Deep learning is now being used in the medical field to detect ailments such as cancer, diabetes, kidney disease, etc. Kidney disease has become a major public health concern around the globe. If CKD is not diagnosed at an early stage, it can lead to loss of kidney function and which require costly treatments like dialysis and a kidney transplant. Using CNN, an automatic model has been developed to predict chronic renal disease. The major goal of this study is to predict CKD and non-CKD patients. Convolutional neural networks are used to classify chronic renal disease in this proposed system, and the batch prediction method is evaluated for CKD prediction. The precision with which renal disease can be predicted is 95%, and the accuracy for the classification of CKD ultrasound images using CNN is 80%.
The overall development of the internet allows internet attacks to occur, which can cause damage to a system. Threats and attacks on internet networks are more vulnerable to the surface because the internet is fully open to users. We need data protection from threats and attacks to maintain confidentiality, availability, and system information. Threats or disturbances can be referred to as anomalies. Anomaly detection is needed to prevent changes in traffic flow. Anomaly detection is one of three techniques of the Intrusion Detection System (IDS). Network characteristics tracked by network anomaly detection programs at scale include packets, bandwidth, bytes, traffic volume, and the used protocols. Suspicious events are recorded in Interface, IP Group, Transmission Control Protocol (TCP), User Data Protocol (UDP), and Internet Control Message Protocol (ICMP) reports. Therefore, this research was carried out to detect anomalies using the Machine Learning algorithm: Decision Tree. This study analyzed 4998 records with 34 attributes, with one attribute as a class. Using the decision tree method, the highest accuracy results are 99.95%.
A triple band flexible microstrip (FM) antenna is presented for Wireless Local Area Network (WLAN), Worldwide Interoperability for Microwave Access (WiMAX), and Ultra Wideband (UWB) applications. The basic C-slot antenna has been modified with an extra I-slot added in the middle. Felt substrate, with permittivity 1.22 and height 4 mm, is taken as substrate which adds flexibility to the design and makes it more robust. The proposed FM antenna resonate at three frequencies, viz. 2.5/5.4/7.1 GHz with 10-dB impedance bandwidths of 95/367/568 MHz and voltage standing wave ratios (VSWR) of 1.44/1.9/1.5 respectively. It also provides a significant overall directive gain of 8.75 dB. With fair certainty, it can be declared that the suggested antenna is suitable for usage in all three of the aforementioned applications due to its straightforward construction, small size, flexible nature, and acquired results. High Frequency Structure Simulator (HFSS) has been used for the design and analysis.
“All-media“ is the result of the deep integration of various media under the support of information and network technology. With the advent of the era of all-media, all-effective media, all-member media, holographic media and full-scale media are gradually formed, and life is more intelligent and convenient. With technology empowerment, reading content is getting richer and richer, and ‘‘ Audio Reading’’ has developed as both a carrier and platform for content, and a medium to connect content dissemination and recipients. This paper develops an interactive platform for Audio Reading based on the progressive framework Vue, which simplifies the event-driven user interface; based on the modular idea, it designs the specific module composition of the system interface and interaction, and builds a reader based on FBReader to meet the needs of audio reading in the era of allmedia.
Visual Tracking is an approach through which a moving object can be tracked at distinct pixel points and coordinates. It is a part of machine vision technology where a machine can recognize the objects as per the patterns and track in the whole frame accordingly. There are so many challenges to recognize the object and track without any overflow. Conventional model is not able to detect the object in the entire frame with high level of accuracy. Machine learning is in trending phase where a dataset can be trained to obtain the better result in tracking the object with more preciseness. There are several researches have been done in this field but certain flaws present in the existing systems for few challenges. Each and every system has its own flaws where accuracy may degrade and system encounters more overlap. There are various benchmarks are available to test the system and obtain the result accordingly to pertain the precision rate. The intention of this paper is to find out the flaws present in the existing algorithms and their results as per the benchmark selected. There are various challenges present in the benchmarks which need to be revised with better accuracy.
In recent years, with the continuous development of digitalization, all kinds of data on the Internet have increased rapidly, and knowledge graphs have emerged. Knowledge graphs have become one of the important means for us to manage and utilize knowledge. Knowledge reasoning is part of building a knowledge graph. There are many different methods of knowledge reasoning, which are mainly divided into traditional knowledge reasoning and knowledge reasoning over knowledge graph. The knowledge reasoning manner based on neural network has stronger thinking ability and generalization ability. The inference effect is better. The utilization rate of the relation, entity, attribute and text information in the knowledge base is higher. In this paper, the basic ideas of knowledge graph are introduced in detail. The basic principle of knowledge reasoning is expounded. In addition, from the three dimensions of semantics, structure and auxiliary storage. Three inference methods are introduced. Furthermore, the problems of neural network are summarized, and the challenges of knowledge reasoning are described. Finally, The development prospect of neural network and knowledge reasoning are prospected.
The development of brain tumor cells causes the intracranial pressure inside the skull to rise, which ultimately threatens the minuscular life cycle. So, medication is the only way to increase life expectancy. It may cause death if not treated in the beginning stages. In this research paper, we aim to increase the effectiveness of magnetic resonance imaging (MRI) devices to identify cancerous brain tumor cells. We classify brain tumor cancer cells using five pre-trained convolutional neural network architecture models: ResNetl52, DenseNet201, VGG19, MobileNetV2, and InceptionV3. The ResNetl52 model was found to have the highest accuracy rate after the work was analyzed, with an accuracy value of 98.52%, which is almost 99%. In addition, the VGG19 model also showed a good accuracy rate of 98%. Other models have accuracy values of InceptionV381.30%, DenseNet20183.30%, and MobileNetV286.73%, respectively. Lower ranking means models are unable to identify brain tumor cancer cells compared to other models in the same magnetic resonance image. Based on the result, it has been observed that ResNetl52 is suitable for brain tumor cancer cell detection. The early diagnosis of malignancies before they develop physical adverse effects is benefited from this accuracy.
In the process of agricultural production, a large number of data are produced, these data are managed and classified reasonably and effectively, and their internal laws are found, which provides scientific basis and reference for the development of agricultural modernization, to help agricultural production, management, management, to achieve the organic integration of big data and agriculture [1]. In this paper, the principles of Bayes, K-nearest neighbor algorithm, decision tree and other common classification algorithms are discussed, and the application of big data classification in agriculture is studied.
Lately, sensible Image processing using deep neural networks has become a fervently discussed issue in machine learning and computer vision. Image can be made at the pixel level by learning from a gigantic variety of pictures. Learning to make splendid movement pictures from highdifference draws is not only a captivating investigation issue yet also a reasonable application in innovative delight. In this research, we research the sketch-to-picture mix issue by using prohibitive generative poorly arranged networks. The model can normally deliver reasonable shadings for a sketch. The new model is not only prepared for painting hand-drawn sketches with real tones, yet also allows customers to exhibit supported tones. Test results on two sketch datasets show that the autopainter performs better contrasted with existing picture-topicture methodologies. With creating interest in the development of film, the interest in building a computerized structure to change over the authentic video into action is higher than at some other time. The edge-by-diagram modification of the action age measure is costly and dreary. To help with moving quickly and with no issue in a robotized collaboration we proposed a generative model that changes over genuine pictures into contrasting energy pictures without losing critical nuances of the source picture. We used an assortment of the generative hostile association as a fundamental plan with the custom incident ability to ensure the substance of the source picture, which changed over to an exuberance image.
In today’s increasingly developed Internet, ecommerce has become an extremely popular form, especially in my country, due to the rapid economic development, the development form and content of e-commerce are becoming more and more mature. And the credit evaluation model is a crucial link in the whole process of e-commerce. In order to solve the problems existing in the existing e-commerce website credit evaluation model, the system adopts the MVC architecture and the fuzzy comprehensive credit evaluation model, and introduces each functional module and model overview of the system in detail. The process of analysis and system evaluation shows that the system has excellent applicable value. The system test shows that the buyer’s evaluation rate can affect the credit of the e-commerce platform after the transaction is successful, and the transaction amount has little effect on the credit of the e-commerce platform, so it proves that the credit evaluation system has certain feasibility.
This research design presents microstrip-patchantenna(MPA) for many shapes of antenna, in this paper with and without cuts ofHexagonal Split Ring Resonator at24.5 GHz for 5G applications. The substrate, FR4, has a relative permittivity of 4.4 and a height of 1.6 mm, respectively used in Microstrip feed. The designs are simulated using Ansys HFSS Software, and the antenna parameters, such as gain, SII& VSWR, band width, and directivity, have been researched. According to the results, when the number of rings increases, resonance is attained at the high of gain with a bandwidth ranging from 223. 2MHz to 1021. 5MHz at the very least.
The ability of enterprises to create and use new technologies is critical to their success. This paper aims to study orientation of digital transformation (DT) of power grid enterprises by digital innovation technology. The transformation focuses on technology and becomes a mere formality, which leads to serious homogenization. Without accurately grasping the Internet thinking, we still use the traditional thinking of running newspapers to deal with the digital operation of power grid enterprises; Wrong positioning of its own functions and market, failing to accurately assess the advantages of power grid enterprises and market changes under the Internet environment. In view of the above problems, this paper puts forward some suggestions: the digital operation of power grid enterprises should reposition their functions and markets with the flow thinking and the thinking of serving users; Reorganize the structure of power grid enterprise groups according to the standards of Internet enterprises, and participate in market competition as market players; The digital operation of power grid enterprises must create unique value points, stick to users with personalized products, and be at the core of the industrial chain.
The primary crop used to produce sugar and ethanol in the globe is sugarcane. One issue in the sugar sector is if sugarcane illnesses are not treated and diagnosed early, they can eradicate growing crops that are infected with the disease, costing small-scale farmers money. The continuously growing categories of illnesses and farmers’ limited capacity for disease diagnosis and identification served as the impetus for this investigation. Deep learning techniques combined with computer vision and machine learning provide a solution to this problem. In this work, a deep learning model with a 97% accuracy rate was constructed and evaluated using 1200 photos of healthy and sick sugarcane leaves. By distinguishing between classes of healthy and ailing or diseased sugarcane leaves in photographs of sugarcane, the trained model successfully performed its objective. This study recommends employing a deep learning system to help farmers recognize and classify sugarcane illnesses.
Diabetic retinopathy is a leading cause of blindness in adults, particularly in those with poorly controlled diabetes. It affects around 93 million people globally and is classified into two main categories: nonproliferative and proliferative. Early detection and treatment are crucial to prevent vision loss, but the current gold standard for diagnosis, fundus photography, has limitations such as requiring specialized equipment and trained personnel, and patient cooperation. To address these issues, this research develops a web application that uses CNN algorithms to effectively diagnose the severity of diabetic retinopathy. The web application, FundusNet, was developed through a multi-phase process using transfer learning and a dataset of fundus images. The proposed hybrid model (based on EfficientNETB7 and VGG-16 using Laplacian Sharpening) was found to have an accuracy of 91.8% and a F1 score of 91.5. FundusNet is free, easy to use, and does not require internet access, making it a convenient and scalable solution for diabetic retinopathy detection, especially in resource-limited settings.
In view of the individual differences in student groups, the traditional teaching model cannot meet the needs of students’ all-round development. Stratified teaching focuses on individual differences and learning needs of students, and makes targeted teaching plans and objectives, which is conducive to the improvement of the overall teaching effect. By establishing the difference model in stratified teaching and taking the computer network course as an example, the hierarchical teaching model based on the fast-clustering algorithm (K-means) is designed to achieve the hierarchical learning of students. The simulation experiment shows that the hierarchical teaching model based on the fast-clustering algorithm is more scientific and reasonable than the hierarchical method of expert classification algorithm and student total score ranking, and can provide reference for other disciplines.
In this paper, sidelobe suppression problem in the antenna radiation pattern is taken into consideration. Dealing with this optimization problem, an algorithm that is called Harmony Search Algorithm (HSA) is proposed and compared with existing and used algorithms in the literature, Particle Search Algorithm (PSO) and Artificial Bee Colony (ABC). A statistical comparison is carried out to examine the differences in performance. The results show that HSA can be used to suppress sidelobes with considerably short processing time.
Liver is one among the crucial organs of a human body. If the liver’s normal functioning is compromised it leads to liver disease. The way people live today has completely changed, and many of them consume alcohol, unhealthy foods, energy drinks, self-medication, and other things like environmental pollution. And some of them continue to engage in these activities daily. Human livers will be severely damaged by this, and much suffering will result. Particularly heavy drinking makes many people more susceptible to liver failure and makes living a perilous life for them. The liver is the most important organ in humans, performing a wide range of tasks such as bile generation, bile and bilirubin excretion, protein and carbohydrate metabolism, enzyme activation, glycogen storage, vitamin and mineral absorption, plasma protein synthesis, and the manufacture of clotting factors. As a result, it is increasingly crucial to detect diseases at an early stage, as doing so enables the use of early, low-dose treatments to help prevent disease. However, because the signs of liver illness at the beginning are so minimal, it is too difficult to detect it. Machine learning holds great promise for automated disease diagnosis. The discovery and treatment of illnesses have undergone a complete transformation this is because of machine learning development in medicine. Machine learning model application in healthcare research includes knowledge extraction and decision support.
An event-driven spectrum-aware routing algorithm based on the Hungarian algorithm (ESRH) is proposed to improve the spectrum utilization and the energy efficiency of cognitive radio sensor networks(CRSN). The method performs clustering in a distributed and self-organized manner and selects the node with the largest weight as the cluster head (CH). The Hungarian algorithm is used to assign the channels with smaller occupancy probability of primary user (PU), longer idle time and higher throughput to sensor nodes (SUs), which reduces the channel competition from SUs to PUs and improves the data transmission success rate. The gateway nodes and packet forwarding nodes are used for relay communication between clusters. Simulation results show that ESRH outperforms ESAC and ERP algorithms in terms of transmission performance, routing stability and network lifetime in a multi-round event-driven CRSN.
Since the industrial revolution, the world has been seeking new advancements in technologies. The global movement in the advancement of technology since 2008 brought in the advent of Information Technology and Web development globally. Today, we aim to make everything more and more advanced, convenient, and at our fingertips. Every technology and software today focused on decreasing human effort. The development of the web browser and the Internet has played a vital role in the all-around IT technology development. In this vast system of services, a browser extension plays a small role in managing the browser. A browser extension is like a mini-website with a graphical user interface which allows the user to change the functionalities of a browser. The base idea is to change or modify the browser’s functionality, providing a better user experience. This screen time tracker and task management browser extension enables the user to get the status of their screen time and access a platform that manages all their tasks. It helps boost the productivity of an individual. The main objective of Free Focused is to ease the social and organizational complications that may occur in one’s work life.
