Search

Recent years have witnessed the rise of supercapacitor as effective energy storage device. Specifically, carbon-based electrodes have been experimentally well studied and used in the fabrication of supercapacitors due to their excellent electrochemical properties. Recent publications have reported the use of Machine Learning (ML) techniques to study the correlation between the structural features of electrodes and supercapacitor performance metrics. However, the poor R-squared values (i.e., large deviations from the ideal value of unity) and large RMSE values reported in these works reflect the lack of accurate models’ development. This work reports the development and utilization of highly tuned and efficient ML models using hyperparameter tuning, that give insights into correlation between the structural features of electrodes and supercapacitor performance metrics namely specific capacitance, power density and energy density. Artificial Neural Networks (ANN) and Random Forest (RF) models have been employed to predict the various in-operando performance metrics of carbon-based supercapacitors based on three input features such as mesopore surface area, micropore surface area and scan rate. Experimentally measured values of these parameters used for training and testing these two models have been extracted from a set of research papers reported in literature. The optimization techniques and various tuning methodologies adopted for identifying model hyperparameters are discussed in this paper. The R2 values obtained for prediction of specific capacitance, power density and energy density using RF model are in the range from 0.8612 to 0.9353 respectively, while the RMSE values of the above parameters are 18.651, 0.2732 and 0.5764 for respective input parameters. Similarly, the R2 values obtained for prediction of specific capacitance, power density and energy density using ANN model are in the range from 0.9211 to 0.9644 respectively, while the RMSE values of the above parameters are 18.132, 0.1601 and 0.5764 for respective input parameters. Thus, the highly tuned ANN and RF models depict higher R-squared and lower RMSE values in comparison to those previously reported in literature, thereby demonstrating the importance of hyperparameter tuning and optimization in building accurate and reliable computational models.

With advances in the mobile communications, many service-related tasks can be made quickly and easily. Mobile banking is one such service that has eliminated the need for a consumer to go to a branch to carry out many common transactions. In a country like India, where the last mile reach through brick-and-mortar banking facilities, mobile phones can complement the reach. This paper describes how mobile cloud architecture can be employed for banking and services to customers to enhance their banking experience as well as ensuring information security. This paper focuses on cloud-based risk architecture for banking solutions to address various issues related to mobile banking such as processing speed and storage capacity. Improved random forecast algorithm is used for the evaluation of the system. This proposed system achieves 99% of the system.

The primary objective of this particular paper is to classify the health-related data without feature extraction in Machine Learning, which hinder the performance and reliability. The assumption of our work will be like, can we able to get better result for health-related data with the help of Tree based Machine Learning algorithms without extracting features like in Deep Learning. This study performs better classification with Tree based Machine Learning approach for the health-related medical data. After doing pre-processing, without feature extraction, i.e., from raw data signal with the help of Machine Learning algorithms we are able to get better results. The presented paper which has better result even when compared to some of the advanced Deep Learning architecture models. The results demonstrate that overall classification accuracy of Random Forest, XGBoost, LightGBM and CatBoost, Tree-based Machine Learning algorithms for normal and abnormal condition of the datasets was found to be 97.88%, 98.23%, 98.03% and 95.57% respectively.

The major challenges, which come across face recognition system, are to find the age and gender in 2D/3D image of the person specifically in cloud environment. This paperis centered on face detection with MAC (Media Access Control) and biometric technology. Face scanning along with machine’s MAC address and biometric technologies has been shown to improve security controls. Face recognition can be used to search and label users and their assigned machines for sensitive purposes. Following that, it was stored in a specific database with their unique ID. In addition, the verification process has begun by comparing the models in the database. Face scanning along with speech and biometric technologies is used to improve security controls. Face recognition system may also be set up in high security machines to improve protection by allowing only registered individuals or others users. Related strategies for determining the age and gender and 2D/3D image from a specific picture are explored, as well as several modern methods for preserving protection. In this paper, the full model is explored independently with security implemented in cloud environment. The proposed model of the paper provides the integrated security features using MAC address of machine and face recognition of the machine user.

Aiming at the problem that the current feature selection algorithm can not adapt to both supervised learning data and unsupervised learning data, and had poor feature interpretability, this paper proposed a hybrid feature selection model based on machine learning and knowledge graph. By the idea of hybridization, this model used supervised learning algorithms, unsupervised learning algorithms and knowledge graph technology to model from the perspective of data features and text features. Firstly, the data-based feature weights were obtained through the machine learning model, and then the text-based weights were obtained by using the knowledge graph technology, and the weight sets are combined to obtain a feature matrix with good explanatory properties that meets both the data and text features. Finally, the case analysis proves that the method proposed in this paper has good effects and interpretability.

Data innovation worldview empowers pervasive access to shared pools of configurable framework assets and more significant level administrations that can be quickly provisioned with minimal administration exertion, regularly over the Internet. Cloud computing provides dynamic asset administrations for the client on request, and it is becoming increasingly popular. Cloud administrations are conveyed from server farms all through the world (as it is otherwise called Networks of systems). Outsider associations are additionally present there, because of which some significant security dangers came and make organize helplessly so the security of cloud information is imperative to keep over Cloud architecture to offer better support, ease administrations, and so forth. Despite that there are a few undermined hazards in cloud computing security. We will be concentrating majorly on cloud computing fraught with threats and how to overcome those threats. This paper is intended to introduce information about the most current attacks on cloud computing, just as safety efforts. The seriousness and impact of these attacks are talked about alongside genuine instances of these attacks. The paper likewise proposes alleviation strategies that can be utilized to decrease or dispose of the risks of the threats talked about. Moreover, the general cloud security proposals are given.

A formal-semantic normative system (FSNS) is presented in this paper. The FSNS allows us to describe the links and functional units of typical information processes. The possibility of using the FSNS to automate graphical-analytical modelling and machine learning is shown. Properties of the FSNS are described. On its basis, a method is developed for specifying a formal alphabet of system connection nodes that has a specific subject content. The rules for manipulating the received alphabetic characters are also presented. The possibility and expediency of building a normative system with increased expressive capabilities is substantiated. This paper provides a brief description of the «Unit-Function-Object» system-object approach. Considered the possibility of using DL for the FSNS in an unambiguous, formalized form. An example of using the FSNS is presented.

Bioenergy is a renewable energy obtained from biomass and its main benefits are the reduction of greenhouse gases and the reduction of waste disposal. A common problem in feasibility analysis is the biodigester location due to biomass availability and its spatial dispersion. Therefore, decisions regarding to biodigester location is an important aspect of feasibility design when multiple biomass producers are involved in clusters. Studies in this research field of biodigester location address different methodologies, such as mathematical programming, multicriteria approach, scenario analysis, and Geographic Information System (GIS). This paper presents the state-of-the-art in the biodigester location decisions considering clustering situations. To accomplish that, a systematic literature review using Prisma approach was carried out. As results, a set of feasibility opportunities was arranged after reading 24 papers, allowing to identify that the main methodologies used for biodigester location decisions were mathematical programming (41,16%), followed by multicriteria approaches (29,16%) and GIS (12,50%).

In order to put an end to the product quality accidents caused by cutting tool breakage or severe wear in machining process, the paper explored an ELM model detection method based on voice recognition. In this paper, firstly it analysed the features of cutting sound signal in time-frequency domain, then discussed the extraction method of tool working state sensitive spectral energy statistical feature based on wavelet packet decomposition, and finally constructed an ELM fast detection model based on sound feature recognition. The experimental results demonstrated that the ELM detection model could achieve higher detection accuracy and faster response time. The simulation results show that the ELM model is effective and applicable in detecting tool wear with the help of sound recognition.

New technologies such as artificial intelligence have developed at a rapid pace in recent years and are increasingly being used in the process of managing news in bulk. The development of deep learning has facilitated unprecedented progress in the field of computing and has opened our eyes to the possibility of using AI for news text classification. In this paper, based on the system requirements analysis, we describe the process of functional modules arising from the requirements analysis, design the internal details of functional modules, including algorithms and detailed principles, and finally obtain a prototype of news text classification software, which results in the pre-design expectations. The research in this paper makes the system development work more concrete, while providing software users, software developers, and analysts and testers with a unified and comprehensive understanding of the system’s functional implementation.

Cloud computing not only provides convenience for users, but also brings certain information security threats to users. How to effectively protect the user's information security has become one of the core issues of network security. This paper analyzes the problems and reasons of network security in cloud computing environment, and points out the main ways to improve network security. This paper analyzes the network security problems in the cloud computing environment, and discusses the information security technology in the network. This paper studies the definition, characteristics, classification and core technology of cloud computing security, analyzes the security risk of cloud computing network, and puts forward the cloud computing network security model and security level evaluation method from different angles. Level experiment results show that the network security model in cloud environment is feasible, computing can fully meet the development requirements of the current network information age, and effectively standardize the network security operation environment. The results show that: in 2016, the number of articles related to cloud computing is at least 2653, and in 2020, the number of articles related to cloud computing is at most 6438.

Nowadays, investigators of phrases and text sentimental orientation identification mainly divided into two ways: semantic comprehension and machine learning. Machine learning needs a lot of training corpus and also cannot handle general field words effectively, meanwhile, semantic comprehension cannot get high scores at precision and recall. Therefore, in this paper, we propose a new fusion method based on our semantic comprehension for judging phrases’ polarity. In this paper, firstly, we modify the knowledge of Hownet, on the basis of its four primitives, the fifth primitive—sentimental primitive was proposed by our research, and annotated it into Hownet manually; secondly, we propose a new phrase sentimental similarity calculation method to compute word’s sentimental value; at last, we integrate transductive learning into this method to identify phrases’ sentimental orientation. The performance of experiment show that, compared with SVM or traditional semantic comprehension, it can get better results.

Targeting at the problem that the existing hybrid database solutions can not simultaneously realize the client’s verification and access function to the encrypted data in the public cloud database and the cross database query function in the heterogeneous database environment, this paper designs an access and integration middleware to realize the client’s verification and access function to the heterogeneous hybrid database in the public cloud and the cross database query function. Access and integration middleware is composed of verification access component and heterogeneous database data integration method. The verification and access component reconstructs the protocol interaction process based on MySQL protocol, and encrypts and decrypts the encrypted data in the public cloud database through the existing hybrid database scheme, so as to realize the client’s verification and access to the encrypted data in the cloud database. The heterogeneous database data integration method unifies the data from different types of databases, and then integrates them into the same database, and uses this database for query operation to realize cross database query. This paper first analyzes the hybrid database system model, then designs from the aspects of MySQL protocol analysis, verification access, data integration, and finally analyzes the access and integration middleware from the aspects of functionality and security.

In present time, meta-heuristic algorithms have been widely adopted for solving diverse classes of optimization problems. These procedures/algorithms have advantage over classical algorithms due to strong exploration and exploitation capabilities. In recent time, a Bat algorithm is developed on the basis of echolocation characteristics of bat. This algorithm has applicability to solve number of constrained and unconstrained optimization problems. The bat algorithm gives state of art results as compared to same class of metaheuristics. The objective of this paper is to explore bat algorithm in terms of modifications, improvements, search strategies and nature of problems being solved. Hence, this paper presents the mini review on bat algorithm. Further, the applicability of bat algorithm in diverse filed has been discussed.

In recent years, big data is used in finance, education, biotechnology, and other fields. In the field of education, thanks to the overall characteristics of big data, it can be applied to understand the objective learning rules of students, to provide effective decision-making suggestions for educators. With the development of electronic communication and media technology, Massive Open Online Courses (MOOC), which contains a large number of user behaviors and information, has gradually become one of the mainstream ways of education. However, MOOC have also been criticized for their high dropout rate. Through the analysis of big data contained in MOOC real data set, it can help teachers master students’ learning rules and learning behaviors in time, which is of great significance to improve the level of education and the dropout rate of MOOC. First, it is helpful for teachers to adjust the course content in time according to the learning status. Second, teachers can carry out benign interventions on students’ learning behavior. In this paper, based on the analysis of the real MOOC data of students’ learning behaviors, we find that different groups of students have the same learning behavior at the same learning period, and students’ learning situations can be divided into three stages. Then ARIMA model is used to predict the learning behaviors of students, to further analyze the factors causing the difference in prediction accuracy, we propose an error analysis method based on MSE. Through the experimental observation, the method proposed in this paper can effectively find and predict the characteristics of students’ learning behavior, to give suggestions for improving the quality of teaching.

Deep Learning (DL) models have tested to be very powerful in solving many hard problems. Especially, those are related to computer vision, text, speech, and classification. However, the blueprint of such models requires large space and elaboration that needs to be examined. Convolutional Neural Network (CNN) is the most popular neural network that can extract the features automatically as compared to conventional machine learning algorithms (CMLA). Our aim in this paper is to lessen the human attempt required to layout architectures by the use of a gadget architecture development process that allows the exploration of huge design space by automating sure version construction, alternative generation, and assessment. The main operations in CNN are Convolution, Pooling, Flattening, Full Connection between the input and output layer. The dataset taken as CIFAR 10 having 60,000 color images of 10 different classes is considered for the study where, various classes represent the images of cars, trucks, frogs, horses, trucks, cats, cars, airplanes, ships, and deer. It is expected that the performance of the CNN model can be further improved by using the deeper network architecture, or by an increasing number of epochs or data augmentation. In this paper, an attempt has been made to explain simple and deeper CNN models on the CIFAR 10 dataset and the comparison has been carried out to check the accuracy achieved from both the models.

Heart disease is the most dangerous disease among all the non-communicable diseases. Annually 17900 thousand of peoples die due to heart problems. Cardiovascular disease (CVD) is the general term used for most of the heart diseases. There are two types of methods for diagnosing a CVD: (i) Invasive Methods (ii) Non-Invasive Methods. Coronary angiography is an invasive method for diagnosing a CVD which is a costly, painful and complicated process. A variety of Non-Invasive (NI) methods are available for diagnosing a CVD. NI methods generate a lot of data which is mainly of 3 kinds :(i) data based on clinical parameters, lab tests and symptoms (ii)data based on raw heart signals (ECG and PCG) (iii)data based on heart images. Majorly, three different machine learning (ML) frameworks may be developed based on the 3 types of data. First framework is simple and main concern is feature selection and classification. Second and third framework is complicated and requires a lot of techniques (preprocessing, segmentation and feature extraction) prior to classification of heart signals and images respectively. In this paper a comprehensive review is presented that summarizes some recent and prevalent machine learning methodologies in all the frameworks. Most of the papers reviewed in this study are from IEEE Explorer, Science Direct, PubMed, Springer, Hindawi, ACM digital library and MDPI libraries. It is found that Support Vector Machines (SVM) and Artificial Neural Networks (ANN) are superseding in most of the studies in all the frameworks. Deep neural network is comparatively newer machine learning methodology which is giving prominent results in classifying heart sound signals and cardiovascular images. The present study will help to automate diagnosis process of heart disease by providing guidelines and avenues to new researchers in domain of machine learning.

English billboards are common in our daily work and life, and how to effectively recognize them is a problem worthy of study. This paper mainly uses Progressive Scale Expansion Network (PSENet) and Convolutional Recurrent Neural Network (CRNN) to conduct text recognition experiments on English billboards. The English billboards are divided into four categories: text distortion, larger background, special text font, and normal format. PSENet is used for text detection, and CRNN is used for text recognition. The identification results of PSENet and CRNN revealed differences among the four groups of English billboards: normal format text recognition is the best, and large background text recognition is the worst, and the recognition effect of text distortion and special text font is less accurate. Finally, the work of this paper is analyzed and summarized, and the future work is prospected.

With the rapid development of transportation, highway bridges as a transportation hub have played an important role and become the link of economic development in the region. In recent decades, my country has built many types of highway bridges. During the construction and operation of these bridges on highways, many inspection activities are required to ensure the quality of construction and operational safety. This article mainly introduces the highway bridge test detection technology and various application methods based on big data. This paper uses big data to detect the dynamics of the highway bridge structure and establish a load test model. The model is solved by the energy-releasing method to strengthen the rigid structure of the highway bridge, evaluate the road condition, and use historical data to revise the model to improve the accuracy of the highway bridge load. The experimental results in this paper show that the highway bridge detection technology based on big data reduces accidents by 23%, and the test and evaluation results under big data show that the performance of highway bridges is excellent and meets the design.

Recently, machine learning methods have a good performance in the field of classification tasks. Summarizing and comparing the performances of different classifiers in the application of their specific classification tasks has a reference significance. In this paper, five classical machine learning classifiers, including GMM, Random Forest, SVM, XGBoost, and Naive Bayes, are compared to show their computing characteristics. The advantages and disadvantages are analysed in this paper. Based on the different datasets, namely different specific classification tasks, the different classifiers perform similarly. However, the SVM-based classifier has the lowest accuracy while processing the text data to apply the text classification task. This result shows that if the classification task is difficult, the accuracy would not be high. This research summarizes the performances of different machine learning methods in the application of specific classification tasks. And this research has a reference significance for the machine learning-based classifiers.

The host computer is an important part of the computer, which is the basic location for users to store data directly. Therefore, we must strengthen the host security, which will protect the security and privacy of user information. However, hackers, Trojans, worms and other network security threats often attack the computer, which has become an important threat to the application of network security. Therefore, we must strengthen the host network security protection, which will protect the host information from illegal network behavior. By ensuring the security of user host information, we can reduce the waste of resources. In the network security attack defense confrontation, both sides have natural asymmetry, which requires us to deal with network security threats. Through a variety of processing measures, we can protect the computer host security. Firstly, this paper analyzes the decision-making framework of network security threats and protection. Then, this paper puts forward the common types of network security threats. Finally, this paper puts forward the current mainstream treatment measures.

In this paper, we measure the performance of supervised learning algorithms on imbalanced class datasets. Supervised learning is considered to be the most advanced and mature from other types of learning in machine learning. On the other hand, imbalanced class data sets refer to an unequal amount of data for each class in the data sets. Many real-world datasets exhibit an imbalanced class distribution. Hence, this paper aims to compare the supervised learning algorithms in classifying the output for imbalanced dataset. This paper also focused on finding the significance of balancing the dataset to the results. This study is conducted by using three different datasets and made a comparison with three supervised learning algorithms chosen which are Logistic Regression, Random Forest and k-Nearest Neighbours. In the experiments, all three datasets used are labelled data with imbalanced class. Due to the imbalanced class for all datasets chosen, a sampling technique with combination of under-sampling and oversampling is implemented to all datasets in data preparation steps. The performance of the algorithms is measured through Classification Accuracy (CA) value. From the experiments, it is proven that Random Forest shown the best results, for both imbalanced and balanced datasets.

The analysis of data useful for further process and find the usefulness for the futher. In education sector, Admission is hottest topic. For the counselling of the admission in reputed institution or foreign university, the past data of the person are taken into the consideration. In this Paper, Admission data taken from the reputed Website (Dataset Weblink) and conduct the study of classification methods like Naïve bayes and its accuracy using Matlab. The performance of different classification methods from MATLAB classifier learner toolbox for admission data. Finally, the paper choosen the suitable classification method for the admission data set.

Skin is an extraordinary human structure. It frequently suffered from many known and unknown disease. Therefore, diagnosis of human skin diseases is the most uncertain and complicated branch of science. It has been observed that most of the cases remain unnoticed because of the lack of better medical infrastructure and facilities. This paper is devoted to solve this challenge. Therefore, this paper effectively proposed (CNN-SVM-MAA) system which combines Convolutional Neural Network with Support Vector Machine classifier to develop a Mobile Android Application. Thus, to evaluate the performance of the proposed system several experiments are conducted on our dataset. This dataset consists around 3000 images which collected from a lot of sources like Beni-Suef University Hospital, Cairo University Hospital and various websites as well to be more accurate and realistic. A comparative study of applying different Feature extraction algorithms with different classifiers was accomplished. The results obtained showed the adequacy of the proposed (CNN-SVM-MAA) system how many skin diseases images have been detected from skin disease dataset. Which lead to detect skin disease and provide the user with the disease name and treatment related prescription with high accuracy.

In recent years, the problem of excessive obesity and poor physique of teenagers has become increasingly prominent. Teenagers lack of life experience and labor training. Their psychological quality is also very poor. The limitation of exam oriented education can never be broken through. In order to solve these problems, the Ministry of education put forward the rectification measures of sports professional education. Let physical education become a compulsory course for children. Put students’ physical and mental health first. Nowadays, the learning method of network teaching is more and more popular. Based on the Internet, the sharing of physical education curriculum teaching resources has become an innovative way for students to learn sports curriculum content. This paper puts forward the current situation of the sharing of physical education teaching resources. This paper also puts forward the design principle of shared resource platform. On this basis, this paper puts forward specific strategies for the construction of shared resources.

According to the analysis of big data, With the rise of online education and teaching in domestic universities, students’ learning behavior data has grown exponentially. How to deal with the massive data, explore the deficiencies in the teaching environment, help the teachers optimize the teaching content and teaching process, and gradually become the focus of education researchers. Based on the analysis of the data characteristics of the teaching platform, this paper proposes a series of visual big data methods for teaching scenarios. Through the programming realization of visual big data method, teachers can intuitively present knowledge for students, timely understand the students’ learning status, and carry out targeted classroom teaching and teaching evaluation. Students can form knowledge framework and interconnection system to promote knowledge expression and Internalization, and complete self-assessment and discover their own learning problems; teaching managers can master the teaching effect of teachers and students’ learning, and adjust the goals of teaching management and decision-making. This paper combines the self-developed online education and teaching - Shidayun classroom, to apply the visual big data method to the actual teaching, which has produced good practical effects and effectively promoted the development of wisdom education.

– Data Communication in customer oriented sector has significant information which is considered as a vital part in enhancing and improving customer base. Loyal customers in business sector are based on buying and selling of products or making use of services provided. Generally, the customers may switch on to variant enterpriser who affords best services with best price hence retaining them is a tedious task unless their communiaction data regarding service is analysed in periodic and to keep track of customer, the enterpriser should know the purchasing pattern and needs of loyal customersThe discovered pattern is also helps to afford proper discounts to the customers at the right time. This research paper experiments customer data in Telecommunication sector to discover useful and interesting patterns. There are so many data mining techniques put forward in revealing the pattern and this paper discusses all the possible mining techniques and evaluates an association rule mining with clustered data to create best rules. For clustering, Hierarchical agglomerative is used and FP-Growth algorithm is used for association rule mining. Both clustered data and association rules results regarding data communication are presented. The research work is implemented in weka tool and assessed with suitable evaluation metrics.

Since the Internet was proposed many years ago, the application of the Internet has become diverse. Now, with the importance of education, the Internet is gradually being used in teaching. Therefore, the purpose of this paper is to explore the application of online and offline teaching mode in embedded system courses under the background of the Internet. In this paper, after consulting the literature on embedded systems and online and offline education models, the data model is constructed by using BP neural network algorithm to process and analyze the relevant data, and the experimental results are drawn. Experimental results show that the Internet-based online and offline teaching model is of some help to the embedded system curriculum.

The establishment of online learning platform through the Internet is a more popular way of learning. However, the existing Internet learning platform is still not intelligent enough, especially in data integration and analysis. Therefore, this paper proposes the application and research of computer BD(Big Data) based on structure in Internet learning. In this paper, the structure of computer BD and Internet learning methods for in-depth research and analysis, the existing problems in the Internet learning, combined with structural computer BDAT(Big Data Analysis Technology), established a new Internet learning BD analysis method. In view of the complex source of Internet learning data and large amount of invalid information, this paper optimizes and improves the traditional BD analysis data preprocessing method according to the characteristics of Internet students’ learning. Through the optimization and improvement of the method in this paper, the data analysis ability of the system is improved. In order to further verify the practical application effect of structural computer BD in Internet learning, this paper takes the freshmen of computer major of grade 2018 in a university as the research object of this experiment and carries out relevant investigation and research. According to the survey, 90.3% of students think that BD analysis can improve their mastery of knowledge in online learning. The analysis shows that the structure-based computer BD plays a good auxiliary role in students’ Internet learning.

The late 2019 outbreak of Coronavirus Disease (COVID-19) had an indelible imprint on the humanity. The world is recovering from the outbreak but there is danger of a second wave of the outbreak. To get rid of the outbreak it is necessary to prevent the viral transmission and it is need of the hour to maintain social distancing and wear masks in public areas. The governments are providing strict guidelines to wear masks in public places. It is not manually feasible to check if people are wearing masks or not. In this paper, process of detecting face masks in public places is automated using Convolutional Neural Networks by performing comparative analysis on Sequential bi-layered CNN, VGG-16 CNN and MobileNetV2 CNN architectures. Among these three architectures MobileNetV2 outperformed with a performance accuracy of 99.2%. The efficient Deep Learning architecture of detecting face masks can be achieved with the help of IoT (Internet of Things) devices and cameras, of those who are not following guidelines in public places. Such a system is very useful in post outbreak period and can be installed in public places such as Railway Stations, Airports, Parks, Schools, colleges, offices etc. to track and ensure wearing of masks by people. The contribution of this paper is not to reel-off the finding from the original paper on Face Mask detection with various architectures rather to provide results on the efficiency of using the MobileNetV2 architecture in comparison with Sequential CNN and VGG-16 architectures for crowd analysis mask detection.

To order and provide services, IoT devices need to communicate and talk to many other devices before these services are provided. However, in complex and heterogeneous systems, certain devices need to trust each other. In this paper, we present an approach that is combined of two lines of defense to help build trust in IoT scenarios and to provide more secure interaction. The paper presents a precisely engineered ontology that serves as the main knowledge base for the definition and semantic registration for the entities within the IoT perimeter to facilitate their automated deployment. We show, in this paper, how trust can be used as a pre-defense step for authorization in flexible and an adaptive manner using semantic web technologies. We develop a proof of concept implementation and give the complexity analysis for our approach.

Executive trading is a kind of investment strategy that analyses and models the financial market with mathematical methods. Machine learning requires computer programs to improve their performance at specific tasks by learning on data sets. Both require the extraction of information from data, so with the success of machine learning in recent years, there has been a strong interest in the combination of executive trading and machine learning methods in both industry and academia. In this paper, based on machine learning is complete, a more complete model of data processing - build - generated forecast - trading strategy - back-test analysis “research framework, and discusses the structures of branching and merging model structure set up, based on the value relevance of the output value and supervision model selection criteria, independently of the back and trading strategy model assessment method and a series of important issues. This paper focuses on the innovative exploration of the application of the loss function, and puts forward several loss functions based on the portfolio point of view, among which the negative cosine loss function shows significant advantages and strong practicability in stability and gain. Corresponding empirical research also reveals the application value and limitation of other loss functions.

Family harmony is an important part of social stability and harmony. Facing the rapid growth of divorce disputes, how to hear the cases quickly and fairly is an urgent matter to be solved. Based on judgment documents from courts located in Southwest of China, this paper studies how to evaluate and predict whether or not a case should be divorced. These documents are characterized by their unbalance and sparsity; besides that, most of variables are bearing with missing value. We propose a feasible method that shows high accuracy on both training and testing datasets. Concisely, oversampling and “clustering” are exploited to data pre-preparation, and recursive feature elimination is applied to deal with variables selection. In this paper we combined Random Forest and XGboost to derive a more feasible and precise model that achieves ninety percent of accuracy.

In this paper more than one approaches are evaluated to optimise machine learning models for diabetes disease diagnosis. The main goal is to sort the medical data computation and choose the most suitable parameters to construct a faster and more perfect model using feature selection. Reducing the number of features to construct a model could direct to more useful machine learning algorithms which helps the doctors to focus on what are the most significant assessment to take into story. Feature selection is one of the process in machine learning which choose a subset of topical features namely variables for construction of models. In this research paper we use three feature selection techniques like Recursive Feature Elimination (RFE), Genetic Algorithm (GA) and Burota Package. After using feature selection at the end we use Decision Tree to predict the diagnosis Diabetes using a dataset named Pima Indian Diabetes Dataset and verify the performance of result model.

The global nature of the Internet of things obviously has the problem of cross domain communication, so the Internet of things also needs a set of perfect resource addressing technology to meet its resource addressing needs and promote the interconnection of the Internet of things. Through the analysis of five key elements of resource addressing, such as resource name, resource address, addressing mechanism, update mechanism and security mechanism, this paper summarizes the resource addressing characteristics of Internet of things, which is not only conducive to the design of resource addressing model of Internet of things that meets the demand of resource addressing, but also the necessary premise to solve the key technology of resource addressing of Internet of things. In order to solve the address conflict and addressing problem between different standards of radio frequency identification (RFID), a hierarchical addressing model based on RFID resource name space is proposed. This paper analyzes the characteristics of RFID resource addressing and the existing RFID addressing technology, finds out the defects of addressing technology based on domain name system, introduces the formal expression method of RFID resource namespace, and designs a hierarchical addressing model.

The research of face modeling and animation technology has a history of 30 years. With the development of virtual reality, video entertainment, videophone, human-computer interaction and other applications, this research field has attracted more and more attention. Based on machine learning algorithm, this paper studies the personalized 3D face modeling and real-time rendering of virtual hosts. The research lines of this paper are the establishment of neutral face model data structure, face feature extraction, improvement of personalized face model matching and real-time rendering in scene roaming. A realistic 3D facial lip animation system based on MPEG-4 standard and driven by voice files is designed and implemented. The system is easy to operate, versatile and efficient, and meets the needs of real-time voice-driven lip animation.

Machine learning has become a great attention to find optimize solutions in different areas and is anticipated to play a vital role in our upcoming technologies. This paper presents a comprehensive review on basic optimization algorithms for micro-strip patch antenna design using machine learning. Classification of machine learning based algorithms: deterministic, stochastic and surrogate model assistant is discussed. Further machine learning models training for optimizing output and for prediction of antenna parameters is presented in this paper. This paper is useful to the readers who work on a particular antenna using the Machine Learning Techniques.

Traffic flow prediction is an important part of intelligent traffic management system. Because there are many irregular data structures in road traffic, in order to improve the accuracy of traffic flow prediction, this paper proposes a combined traffic flow prediction model based on deep learning graph convolution neural network (GCN), long-term memory network (LSTM) and residual network (RESNET). GCN is used to extract the features of topology structure in traffic data, LSTM is used to extract the features of time structure, combined with ResNet to optimize the overall model, reduce the occurrence of gradient disappearance or explosion in network degradation, and finally achieve the prediction of traffic flow. According g to the experimental results, the combined traffic flow prediction model used in this paper is closer to the actual traffic flow occurrence than the traditional convolutional neural network model (CNN), and the accuracy is improved.

Energy conservation and emission reduction is an important part of enterprise management. Energy costs directly affect the economic benefits of enterprises. Energy control of industrial air conditioning has long been the focus of the attention of enterprises. At present, enterprises often adopt fixed pre-cooling and pre-heating time for air conditioning start-up strategy, and there is no unified and scientific standard. By collecting and cleaning the historical data (climate, capacity) of air conditioning operation of enterprises, this paper establishes a scientific model of air conditioning energy consumption and a prediction model of air conditioning starting time. In addition, the theory and technology proposed in this paper are applied in practice to design and implement the air-conditioning energy control system of smart factory, and to predict the starting time of the air-conditioning in the production workshop of enterprises. The results show that the system can help enterprises achieve the goal of intelligent energy management and control.

Brain is the regulatory unit in human body. It controls the functions such as memory, vision, hearing, knowledge, personality, problem solving, etc. The main reason for brain tumour is the abandoned progress of brain cells. Many health organizations have recognized brain tumour as the second foremost dispute that causes a large number of human deaths all around the world. Identification of brain tumour at a premature stage offers a opportunity of effective medical treatment. Use of Magnetic Resonance Imaging images have been recognized as more detailed and more consistent images when compared to Computed Tomography images. There are various techniques to detect brain tumour or neoplasms. The most competent and effective algorithms are discussed in this paper after studying a number of appropriate research papers. Pre-processing brain images, segmenting them, feature extraction, clustering and detection of the tumour are the methodologies in most researches.

The paper proposes the use of support vector machines to establish a financial early warning system and a method for financial risk monitoring, gives a financial evaluation index system and its quantitative methods, and uses the classification capabilities of support vector machines to establish a financial early warning model. At the same time, the paper explores and studies the combination of enterprise group’s financial risk identification, assessment, response and management information system, and proposes an enterprise group’s financial risk control information system. Through experimental comparison, the model based on the support vector machine has good consistency with the experimental data, which is in line with the expected results.

This article proposes an automatic grammatical correction method for typos and word order errors that may occur in Chinese writing by beginners. The thesis first constructs heuristic rules and expands the corpus by analyzing the characteristics of different grammatical error types in the data set. Secondly, when the paper uses classification methods to detect grammatical errors, it extracts sentence-level binary and ternary part-of-speech combinations, n-gram models based on part-of-speech statistics and other three types of features to construct single classification and ensemble classification models, and then use convolutional neural the network constructs classification models from different angles. Finally, when the paper adopts the method based on sequence labeling for grammatical error detection, it mainly uses dependency syntax tree features, and realizes grammatical error detection by constructing a conditional random field model. This method can automatically detect grammatical errors while also identifying the sentence the location of the error. On this basis, the paper implements a simple Chinese grammatical error automatic detection system, which can provide help for the optimization of questions and answers in the question-and-answer system

Community search is to explore valuable target community structure from a large social network. In real community, every point has a geographic location information, and many edges are uncertain. Such a network is called spatial uncertain network. In order to meet the existing needs, this paper studies the community search in the spatial uncertain network so that the searched communities can be relatively close in the geographical location and the relationship between each other is also very close. Based on the spatial uncertain graph, this paper proposes K-R-τ community, and proposes a linear algorithm and a two-dimensional algorithm for community search in the spatial uncertain network. A large number of experiments are carried out on several real datasets, and the results show that the algorithm proposed in this paper is effective and accurate.

Transfer learning from a pre-trained and fine-tuning methodology has been utilized for the image classification. In this paper, we classify images of cats and dogs. It is much faster and easier than training from scratch using a pre-trained network with transfer learning. This saved network was trained previously on a huge dataset and known as a pre-trained model. The pre-trained model can be used in two ways, either it is used in the same way or it utilizes transfer learning approach to adapt this pre-trained network to a specific goal. Since the idea for image classification using transfer learning is that it is trained on the general and large dataset and further this model will be efficiently worked on the visual perception task of the real world. In fine-tuning, instead of tuning with the weights of generic feature map of based model, it is tuned accordingly to the dataset. In this paper, we take benefit of these learned feature maps deprived of having to start from scratch by utilizing a deep learning technique on a huge dataset and analyze the various methodology. The experimental results show that fine-tuning methodology outperformed over the pre-trained with feature extraction on accuracy.

The frequent cross-ship communication has raised the complexity of ship network, therefore raised the requirements of monitoring system on the complicated local network on board. The traditional monitoring system is expensive and can hardly monitor cross-OS local network. Aimed at the specifications of ship network, this paper puts forward using embedded web server to design monitoring system. Through designing monitoring layer with LibPcap, designing application layer with CGI and designing user layer with JavaScript and REST, this paper finish the complete monitoring system and high improves the performance on latency and threat identification success rate.

The backpropagation algorithm has many training and activation functions that can be used to influence or maximize prediction results, all of which have their respective advantages and disadvantages. The purpose of this paper is to analyze one of the training functions of the backpropagation algorithm which can be used as a reference for use in data prediction problems in the form of models and best performance. The training function is the Bayesian Regularization method. This method is able to train the network by optimizing the Levenberg-Marquardt by updating the bias and weights. The research dataset used to analyze the data in this paper is Formal Education Participation in Indonesia 2015-2020 which consists of the School Participation Rate, the Gross Enrollment Rate, and the Pure Enrollment Rate. The 2015-2016 dataset is used as training data with a 2017 target, while the 2018-2019 dataset is the test data with a 2020 target. The models used are 2-10-1, 2-15-1, and 2-20-1. Based on the analysis and calculation process, the results of the 2-15-1 model are the best with an epoch of 217 iterations and an MSE of 0.00002945, this is because the epoch is not too large and has the smallest MSE compared to the other 2 models.

Knowledge graph has shown great value in search engine, natural language Q&A, recommendation system and other application scenarios in recent years. The basic elements of a knowledge graph are entities and relations therein, so how to automatically extract entities and relations from natural language texts becomes a key issue in knowledge graph construction. In this paper, we propose an unsupervised method to extract space domain entities and relations with the goal of building a space knowledge graph. Firstly, a neural network model is used to extract implicit semantic features of domain words represented by dense vectors from original space domain corpus, and then new entities are discovered by clustering in vector space through a small number of labeled data. By concatenating space domain-specific word vectors and general domain word vectors, universal vector representations of entities are obtained, which include general features and domain features as well. On this basis, semantic vectors of relations between entities are calculated, and more new entity relations can be extracted from the corpus by using semantic vectors of relations. Compared with supervised method, the entity and relation extraction method proposed in this paper only needs a small amount of labeled data, thus is quite suitable for the construction of knowledge graph in space domain where labeled data is rather rare and expensive.

The invasion of exotic species should be paid attention to because of its severe consequences, and governments have constructed monitoring and tracking system to collect people’s sightings of species. For example, Asian giant hornet, an exotic species for North American countries was first discovered in September, 2019. Interpretations based on received reports are of essential to master the current situation of species and to predict its spread. However, there is a considerable number of mistake reports, it is necessary to select the most likely positive reports out of mountains of reports to be detailly investigate first. In this paper, two main problems are focused: interpretation of historical reports and strategy of prioritization. First of all, this paper preprocesses the image data based on Gaussian filtering and convolution neural network to make it as balanced as possible. In addition, the image consists of a wide range of array data. Proper treatment can improve the efficiency of analysis. Then, construct the GPSM model (Gray Predicting Spread Model) based on gray prediction to measure the spread of the pest. From the historical positive sighting location provided in the dataset, the GPSM model is able to output the predicted next location following the previous principle. If the distance between predicted location and current outbreak area are significant, a spread can be concluded. Tests for the GPSM model shows the model has a very high level of precision

In order to adapt to the development of economic and social aging, solve the practical problem of incomplete medical security system in China, and provide higher quality medical services to more elderly people. In this paper, a comprehensive and effective survey and data collection on the status quo of medical services for the elderly were conducted by means of questionnaire survey, and an evaluation index system affecting the satisfaction with medical services for the elderly was established by using SERVQUAL model. On this basis, this paper will survey the one dimensional data into two dimensional data, taking advantage of computer vision in depth study using convolution neural network model to generate satisfaction evaluation system for the weight of each index, and then using the trained network model as an effective assessment tools, effective training sample sets of the sample satisfaction accurately forecast, forecast results to improve the elderly health work provides some feasible suggestions.

RESTful is an emergent architectural style to develop web services/applications due to its lightweight nature. Though more and more organizations are adopting RESTful services, when it comes to choosing the right platform to implement one, teams usually gravitate towards reusing the existing skillset and prefer the platform that they currently support instead of looking for a better performing alternative. Considering that internal architecture and implementation of each platform can drastically affect its efficiency and performance, the objective of this research paper is to assess and compare the performance of RESTful web applications implemented in Spring Boot Java and MS.NET Core. An identical business use case to provide the basic four Create, Read, Update, and Delete (CRUD) operations was developed in both implementations. Apache JMeter 5.2.1, an automated performance testing tool is used in this experiment to create a swamp of concurrent virtual users to load both the applications at regular intervals. The comprehensive experimental results are presented at the end of the research paper.