Asian Journal of Research in Computer Science
Not a member yet
792 research outputs found
Sort by
Improved the Least Square Regression Line Method to Develop A Predict Method for Discriminate the Trend of Stock Price in Future
No full textThe paper described the least square regression line method has been improved as a novel method for tendency discrimination on future stock price. A new method is established, which is obtained from one dataset of known strain variables. The result is been calculated from 10 strain variables consist of one dataset through a few unique managing approaches to calculate out four different tendencies, and encode them. Those codes are added into the least square regression line method by the application software of MATLAB to develop a diagnostic method, which can predict the tendency by the text and graphics. The new method predicts trends more clearly and easily than the least square regression line method. In this paper, firstly establish any of the four standard graphics that can be generated from known data. Finally, it is verified by historical data and the graphics are compared. The result is consistent with both text recognition and graphics trend. The new novel method of the least square regression line was as accurate and alike as we had expected. So that it is enough to prove that it is not only easy to understand but also easy to operate for discriminating the trend of stock price in future
Image Compression Based on Deep Learning: A Review
No full textImage compression is an essential technology for encoding and improving various forms of images in the digital era. The inventors have extended the principle of deep learning to the different states of neural networks as one of the most exciting machine learning methods to show that it is the most versatile way to analyze, classify, and compress images. Many neural networks are required for image compressions, such as deep neural networks, artificial neural networks, recurrent neural networks, and convolution neural networks. Therefore, this review paper discussed how to apply the rule of deep learning to various neural networks to obtain better compression in the image with high accuracy and minimize loss and superior visibility of the image. Therefore, deep learning and its application to different types of images in a justified manner with distinct analysis to obtain these things need deep learning
BPS: Blockchain Based Decentralized Secure and Versatile Light Payment System
No full textIn the presentage, online paymentsystem is a very simple practice.But manypeople use this system to manipulate people’s money. Many are trying for finding a variety of solutions. Butthereis no way to stopthatcrime.Blockchain’syoke is a blessing.Usingblockchainisaveryeasywaytocompletea paymentwithoutmakingany mistakes.Hackerwill never find a way to do theirworkin this kindof system. OurSystem is full workedwith Blockchain.Basically, we choose blockchainas our projectbecauseit is the most secureway to do a transaction in everyonlinesystem.Thecentralbusinessmodelis basedon a database management system. Once accomplishedthe security of the transaction can no longer be guaranteed. On the otherhand, itis really expensivetoresolvepossiblefraudtransactionsby a middle man.Aiming at solving issues concerningsecurityand worthlessness,there is a proposalof a model which is completely madeofblockchainsystem.InOursystemtherearemany blocksof information of eachandeverytransaction. Wehave proposedan algorithm.The algorithm will make consumersable to transact through cryptocurrency in blockchainnetworks.It is totally different from the fiat system where consumers will be able to transact without the help of thirdpartiesand vendors can also be relievedwith theirtransaction. Thistypeof transaction will be very comfortablefor both consumersand vendors. Consumers along with vendorscan see the whole transaction date, time and everythingthatthey dealtwith when the transaction was held
Emerging Approach for Detection of Financial Frauds Using Machine Learning
No full textThe growth of regularly generated data from many financial activities has significant implications for every corner of financial modelling. This study has investigated the utilization of these continuous growing data by a means of an automated process. The automated process can be developed by using Machine learning based techniques that analyze the data and gain experience from the underlying data. Different important domains of financial fields such as Credit card fraud detection, bankruptcy detection, loan default prediction, investment prediction, marketing and many more can be modelled by implementing machine learning methods. Among several machine learning based techniques, the use of parametric and non-parametric based methods are approached by this research. Two parametric models namely Logistic Regression, Gaussian Naive Bayes models and two non-parametric methods such as Random Forest, Decision Tree are implemented in this paper. All the mentioned models are developed and implemented in the field of Credit card fraud detection, bankruptcy detection, loan default prediction. In each of the aforementioned cases, the comparative study among the classification techniques is drawn and the best model is identified. The performance of each classifier on each considered domain is evaluated by various performance metrics such as accuracy, F1-score and mean squared error. In the credit card fraud detection model the decision tree classifier performs the best with an accuracy of 99.1% and, in the loan default prediction and bankruptcy detection model, the random forest classifier gives the best accuracy of 97% and 96.84% respectively
A Survey of Data Mining Activities in Distributed Systems
No full textDistributed systems, which may be utilized to do computations, are being developed as a result of the fast growth of sharing resources. Data mining, which has a huge range of real applications, provides significant techniques for extracting meaningful and usable information from massive amounts of data. Traditional data mining methods, on the other hand, suppose that the data is gathered centrally, stored in memory, and is static. Managing massive amounts of data and processing them with limited resources is difficult. Large volumes of data, for instance, are swiftly generated and stored in many locations. This becomes increasingly costly to centralize them at a single location. Furthermore, traditional data mining methods typically have several issues and limitations, such as memory restrictions, limited processing ability, and insufficient hard drive space, among others. To overcome the following issues, distributed data mining\u27s have emerged as a beneficial option in several applications According to several authors, this research provides a study of state-of-the-art distributed data mining methods, such as distributed common item-set mining, distributed frequent sequence mining, technical difficulties with distributed systems, distributed clustering, as well as privacy-protection distributed data mining. Furthermore, each work is evaluated and compared to the others
Approach to Kirana Store Product Arrangement Using Machine Learning
No full textMarket Basket Analysis (MBA) is a method for determining the association between entities, and it has often been used to study the association between products in a shopping basket. Trained Computer vision models are able to recognize objects in photos so accurately that it can even outperform humans in some instances. This study shows that combining objective detection techniques with market basket analysis can assist Stores/Kirana in organizing the products effectively. With the use of MBA and Object detection, we formulated recommendations for store arrangements along with putting a recommendation engine on top to help shoppers. After deploying this to local Kirana stores, the Kirana store was able to see an increase of 7% in the sale. The recommendation engine performed better than just the domain knowledge of the kirana store
Discussion and Improvement on the Calculation of the Relationship between the Base Emitter of the Bipolar Junction Transistor and the Temperature by Using MATLAB Software
No full textThe purpose of this article is to explore and improve the effect of Bipolar Junction Transistor amplifier base-emitter on temperature changes. The prelude starts with Shockley\u27s theory and its formula calculation. Research motivation, literature data and book principles cooperate with MATLAB application software skills to develop programs; it is used to calculate the relationship between various component parameters and temperature changes. The artificial calculation steps are too cumbersome and prone to clerical errors. Therefore, the program approach has been developed with report-style calculation results with both text and pictures. The feature of computer program calculation is the ability to compare and analyze the results produced at different temperatures, rapidly and repeatedly. The lack of known Bipolar Junction Transistors is replaced by Field Effect Transistors that are not affected by temperature. If the Bipolar Junction Transistor is used as the basis for the design, temperature changes must be considered to ensure the design quality. The purpose of this article is to introduce that the process of calculation has made the shift from an artificial-based way to a computer-based one
Efficient Data Mining Techniques for Heart Disease Prediction and Comparative Analysis of Classification Algorithms
No full textData mining techniques are used to extract interesting patterns and discover meaningful knowledge from huge amount of data. There has been increasing in usage of data mining techniques on medical data for determining useful trends and patterns that are used in analysis and decision making. About eighty percent of human deaths occurred in low and middle-income countries due to heart diseases. The healthcare industry generates large amount of heart disease data which are not organized. These data make the prediction process more complicated and voluminous. Data mining provides the techniques for fast and accurate transformation of data into useful information for heart diseases prediction. The main objectives of this research is to predict heart diseases more accurately using Naïve Bayes, J48 Decision Tree, Neural Network, Random Forest classification algorithms and compare the performance of classifiers. The research uses raw dataset for performance analysis and the analysis is based on Weka Tool. This research also shows best technique from them which is Random Forest on the basis of accuracy and execution time
Effectiveness of Classification Methods on the Diabetes System
No full textIn today’s world using data mining and classification is considered to be one of the most important techniques, as today’s world is full of data that is generated by various sources. However, extracting useful knowledge out of this data is the real challenge, and this paper conquers this challenge by using machine learning algorithms to use data for classifiers to draw meaningful results. The aim of this research paper is to design a model to detect diabetes in patients with high accuracy. Therefore, this research paper using five different algorithms for different machine learning classification includes, Decision Tree, Support Vector Machine (SVM), Random Forest, Naive Bayes, and K- Nearest Neighbor (K-NN), the purpose of this approach is to predict diabetes at an early stage. Finally, we have compared the performance of these algorithms, concluding that K-NN algorithm is a better accuracy (81.16%), followed by the Naive Bayes algorithm (76.06%)
Extracting a Bounded Region from a Map Using Flood Fill Algorithm
No full textExtracting the needed portion from a bounded region is an important task in image processing. Editing a map and extracting a region from the map is challenging. It is useful in some contexts to have a region in a separate sheet. In this image processing, we have used the Flood Fill algorithm to extract a region from the image map. To achieve that goal, we had worked in our study to separate a bounded region on a map. Usually, a scanned map may contain a lot of useless information. So we have to process the image to remove useless information from the map. We had quantized the image to a binary one. In the second phase, we have applied a gray color to separate the desired position from a map. Our main objective of the study to extract a bounded region from mapping an image that contains useless information and removes it. We have experimented with several maps and it works successfully