses such as diabetic retinopathy, cataract, and glaucoma. Here, cataract disease is identified as the one of the leading and most common ophthalmological problems that occurs due to aging. A computer-assisted cataract detection and diagnosis support system is required by the ophthalmologists to overcome the error that occurs during manual screening process. So, a cloud-based cataract recognition system is proposed using the convolutional neural network with support vector machine classifier model to improve the prediction accuracy, sensitivity, specificity, precision, recall, F1-score, and Mathews correlation coefficient. Moreover, the four-layer convolutional neural network is finetuned with a rich set of features and trained with various network models such as Inception V3, MobileNet, VGG-16, VGG-19, and ResNet-101. Therefore, the proposed hybrid combination of ResNet-101 with support vector machine classifier makes better cataract detection and outperforms the existing classifier models in terms of above-mentioned performance evaluation metrics. Moreover, the proposed hybrid approach provides the better telemedical solution to remote people by providing accurate disease prediction and severity grading such as normal, mild, premature, and severe cataract. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Towards various applications of Big Data and related issues and challenges
A new trend in feature abstraction is Big Data Analysis combined with computational techniques. This includes gathering knowledge from reputable data sources, analyzing information quickly, and forecasting the future. Big data entails vast amounts of data that are challenging to analyze using typical database and software approaches. When using big data applications, a technological hurdle arises when transporting data across several locations, which is quite expensive and necessitates a huge primary memory for storing data for processing. Big data refers to the transaction and interaction of datasets whose size and complexity transcend the usual technical capabilities of acquiring, organizing, and processing data in a cloud environment. This article provides an in depth study of various applications of big data. It also provides a detailed view on various problems and challenges in Big Data. 2021 IEEE. -
Stir Speed and Reinforcement Effects on Tensile Strength in Al-Based Composites
This study focuses on the preparation of Al-based hybrid composites using AA7475 as the main alloy reinforced with two materials, ZrO2 and SiC. The combination of stir-squeeze processing techniques was employed to create various specimens by varying four parameters: Stir-speed, Stir-time, reinforcements, and squeeze pressure. Taguchi design was utilized to generate specimens for analyzing their mechanical properties, specifically tensile strength, hardness, and porosity.The results indicated that the highest porosity (4.44%) was observed in the L16 test, with a combination of 700rpm stir speed, 25 mins stir time, 2wt% reinforcements, and 80MPa squeeze pressure. On the other hand, the lowest porosity (2.61%) was found in the L7 test, with 800rpm stir speed, 30 mins stir time, 2wt% reinforcements, and 100 MPa squeeze pressure.Regarding tensile strength (UTS), the maximum value (285.23MPa) was achieved in the L13 experiment, while the minimum value (187.58 MPa) was observed in the L1 experiment. This variation in UTS can be attributed to the applied load, the strengthening effect of the reinforcements, and the grain size of SiC. 2024 E3S Web of Conferences -
Application of AI in Determining the Strategies for the Startups
Startups face unique challenges in developing viable strategies to create a competitive advantage and achieve long-term success in today's rapid and global business climate. Using tools powered by AI and algorithms, startups may harness massive amounts of data, generate relevant insights, and make intelligent choices across a variety of business processes. The research begins with an examination of the fundamental concepts beneath AI and how they are used in the business sector. It illustrates how organizations may simplify and improve important activities such as market research, customer segmentation, and trend analysis using artificial intelligence (AI), natural language understanding (NLU), as well as predictive analytics. The report also delves into extensive case studies and real-world examples of businesses that have effectively integrated AI into their business decision-making processes. These examples highlight the practical benefits of AI-driven insights that include enhanced resource allocation, customer targeting, as well as operational performance. It highlights the importance of ethical AI methodologies, transparency, and safeguards to ensure unbiased and fair decision-making. Finally, this study demonstrates how AI has the potential to profoundly transform how entrepreneurs design and implement their strategies. By leveraging AI-driven perspectives, startups may handle complex market dynamics with more precision and agility, increasing their chances of enduring and succeeding in a competitive business climate. The study's findings provide a road map for organizations wishing to apply AI in strategic decision-making processes. 2024 IEEE. -
Lung Cancer Detecting using Radiomics Features and Machine Learning Algorithm
Lung Cancer Incidence across the globe is the second leading cancer type tallying to about 2,206,771 during 2020 and is estimated to rise to about 3,503,378 by 2040 for both male and female sexes and for all ages accounting to 11.4% as per Globocan 2020 [1]. It is the leading death-causing cancer. Lung Cancer [2] in broad terms encompasses Trachea, bronchus as well as lungs. Purpose: The study is aimed to understand Radiomics based approach in the identification as well as classification of CT Images with Lung Cancer when Machine Learning (ML) algorithms are applied. Method: CT Image from LIDC-IDRI [4] Dataset has been chosen. CT Image Dataset was balanced and image features by PyRadiomics library were collected. Various ML features classification algorithms are utilized to create models and matrices adopted in judging their accuracies. The models, distinctive capacity is assessed by receiver operating characteristics (ROC) analysis. Result: The Accuracy scores and ROC-AUC values obtained for various Classification Model are as follows, for Ada Boosting, the accuracy score was 0.9993 ROC-AUC was 0.9993 and followed by GBM, the accuracy score was 0.9993, was 0.9992. Conclusion: Extracting texture parameters on CT images as well as linking the Radiomics method with ML would categorize Lung Cancer commendably. 2023 IEEE. -
Kakkot List- An Improved Variant of Skip List
Kakkot list is a new data structure used for quick searching in a well ordered sequence of list like Skip list. This ordered sequence of list is created using linked list data structure and the maximum number of levels here will be limited to log n in all input behavioral cases. The maximum number of items in each level is halved to that of previous levels and thus guarantees a fast searching in a list. The basic difference between Kakkot list and Skip list lies in the creation of levels and decision of when an item has to be included in the higher levels. In skip list the levels are created and items are added to each level during the insertion of an item where as in Kakkot list this will be done at the time of searching an item. This modification have made drastic impact in searching time complexity in the Kakkot list. Another issue in Skip list is that it is not cache friendly and does not optimize locality of reference wherein this problem is also addressed in Kakkot List. 2020 IEEE. -
Machine Learning Based Crime Identification System using Data Analytics
Poverty is known to be the mother of all crimes, and a vast percentage of people in India live below the poverty line. In India, the crime rate is rapidly rising. The police officers must spend a significant amount of time and personnel to identify suspects and criminals using current crime investigation. In this research, the method presented for designing and implementing crime identification and criminal recognition systems for Indian metropolitans is utilizing techniques of data mining. These occurrences are represented by 35 predefined crime attributes. Access to the crime database is protected by safeguards. The pending four subjects are important for crime unmasking, identification and estimation of criminals, and crime authentication, in that order. The detection of crime is investigated with the help of K-Means clustering, which iteratively builds two crime batches based on congruent criminal features. Google Maps is to enhance the k-means visualization. K-Nearest Neighbor classification is used to examine criminal identification and forecasting. This is used for the authentication of the results. The technique benefits society by helping investigative authorities in crime solving and criminal recognition, resulting in lower crime rates. This research study describes a way for creating and deploying crime solving and criminal recognition systems for Indian metro's using data mining tools in this study. The method consists of data evulsion, data pre- processing, clustering, Google map delegation and classification. The first module, data evulsion, retrieves unformed or unrecorded crime datasets from several criminal sources online from 2000 to 2012. In the second module, Data pre-processing cleans, assimilates, and reduces the obtained criminal data into organized 5,038 crime occurrences. Several predefined criminal traits represent these instances. Safeguards are in place to prevent unauthorized access to the crime index. The remaining components are critical for detecting crimes, criminal identity and prediction, and crime verification, in that sequence. The investigation of crimes is investigated using k-means clustering, which gives results repeatedly. 2023 IEEE. -
Artificial Intelligence and Deep Learning Based Brain Tumor Detection Using Image Processing
In the field of medical science, applications that are particularly used for diagnostic purposes, are used in the detection of brain tumors since detecting an error in MRI scanning is becoming a major task for radiologists and requires a lot of their focus. Flaws that are prevalent during tumor detection must be taken care of to avoid further complications. MRI scanning is one of the most recently developing technologies. The radiologist is a key player in the identification of the brain tumor. Radiologists have to check every image perfectly to avoid the errors in identifying the brain tumor. There is a probability that sometimes cerebral fluid may also appear as mass tissue during the MRI scan. The model that is proposed in this research uses a machine learning algorithm which helps to improve the validity of the classification of the images that are taken in MRI scans. The study focuses on having an automated system that carries out an essential role in determining whether a lump is present in the brain or not. The study tries to resolve the primary flaws in detection necessary to evade further complications in MRI images in brain detection. The main aim of this study is to train the algorithm in a more extensive dataset and to check the patient-level validity with the help of various new datasets. 2023 IEEE. -
A Compact Workflow Model for Cloud Computing
Scheduling tasks in the cloud computing environment, particularly for data intensive applications is of great importance and interest. In this paper, we propose a new workflow model presented in a rigorous graph-Theoretic setting. In this new model, we would like to incorporate possible similarities between requisite files which are needed to complete the given set of tasks. We show that it is NP-Complete to compute the make span in this model even with oracle access to the cost of retrieving a file. 2015 IEEE. -
Evaluation of machine and deep learning models for utility mining-based stock market price predictions
Considering the extreme volatility of stock market returns and hazards, accurate price prediction has attracted the attention of both financial institutions and regulatory bodies. Stocks, due to their historically strong returns, have long been considered by investors to be an excellent asset allocation strategy. Predicting stock prices has never ceased being a hot topic of study. Many early-day economists sought to foretell future stock values. In subsequent years, as computer technology has advanced rapidly and mathematical theory has been extensively studied, it has been shown that mathematical models, like the time series model, may be very effective in predicting due to their simplicity and superiority. Over time, the time series model is put into practice. Over time, the horizon widened. Support vector machines and other ML techniques have challenges when applied to stock data because of its non-linearity. In subsequent years, thanks to advancements in deep learning, models like RNN and LSTM Neural Networks were able to analyze non-linear input, remember the sequence, and remember valuable information,Stock data forecasting cannot be done without it. 2024 Author(s). -
English to Hindi Translation System Using Hybrid Techniques
Good communication is critical for overcoming cultural and linguistic divides in today's internationalized society. An essential communication component is the Translation of written materials, primarily academic papers, from one language into another. This abstract focuses on the research involved in translating academic publications from Hindi to English. Translating Hindi academic papers into English is naturally hard due to the significant linguistic and cultural differences between the two languages. The proposed work provided an analytical analysis of various models used in language translation, including the seq-to-seq model, MT5, and LSTM, with the help of BLEU score, Learning rate, and average loss. MT5 model outshines others in terms of an average loss of 4.75; meanwhile, LSTM has an average loss of 5.56, and the seq-to-seq model has an average loss of 6.09, implying weaker Translation. 2024 IEEE. -
State-of-art Techniques for Classification of Breast Cancer: A Review
Cancer is an unexpected and unclear disease that puts many people at risk. Breast cancer has surpassed prostate cancer as the most common cancer in women, as well as the main cause of cancer-related mortality in women. Breast cancer rates have been rising in India for several years, with 100,000 new cases recorded each year. In India, there are up to one million breast cancer patients at any given moment. The survival rate of breast cancer has increased in recent years as a result of advances in technology, effective treatment, and medical care delivery. It extends the lives of the sufferers and improves their quality of life. Breast cancer can be detected using a variety of imaging methods. Radiologists can utilize a computer-aided diagnostic technique to discover and diagnose irregularities earlier and more quickly. Many Computer-Aided Diagnosis methods have been developed to identify breast cancer in its early stages using mammography images. The computer aided diagnostics systems mostly focus on identifying and detecting breast nodules. Staging breast cancer at its detection needs to be focused on, as the treatment is based on the stage of cancer. As a result, this study focuses on producing evaluations on computer aided diagnostics approaches for segmenting nodules and identifying different stages of breast cancer, thereby assisting radiologists in assessing the illness. 2022 IEEE. -
Demography-Based Hybrid Recommender System for Movie Recommendations
Recommender systems have been explored with different research techniques including content-based filtering and collaborative filtering. The main issue is with the cold start problem of how recommendations have to be suggested to a new user in the platform. There is a need for a system which has the ability to recommend items similar to the users demographic category by considering the collaborative interactions of similar categories of users. The proposed hybrid model solves the cold start problem using collaborative, demography, and content-based approaches. The base algorithm for the hybrid model SVDpp produced a root mean squared error (RMSE) of 0.92 on the test data. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
On the k-Forcing Number of Some DS-Graphs
Amos et al. introduced the notion of k-forcing number as a generalization of Zero forcing number and is denoted by Fk(G) where k> 0 is any positive integer, the k -forcing number of a graph is the minimum cardinality among all k -forcing sets of a graph G. In this paper, many bounds for k -forcing number of degree splitting graph DS(G) for different graph classes are found. We evaluate the value of k -forcing number of degree splitting graph of some of the Cartesian product graph for different values of k. Also we observed that for Tur graph Tn , t, upper and lower bound is given by, Fk(Tn , t) ? Fk(DS(Tn , t) ) ? Fk(Tn , t) + 1. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Voltage stability analysis using L-index under various transformer tap changer settings
Voltage stability is a major problem in power system which depended on many factors like improper load forecasting, generator outage, line fault and shortage of reactive power supply etc. For a secure and economic power system operation voltage stability should be maintained within permissible limit. Voltage stability is a measure of whole power system quality. Voltage stability studies can done by analyzing reactive power production, transmission of power and consumption. In this paper voltage stability analysis of an IEEE 14 bus system is done by calculating L-index of the buses. From load flow studies optimized voltage is chosen, and by using these voltage values L-index is calculated. From the calculated L-index values we can find out vulnerable buses. How the transformer tap changing effect the voltage stability is also calculated here. 2016 IEEE. -
An Intelligent Portfolio Management Scheme Based On Hybrid Deep Reinforcement Learning and Cumulative Prospective Approach
Stock markets retain an extensive role towards economic growth of diverse countries and it is a place where investors invest assured amount to earn more profit and the issuers pursue the investors for project investing. However, it is deliberated as a challenging task to buy and sell because of its explosive and complex nature. The existing portfolio optimization models are primarily focused on just improving the returns whereas, the selection of optimal assets is least focused. Hence, the proposed research article focuses on the integration of stock prediction with the portfolio optimization model (SPPO). Initially, the stock prices for the next period are predicted using the hybrid deep reinforcement learning (DRL) model. Within this prediction model, the gated recurrent unit network (GRUN) model is utilized to simulate the interactions of the agent with the environment. The best actions in the prediction model are determined throughout the prediction process using the quantum differential evolution algorithm (Q-DEA). After the prediction of best assets, the optimal portfolio with the best assets is selected using the cumulative prospect theory (CPT) model. The work will be implemented in python and evaluated using the NIFTY-50 Stock Market Data (2000 -2021) dataset. Minimal error rates of 0.130, 0.114, 0.148 and 0.153 is obtained by the proposed model in case of MSE, MAE, RMSE and MAPE. 2024 IEEE. -
Enhanced Process Model and Analysis of Risk Integration in Software effort estimation
The development of software within the estimated effort is remaining as a challenging task. The process of effort estimation is a critical activity in a software project, where effort estimates are utilized to arrive at the schedule, resources, and cost. Though many software effort estimation techniques exist, effort overrun occurs in a project. Identification of risks and their consideration in software lifecycle activities play a significant role in the successful execution of a software project. It would be required to account for uncertainty and the key factors that contribute to it. This study focuses on the need to include project risk score in the software effort estimation process to arrive at better effort estimates. This paper depicts the standard and enhanced process frameworks for estimation of software development efforts. A multi-layer perceptron model was built and the results indicated the relevance of considering project risk score in the effort estimation process. The usage of an enhanced gradient boosting technique for predictive modelling revealed a decrease in standard deviation of the residuals, thus indicating a better fit for the effort estimation model through integration of risks. 2019 IEEE. -
CBMIR: Content Based Medical Image Retrieval Using Hybrid Texture Feature Extraction Method
Due to the revolution of digital era in the medical domain at various hospitals across the world, the online users on the internet access have been increased. So the amount of collections of digitized medical images has grown rapidly and continuously. As well it is ratting significant to mention that the images are globally used by radiologists, professors in medical colleges and Lab technicians, etc. These Images are increasingly applied to communicate information about patient history. In this context, there is a necessity to develop appropriate systems to manage these medical images in storage and retrieval for diagnosis of the patient information. Another big issue is the convolution of image data and that can be interpreted in different ways. In order to manipulate these data and establish policies to its content is very tedious job. This will raise another big question. These issues motivated the researchers to give more focus on the image retrieval area whose goal is trying to solve those problems to provide an efficient retrieval system to the user community. In this perspective, this work has been proposed to facilitate radiologists, professors in medical colleges, lab technicians, and all other medical image user communities for their purpose for easy access from the remote location. 2022 IEEE. -
A Review on Artificial Intelligence Techniques for Multilingual SMS Spam Detection
With social networks increased popularity and smartphone technology advancements, Facebook, Twitter, and short text messaging services (SMS) have gained popularity. The availability of these low cost text-based communication services has implicitly increased the intrusion of spam messages. These spam messages have started emerging as an important issue, especially to short-duration mobile users such as aged persons, children, and other less skilled users of mobile phones. Unknowingly or mistakenly clicking the hyperlinks in spam messages or subscribing to advertisements puts them under threat of debiting their money from either the bank account or the balance of the network subscriber. Different approaches have been attempted to detect spam messages in the last decade. Many mobile applications have also evolved for spam detection in English, but still, there is a lack of performance. As English has been completely covered under natural language processing, other regional languages, such as Urdu and Hindi variants, have specific issues detecting spam messages. Mobile users suffer greatly from these issues, especially in multilingual countries like India. Thus, this paper critically reviews the artificial intelligence-based spam detection system. The review lists out the existing systems that use machine and deep learning techniques with their limitations, merits, and demerits. In addition, this paper covers the scope for future enhancements in natural language processing to efficiently prevent spam messages rather than detect spam messages. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Multi-lingual Spam SMS detection using a hybrid deep learning technique
Nowadays, the incremental usage of mobile phones has made spam SMS messages a big concern. Sending malicious links through spam messages harms our mobile devices physically, and the attacker might have a chance to steal sensitive information from our devices. Various state-of-the-art research works have been proposed for SMS spam detection using feature-based, machine, and deep learning techniques. These approaches have specific limitations, such as extracting and selecting signifi-cant and quality features for efficient classification. Very few deep learning techniques are only used for classifying spam detection. Moreover, the benchmark spam datasets written in English are mostly used for evaluation. Very few papers have detected spam messages for other languages. Hence, this paper creates a multilingual SMS spam dataset and proposes a hybrid deep learning technique that combines the Convolutional Neural Network and Long Short-Term Memory (LSTM) model to classify the message dataset. The performance of this proposed hybrid model has been compared with the baseline deep learning models using accuracy, precision, recall, and F1-score metrics. 2022 IEEE.