Handwriting Identifies basic graph-like problems and has a high real-world value in areas such as cloud accounting, finance, and postal administration.Due to the unrestricted problem of handwritten numbers when writing, it is relatively difficult to achieve rapid and effective recognition.With the emergence of deep learning-related algorithms and the rapid development of computer hardware technology, image classification methods based on Convolutional Neural Network (CNN) have gradually become a research hotspot.Because the convolutional network has a strong letter numbering ability and network generalization ability, the recognition rate can often exceed the traditional graph classing method.Therefore, the study of hand-written word recognition should be implemented using CNN through the network.Handwriting Word Recognition is the key technique for self-identification.Therefore, summarizing and analyzing the existing handwritten digit recognition algorithms, two handwritten digit recognition algorithms based on Convolutional Neural Network (CNN) are proposed.To improve the recognition performance of the CNN model, this article proposes a handwriting recognition algorithm based on the change to CNN.To extract the image feature information more fully, this paper proposes a handwriting recognition algorithm based on feature fusion and SVM.First, using the modified CNN model and the Gabor filter that introduces curvature systems, extract the CNN and Gabor characteristics of the character image; Second, the characteristics of its progress are fused to obtain more effective new features; Finally, the fusion feature is entered into the SVM classifier into the line number of words to recognize.The results of the experiment show that the algorithm can effectively improve the recognition effect of handwritten
In this paper, we propose an optimization framework for a robust deep learning algorithm using the influences of noisy recurring on artificial neural networks.Influences between nodes in the neural network remain very steady in the convergence towards a superior node even with several types of noises or rouges.Several characteristicss of noisy data sources have been used to optimize the observations in a group of neural networks during their learning process.While the standard network learns to emulate those around, it does not distinguish between professional and nonprofessional exemplars.A Collective system can accomplish and address such difficult tasks in both static and dynamic environments without using some external controls or central coordination.We will show how the algorithm approximates gradient descent of the expected solutions produced by the nodes in the space of pheromone trails.Positive feedback helps individual nodes to recognize and hone their skills, and covering their solution optimally and rapidly.Our experiment results showed how longrun disruption in the learning algorithm can successfully move towards the process that accomplishes favorable outcomes.Our results are comparable to and better than those proposed by other models considered significant, e.g., "large step Markov chain" and other local search heuristic algorithms.
One of the most significant types of Mobile Cloud Networking (MCN) is Cloud-based Wireless Multimedia Social Networks (CWMSNs). We believe that microeconomics theory is a good candidate to model the bandwidth sharing operations in CWMSNs.We model the interactions of mobile users in terms of the barter exchange economy.In our modeling, bandwidth is chosen as the exchangeable commodity and mobile users and desktop users act as players.From a microeconomics point of view, the allocated bandwidth subject to each service plays the role of "endowment" (budget) for players.With this endowment and leveraging the concept of barter exchange, mobile users can interact with each other to gain more quality of service (QoS) in the future.We prove that by applying the exchange economy, users' social welfare could reach to global maximum, known as Pareto efficiency.To the best of our knowledge, the idea of a barter exchange economy has never been employed in any study on cloud computing.Simulation results, obtained through the CloudSim framework, established the robustness of our modeling in terms of significant metrics such as social welfare, number of blocked users, satisfaction level, and Pareto efficiency.
Environmental food and nutritional protection primarily depend on pollination from bees.Historically, beekeeping has been performed in different locations as part of the local food community.Beekeeping is increasing rapidly these days due to the high demand for honey and farmers are taking various forms of beekeeping methods to achieve high yield.Honey production also depends on different types of environmental factors.The main principle of this study is to show the analysis results of various types of environmental factors for three different bee farms by the linear regression model to figure out the best farm among all three farms.To improve the production of honey, farmers have to consider different types of environmental factors and this is the elevated time to support farmers by technology.This study analyzed different types of environmental factors like farm outside temperature, farm inside temperature, farm humidity for three different smart bee farms by using a linear regression model to know about their environmental conditions.The performance of prediction models is measured by R 2 error, Root Mean Squared Error (RMSE), Standard Error values (SE), and Mean Absolute Error (MAE).Based on the outcome, it is observed that the best results giving farm is farm 3 that has been able to give R 2 value 0.95,0.95,and 0.72 for the farm outside temperature, inside temperature, and farm humidity.
This study presents an improved integral value ranked Fuzzy Analytic Hierarchy Process (FAHP) and Geographic Information System (GIS) based Multi-Criteria Decision Making (MCDM) technique to help decision-makers/farmers evaluate and map suitable lands for optimum cassava production.Selected input/ suitability factors chosen from literature and experts' opinion were: pH, organic carbon, cation exchange capacity, slope, aspect, elevation, temperature, relative humidity, rain, distance from river and road.The improved integral value ranked FAHP method was used in prioritizing and assigning weights to each causative factor in the MCDM process due to its effectiveness, consistency, and ease of implementation.Land suitability maps were created using GIS techniques based on the aggregation of the various input factors and their derived weights.The outcome of the aggregation was reclassified into four classes using the standard deviation classification method (this method shows how much a feature deviates from the mean).Results obtained showed that 40% of the total area was highly suitable (S1), 36% was moderate suitability (S2), 20% was marginally suitable (S3) and 4% was not suitable (N).Results also showed that pH and organic content of the soil were the major determinants of soil suitability for cassava cultivation in the study area.This study showed the effectiveness of the proposed approach in assessing and mapping suitable areas for optimum cassava production within the study area.
Text classification is an important problem for spam filtering, sentiment analysis, news filtering, document organizations, document retrieval and many more.The complexity of text classification increases with some classes and training samples.The main objective of this paper is to improve the accuracy of text classification with long short-term memory with word embedding.Experiments were conducted on seven benchmark datasets namely IMDB, Amazon review full score, Amazon review polarity, Yelp review polarity, AG news topic classification, Yahoo!Answers topic classification, DBpedia ontology classification with the different number of classes and training samples.Different experiments are conducted to evaluate the effect of each parameter on LSTM.Results show that 100 batch size, 50 epochs, Adagrad optimizer, 5 hidden nodes, 100-word vector length, 2 LSTM layers, 0.001 L2 regularizations, 0.001 learning rate give the higher accuracy.The results of LSTM are compared with the literature.For IMDB, Amazon review full score, Yahoo!Answers topic classification dataset the results obtained are better than literature.Results of LSTM for Amazon review polarity, Yelp review polarity, AG news topic classification are close to bestknown results.For the DBpedia ontology classification dataset the accuracy is more than 91% but less than best known.
As one of the most important and costly functions of any business, sales analytics has been the target of many studies for some time now. Knowing and tracking the sales of a business proves useful in all data-driven decisions made from inventory management to shelf layouts in a supermarket. However, forecasting sales relies heavily on data and algorithms strong enough to handle unseen data. Since sales data are in nature time series datasets one of such predictive methods is time series analytics. In this paper, the ARIMA modelling with respect to the seasonality of the data is compared with a machine learning technique, support vector regression. These comparisons are carried out on three different and unrelated datasets and these algorithms’ errors when predicting future sales is compared. The results obtained from our analysis shows poor results in general due to datasets having large numbers of oscillation and outliers, but for comparison purposes these datasets and results are fine. We conclude that support vector regression produces better results in comparison with time series analytics on all datasets used in this paper.
This paper described the recognition of the phonetics related to numerical in Indian regional languages such as Marathi & Hindi by Nearest Neighbour rule.The segmentation is based on the location of the start and endpoints of the speech.The exact speech boundaries can be located and evaluated for linear predictive codes.The Linear Predictive Codes of phonetics related to numerical in Indian regional languages such as Marathi & Hindi forms the codebook.The optimum distance between the test and the codebook linear predictive codes can be determined by the Dynamic Time Warping technique.Depending on the distance, the word is recognized by the Nearest Neighbour rule.The accuracy of 88% is achieved with a high reduction in the memory requirements & good SNR.
Recently, proptech, which is a combination of property and technology, has been attracting attention, but it is necessary to apply proptech centered on urban spaces to areas centered on regions such as smart villages.Therefore, the purpose of this study was to derive common technology demands for smart cities for technology demand in urban areas and smart villages for rural technology demands based on the existing prop-tech concept for the real estate industry.As a method of research, we surveyed technology demand.To present a technology platform that can be reflected to prop technology from demand, it was set as a common technology demand range that encompasses this.This is because smart cities and smart villages are based on physical and technical environments in urban and non-urban areas, which means that the scope of the sharing economy and smart real estate technologies that reflect residential and convenient facilities are mutually reflected.Therefore, the common technology demand was divided into general and specialized types.The general type was categorized into cultural, welfare, and living environment services.The specialized type was divided into experience programs, visitor management, public relations, production distribution, accounting management, and facility management to derive detailed technical demands.
Agriculture is the mainstream to keep pace in the Bangladeshi economy.Plant disease became a threat to food security as it is a very important factor to deteriorate the quality and quantity of harvest.Therefore, it is important to detect the plant diseases early which results in interrupting from falling the massive destruction of harvest.But, an erroneous diagnosis of the disease results in the inappropriate use of pesticides.To enhance the production quality and quantity, a deep learning-based approach is proposed to detect the tomato leaf diseases, and then classify the types of the disease using an image dataset.This proposed approach trained two model architectures: inception V3 and Convolutional Neural Network (CNN).Inception V3 performs well and reaches a success rate of 96.11% to identify whether the specific plant leaf is infected or healthy.The success rate is significant and makes this approach a very useful way or early forewarning tool, and this approach might be an essential system to operate real agriculture fields.As the detection accuracy is recorded as 94.72% for CNN.We confirm that it achieves the experimental results with 94.72% and 96.11% for the detection and classification of infected leaves from the dataset for CNN and Inception V3 respectively.
Internet of Things network is based on the distributed infrastructure as large of number of devices connected to the network makes the network an ultra-dense network.The profound devices are becoming capable of connecting to the other devices operating on different networks nature and different architecture thus giving birth to the heterogenic nature of the networks.In such an environment where incident responders face challenges postured by the event that occurred from IoT device networks becomes difficult to gather, analyze and examine its impending traces.This study proposed a contrivance to fetch and provide the information of the IoT devices connected to a certain network using protocols of application layers and associated open ports to the investigators and incident responders.This will help detect and identify the IoT devices connected to the network that will to a significant certainty aided the work of investigators.For this purpose, a tool will be presented through series of experiments and algorithmic development.The results of the experiment show that the proposed tool effectively identified the IoT devices associated with open ports and also classification of the IoT and non-IoT devices is achieved.
In many countries like India, there are more rural areas as compared to cities.In rural areas, due to the lack of medical facilities, people are not much concerned about their health.Even for a routine check-up, they need to travel a long distance.Pregnancy demands a routine check-up and women in rural areas don't do their regular check-ups at an early stage of pregnancy.A routine check-up can help in reducing the fetal mortality rate and to identify and reduce risks (if any) for the mother and the baby.In this paper, the system allows the interaction of doctors with pregnant women through the mobile application.Some crucial parameters of pregnant women like heartbeat rate, blood pressure, temperature, and fetus movement (kicking) are measured and stored in the cloud.The android application can access this information.Whenever there is any fluctuation that happens from the normal value, an alert message is sent to the doctor's mobile application.Hence, doctors can monitor the health of pregnant women.The purpose of this system is to record the parameters of the pregnant woman and deliver the recorded results to the doctor so that the routine health status of the pregnant woman can be monitored without going to the hospital.
Lung nodule classification has been one of the major problems relevant to the Computer-Aided Diagnosis (CAD) system.Lung cancer for both men and women has been one of the leading causes of cancer-related death.Deep learning models have produced promising performance in recent years, outperforming traditional methods in different fields.Nowadays, scientists have attempted numerous deep learning approaches to enhance the efficiency of CAD systems via Computed Tomography (CT) in lung cancer screening.In this paper, we presented a completely automatic lung CT system for cancer diagnosis named Two-step Deep Network (TsDN) and it contains two parts detection of nodule and classification.First, Improved 3D-Faster R-CNN with U-net-like encoder and decoder is used for detection of nodule and then Multi-scale Multi-crop Convolutional Neural Network (MsMc-CNN) is proposed for the pulmonary nodule classification.The multi-scale approach uses filters of various sizes to extract nodule features more efficiently from the local regions, and then the multi crop pooling technique involves extracting the important nodule information that cultivates various regions from the convolutional feature map and then adds numerous times for the maximum pooling.The proposed TsDN is trained and evaluated on LIDC-IDRI public dataset and achieved a sensitivity of 0.885 and specificity of 0.922 with AUC of 0.946.
Image mining is an astonishing data mining concept.To understand the data mining concept prior knowledge is more important to image mining.Image mining deals with the extraction of implicit knowledge, image data relationships, or other patterns not explicitly stored in the images.It is the process of analyzing large sets of domain-specific data and subsequently extracting information and knowledge in the form of new relationships, patterns, or clusters for the decision-making process.Tiger becomes a reserved animal.Conservation of tigers has been a challenging task.This work would add a small account to the herculean task of conserving the species.Several scientific researchers have carried out their research on tiger reserve conservation.This research work proposes a method to find the age of the tiger, using color as a parameter.Color pixel-based image classification and clustering techniques have been used to identify the age of the tiger.This research work mainly focuses on RGB color spaces, which are implemented on real-time tiger images.The objective of the research work is to be done on assessing the age of the tiger using the color pixel-based image classification and clustering is the main task of the research work and to optimize the image filtering and enhancement methods that are used to remove the noise and to improve the quality of pixels or images and assessing the processing Time, Retrieval Time, Accuracy and Error Rate by generating the better results is real-time tiger image database.
To improve the Information Technology (I.T.) graduate skill set, students need to be immersed in as realistic a software development environment as possible. In continuing our work on integrating Agile Methodology into the Capstone Program of our Bachelor of Science in I.T. (BSIT) degree program, this paper discusses the student challenges and difficulties during the software development project, and provides recommendations on improving the student overall learning process in such a program. We collected survey data from the whole population of 90 BSITstudents across four academic years about their experience with their client and the Capstone Program itself. Conceptual content analysis was then applied to discover and describe underlying themes. Also, faculty advisers were tasked with writing about their interactions, thoughts, and observations on their respective student group advisees. These showed issues with time management, communication, and competency. Also, groups that excelled exhibited better team coordination and a complete grasp of the Agile methodology. For future implementations, clearer task definition and reducing the skill gaps are necessary for better execution.
This paper proposes a denoising neural network for real-time ray tracing.The ray-tracing method is applied in graphics to increase the reality and in particular, Monte Carlo Rendering is most effective.However, ray tracing that applies Monte Carlo Rendering has a steep rise in the number of calculations with the increase of the number of rays.Therefore, to solve this problem, various methods are being proposed to reduce the number of rays and to decrease the occurring noise.In this paper, an autoencoder-based neural network that can effectively remove noise while using a small number of rays was implemented.An autoencoder that uses a 1×1 convolution in creating the last feature map was proposed to significantly lower the amount of calculation.The proposed structure can handle an 8196 spp ray-tracing image in 20 seconds at 64 spp.
MANET is a truncation for a special versatile appointed system.It is additionally alluded to as a remote impromptu system and it is a continuous self-designing, framework-free system of phones that are associated with no utilizing wires.In MANET engineering, gadgets can proceed separately in the direction of any path and in this way, change their connections to various gadgets now and then.Since MANETs are transportable, they use remote associations with interface with various systems.In this article, a diagram of MANET alongside its use in remote frameworks will be discussed.A short thought in regards to the kinds of MANET designs and their points of interest and inconveniences in remote correspondence systems is depicted in this paper.
About 5% of the world population, 466million people in figure (432million adults and 34 million children), have hearing impairment.Most people with hearing impairment use sign language as a means to communicate not only with people without disability but also with people with disability.However, there are hundreds of different sign languages and most people without a disability do not know how to use sign language at all.Thus, it is difficult for people without disabilities to communicate with people with hearing impairment in daily living.This study aims to help those with hearing impairment for easier communication.To do this, this study designed a sign language translation program and implemented some of the design to recognize the user's motion by using motion recognition sensors and translating the recognized motions after finding the right words for the motions followed by displaying the words on the screen.
Information on the web is extremely growing in current years with a quicker velocity.This enormous or capacious data has driven intricate troubles for information recovery and information organization.As the data resides in a network with numerous forms, knowledge management on the web is a challenging task.Here the novel 'Semantic Web' concept may be used for understanding the web contents by the machine to offer intelligent services efficiently with a meaningful knowledge representation.The information recovery in the conventional web source is centered on 'page positioning' strategies, though in the semantic web the information recovery forms depend on the 'idea based learning'.The proposed work is gone for the improvement of another system for programmed age of cosmology and RDF to some continuous system information, removed from numerous storehouses by following their URLs and Text Documents.The enhanced altered ordering method is connected for the cosmology age and turtle documentation is utilized for RDF documentation.A program is composed for approving the extricated information from different archives by expelling undesirable information and considering just the report segment of the website page.
This paper presents a comprehensive comparative review of existing floating-point multiplier systems.The study focuses on single, double, quadruple and multi-precision floatingpoint multiplier architectures and seeks to identify engineering techniques involved in their development.A comparison of the performance of these systems in terms of metrics such as path delay, hardware utilization and even power consumption in some cases are carried out.Weaknesses in the systems reviewed along with possible gaps in the area of research are identified.This paper also serves to identify several recommendations and considerations for the development of a multi-precision floating-point multiplier system capable of treating the weaknesses of multiplier systems identified.
