Browse Items (2150 total)
Sort by:
-
Content-Restricted Boltzmann Machines for Diet Recommendation
Nowadays, society is leading towards an unhealthy and inactive and lifestyle. Recent studies show the rapid growth of people suffering from diseases caused due to unhealthy lifestyles and diet. Considering this, recognizing the right type and amount of food to eat with a suitable exercise set is essential to obtain good health. The proposed work develops a framework to recommend the proper diet plans for thyroid patients, and medical experts validate results. The experiments results illustrate that the proposed Content-Restricted Boltzmann Machines (Content-RBM) produces more relevant recommendations with content-based information. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Physical Unclonable Function and OAuth 2.0 Based Secure Authentication Scheme for Internet of Medical Things
With ubiquitous computing and penetration of high-speed data networks, the Internet of Medical Things (IoMT) has found widespread application. Digital healthcare helps medical professionals monitor patients and provide services remotely. With the increased adoption of IoMT comes an increased risk profile. Private and confidential medical data is gathered across various IoMT devices and transmitted to medical servers. Privacy breach or unauthorized access to personal medical data has far-reaching consequences. However, heterogeneity, limited computational resources, and lack of standardization in authentication schemes prevent a robust IoMT security framework. This paper introduces a secure lightweight authentication and authorization scheme. The use of the Physical Unclonable Function (PUF) reduces pressure on computational resources and establishes the authenticity of the IoMT. The use of OAuth 2.0 open standard for authorization allows interoperability between different vendors. The resilience of the model to impersonation and replay attacks is analyzed. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Variable parametric test to improve the machinability of Inconel-718 using Tungsten Carbide tool
The Inconel-718 is a nickel based super alloy containing an old age hardening alloy of nickel-chromium as addition which provides increased strength without its decrease in ductility. It is known as a difficult to cut material due to certain properties like high thermal resistance, high creep, corrosion resistance having the capability of retaining toughness and strength at high temperatures. Inconel-718 has a large number of applications in the world of manufacturing such as aircraft gas turbines, steam turbine power plants, reheaters and reciprocating engines. Due to such superior quality functions, its machining becomes more challenging for which Tungsten Carbide is one of the tools to improve the machinability to 2.64%. In this paper, parametric tests has been carried out in CNC machining to determine the tool performance and improve the machining conditions. 2021 Elsevier Ltd. All rights reserved. -
Ear Recognition Using Pretrained Convolutional Neural Networks
Ear biometrics, which involves the identification of a person from an ear image, is challenging under unconstrained image capturing scenarios. Studies in Ear biometrics reported that the Convolutional Neural Network is a better alternative to classical machine learning with handcrafted features. Two major concerns in CNN are the requirement of enormous computing resources and large datasets for training. The pretrained network concept helps to use CNN with smaller datasets and is less demanding on hardware. In this paper, three pre-trained CNN models, AlexNet, VGG16, and ResNet50 are used for ear recognition. The fully connected classification layers of the nets are trained with AWE, an unconstrained ear dataset. Alternatively, the CNN layers output (the CNN features) are extracted, and an SVM classification model is built. To improve the classification accuracy, the training dataset size is increased through data augmentation. Data augmentation improved the classification accuracy drastically. The results show that ResNet50, with the fully connected classification layer, results in higher accuracy. 2021, Springer Nature Switzerland AG. -
User Authentication with Graphical Passwords using Hybrid Images and Hash Function
As per human psychology, people remember visual objects more than texts. Although many user authentication mechanisms are based on text passwords, biometric characteristics, tokens, etc., image passwords have proven to be a substitute due to its ease of use and reliability. The technological advancements and evolutions in authentication mechanisms brought greater convenience but increased the probability of exposing passwords through various attacks like shoulder-surfing, dictionary, key-logger, and social engineering attacks. The proposed methodology addresses these vulnerabilities and ensures to keep up the usability of graphical passwords. The system displays hybrid images that users need to recognize and type the randomly generated alphanumeric or special character values associated with each of them. A mechanism to generate One Time Password (OTP) is included for additional security. As a result, it is difficult for an attacker to capture and misuse the password. 2021, 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. -
Document Classification for Recommender Systems Using Graph Convolutional Networks
Graph based recommender systems have time and time again proven their efficacy in the recommendation of scientific articles. But it is not without its challenges, one of the major ones being that these models consider the network for recommending while the class and domain of the article go unnoticed. The networks that embed the metadata and the network have highly scalable issues. Hence the identification of an architecture that is scalable and which operates directly on the graph structure is crucial to its amelioration. This study analyses the accuracy and efficiency of the Graph Convolutional Networks (GCN) on Cora Dataset in classifying the articles based on the citations and class of the article. It aims to show that GCN based networks provide a remarkable accuracy in classifying the articles. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Design and Analysis of Vortex Bladeless Wind Turbine
Vortex bladeless turbine antiquates the conventional wind turbine and adopts a radically innovative and novel approach to captivate the moving wind energy. This device effectively captures the energy of vorticity, an aerodynamic instability condition. As the wind passes a structure, the flow steers and cyclical patterns of vortices are generated. Once the strength of wind force is suffice, the structure starts vibrating and reaches resonance. Vortex bladeless is a vortex induced vibration resonant power generator. It harnesses wind energy from a phenomenon of vorticity, called vortex shedding effect. Clearly bladeless technology consists of a cylinder fixed vertically on an elastic rod, instead of tower, nacelle and blades which are the crucial parts of a conventional wind turbine. The cylinder oscillates on a specifically mentioned wind range, which then generate electricity through an alternator and a tuning system. In this paper the vortex turbine is designed with certain existing parameters of dimensions in Solidworks and the same is analyzed for different materials and dimensions of mast, which is an important part in the vortex turbine. Also various performance parameters like displacement, frequency etc. are also compared among different models. 2021 Elsevier Ltd. All rights reserved. -
Rice Yield Forecasting in West Bengal Using Hybrid Model
Agriculture in India is the primary source of revenue, yet farmers still face challenges. The primary goal of agricultural development is to produce a high crop yield. The Datasets collected for the study of real-world time series include a blend of linear and nonlinear patterns. A mixture of linear and non - linear models, rather than a single linear or non - linear model, gives a more precise forecasting models for time series data. The ARIMA and ANN prediction models are combined in this paper to create a Hybrid model. This model is used to predict rice yield for all 18 West Bengal districts during the Kharif season, based on 20years of information(20002019) collected from various sources such as India Meteorological Department, Area, and production Statistics, DAV from NASA, etc. The hybrid model aims to enhance efficiency indicators such as MSE, MAE, and MAPE, demonstrating excellent performance for rice yield prediction in all the districts of West Bengal. In the future, it can be applied to other crops that can support farmers in their farming. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Prediction of Depression in Young Adults Using Supervised Learning Algorithm
Over the years, mental health has achieved an essential role in the pertinent development of a human being, and a large part of the population is affected by it. The most commonly affected community being college-going students, and the most common disorders being Anxiety and Depression. Depression is a leading cause of suicide in individuals, where suicide is the second most prevailing reason for death among 1529-year-olds. This study aims to identify the different reasons and other factors associated with depression to predict and determine whether an individual faces depressive disorders. For this research purpose, the most appropriate classifier is selected. The absolute accuracy of the proposed model is 91.17%, i.e., the model can correctly predict whether an individual has depression 91.17% of the time. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
The Preservative Technology in the Inventory Model for the Deteriorating Items with Weibull Deterioration Rate
An EOQ model for perishable items is presented in this study. The deterioration rate is controlled by preservative technology. This technology only enhances the life of perishable items. So, retailers invested in this technology to get extra revenue. The Weibull deterioration rate is considered for the ramp type demand. Shortages consider partially backlogged, and discount is provided to loyal customers. The concavity of the profit function is discussed analytically. Numerical examples support the solution procedure; then, Sensitivity analysis is applied to accomplish the most sensitive variable. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Reimagining the Digital Twin: Powerful Use Cases for Industry 4.0
Novel cohorts of information technologies are transformation and upgrading the global manufacturing sector. The analysis of product procedure in discrete globe might furnish significant perceptions resting on scheme routine which may change manufacturing product design. Digital twin predictive analysis on both historical and future performances of an organizations physical resources leading to proficient industry functioning. In digital twin, cloud-based virtual image of industrial asset is maintained throughout the lifecycle which can be accessed at any time. Digital twin enhances the degree and functions of manufacturing world by integrating with the physical world. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Automated Organic Web Harvesting on Web Data for Analytics
Automated Web search and web data extraction has become an inevitable part of research in the area of web mining. The web scraping has immense influence on ecommerce, market research, web indexing and much more. Most of the web information is presented in an unstructured or free format. Web scraping helps every user to retrieve, analyze and use the data suitably according to their requirement. There exist different methodologies for web scraping. Major web scraping tools are rule based systems. In the proposed work, an automated method for web information extraction using Computer Vision is proposed and developed. The proposed automated web scraping method comprises of automated URL extraction virtual extraction of required data and storing the data in a structured format which is useful in market research. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Comparison of Full Training and Transfer Learning in Deep Learning for Image Classification
The deep learning algorithms on a small dataset are often not efficient for image classification problems. Make use of the features learned by a model trained on large similar dataset and saved for future reference is a method to solve this problem. In this work, we present a comparison of full training and transfer learning for image classification using Deep Learning. Three different deep learning architectures namely MobileNetV2, InceptionV3 and VGG16 were used for this experiment. Transfer learning showed higher accuracy and less loss than full-training. According to transfer learning results, MobileNetV2 model achieved 98.96%, InceptionV3 model achieved 98.44% and VGG16 model achieved 97.405 as highest test accuracies. The full-trained models did not achieve as much accuracy as that of transfer learning models on the same dataset. The accuracies achieved by full-training for MobileNetV2, InceptionV3 and VGG16 are 79.08%, 73.44% and 75.62% respectively. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
A Systematic Review of Challenges and Techniques of Privacy-Preserving Machine Learning
Machine learning (ML) techniques are the backbone of Prediction and Recommendation systems, widely used across banking, medicine, and finance domains. ML techniques effectiveness depends mainly on the amount, distribution, and variety of training data that requires varied participants to contribute data. However, its challenging to combine data from multiple sources due to privacy and security concerns, competitive advantages, and data sovereignty. Therefore, ML techniques must preserve privacy when they aggregate, train, and eventually serve inferences. This survey establishes the meaning of privacy in ML, classifies current privacy threats, and describes state-of-the-art mitigation techniques named Privacy-Preserving Machine Learning (PPML) techniques. The paper compares existing PPML techniques based on relevant parameters, thereby presenting gaps in the existing literature and proposing probable future research drifts. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Early Prediction of Plant Disease Using AI Enabled IOT
India is an industrialized country, and about 70% of the residents rely on agriculture. Leaves are damaged by chemicals, and climates issues. An unknown illness is found on plants leads to the lowering of quality of produced. Internet of Things is a practice of reinventing the wheel agriculture by enabling farmers to tackle the problems in the industry with practical farming techniques. IoT helps to inform knowledge about factors like weather, and moisture condition. We proposed IoT, ML, and image processing based method to identify the infection. IOT enabled camera to capture the image then required region of interest is extracted. After ROI extraction, image is enhanced to remove the unwanted details form the image and to improve image quality. We compute image features. At the end we do the classification which is a twostep process training and testing and done by SVM. Our proposed method gives 92% accuracy. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Optimization of friction stir welding parameters during joining of AA3103 and AA7075 aluminium alloys using Taguchi method
This paper investigates the optimization of input parameters for the friction stir welding of AA3103 and AA7075 aluminium alloys. The properties of base materials AA3103 are non-heat-treatable alloy, which is having good weldability while AA 7075 is having higher strength. Therefore, the welding of these aluminium alloys will produce superior mechanical properties. Friction stir welding is a rapidly growing welding process which is being widely used in marine, automobile and aerospace industries. Rather than its widespread use, this type of welding has several advantages over normal welding processes like low production of fume, no consumable electrodes are used and can be used in any position. In this paper, optimization of input parameters were conducted based on Taguchi method using the L9 orthogonal array. There were nine experimental runs in total after creating the L9 orthogonal array table in MINITAB software. The input parameters selected for optimization are tool rotation speed, feed rate, tool pin profile the output parameters which are optimized hardness, tensile strength, impact strength. The ANOVA analysis was carried out in the Qualitek 4 software to find out the percentage influence of input parameters on the output parameters. This research work was carried out to find the optimized condition to carry out friction stir welding of above mentioned aluminium alloys. 2021 Elsevier Ltd. All rights reserved. -
Use of zeolite and industrial waste materials in high strength concrete - A review
Concrete is widely used in construction material by the construction industry. It is considered as a vital material because of its properties. Different grades of concrete (M10, M20, M30, M40, M50, M60and M70) are used in construction and are chosen based on the requirements. Higher grade concrete requires cement of different properties. The manufacturing process of cement, releases a huge amount of Carbon footprints. To reduce the emission of CO2, usage of virgin cement can be minimized by partially replacing with pozzolanic materials or industrial wastes like zeolite, metakaolin, silica fume and fly ash. These materials improve the durability and strength of concrete by filling the pores and reduce the porosity and permeability of the concrete without compromising on the desired properties. For sustainable development and protecting the environment, enormous research has been done on concrete by using various industrial waste materials. This article is a state-of-the-art review of research on the use of industrial waste materials to produce High Strength Concrete (HSC). Different materials were studied to prepare HSC by using distinct methods. Different experimental tests were conducted on concrete when cement is partially replaced with industrial waste materials and are compared with conventional concrete. It is observed that the partial replacement of cement with zeolite, metakaolin, fly ash, and silica fume, the properties of concrete increases up to certain age and mixing proportions when compared to conventional concrete. It is observed that there is limited research was done on zeolite with the combination of industrial waste materials for health analysis of the structures at different w/c ratios for large production. So, further investigation is needed on the technical, environmental, economic aspects and educating the public through the use of industrial waste materials as a sustainable approach. 2021 Elsevier Ltd. All rights reserved. -
A Study on the Influence of Geometric Transformations on Image Classification: A Case Study
The present research work involves the study of the geometrical transformations which influences the training and validation accuracies of machine learning models. For the study, rice plant leaf disease dataset of 2096 images consisting of 4 classes with 523 images per class were used. The dataset subjected to 24 models out of which three models namely - DenseNet201, Densenet169 and InceptionResNetV2 are selected based on highest training accuracy and less difference between training and validation accuracy. To evaluate the performance of the selected three models, loss functions and accuracies have been computed. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Movie Success Prediction from Movie Trailer Engagement and Sentiment Analysis
The diverse movie industry faces many challenges in the promotion of the product across different demographics. Movie trailer engagements provide valuable information about how the audience perceives the movie. This information can be used to predict the success of the upcoming movie before it gets released. The previous research works were mainly concentrating on Hindi language movies to predict success. The current research paper includes the success prediction of movies other than Hindi. This paper aims to analyze various Machine Learning models performance and select the best performing model to predict movie success. The developed model can efficiently classify successful and unsuccessful movies. For the current research, the data is collected from various sources through web scrapping and API calls in Sacnilk, The Movie Database (TMDB), YouTube, and Twitter. Different machine learning classification models such as Random Forest, Logistic Regression, KNN, and Gaussian Nae Bayes are tested to develop the best-performing prediction model. This research can help moviemakers to understand the popularity of the movie among the viewers and decide on an efficient promotional strategy to make the movie more successful. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.