درخت حوزه‌های تخصصی

In our research, we provide a statistical syntax parsing method experimented on Kannada texts, which is an official language of Karnataka, India. The dataset is downloaded from TDIL website. Using the Cocke-Younger-Kasami (CYK) parsing technique, we generated Kannada Treebank dataset from 1000 annotated sentences in the first stage. The Treebank generated in this stage contains 1000 syntactically structured sentences and it is used as input to train the syntax parser model in the second stage. We have adopted Probabilistic Context Free Grammar (PCFG) while training the parser model and extracting the Chmosky Normal Form (CNF) grammar from a Treebank dataset. The developed syntax parser model is tested on 150 raw Kannada sentences. It outputs with the most likely parse tree for each sentence and this is verified with golden Treebank. The syntax parser model generated 74.2% precision, 79.4% recall, and 75.3% F1-score respectively. The similar technique may be adopted for other low resource languages.

The Analytic visualization suggests representing knowledge during a visual type that may be charts, graphs, lists, or maps. The COVID 19 detection and analysis of spreading is very important for countries. Database management with respect to virus deep analysis is a critical task to the researcher through conventional algorithms. The RNA, DNA, and biological data are helping to the bio-inspired algorithm but its implementation can be complex by software tools. Therefore, an effective technique is required to cross over the above limitations. So that covid 19 pandemic data analysis is performed through FCNN (Fully conventional Neural Network) pre-training network. The dataset is collected from social media, Kaggle, and GitHub databases. At 1st stage, the auto stack encoding process is applied later same data is processed with FCNN deep learning classifier. In this research work, covid-pandemic affects parameters like infected persons, deaths, active cases, and recovering cases. The FCNN is take care of feature extraction, training, testing, and classification. Finally using a confusion matrix accuracy of 98.34%, sensitivity 97.63%, Recall 98.26%, and F measure 98.83% had been estimated.

Diseases like diabetes are chronic and require long-term management. Inadequate production of insulin results in high blood sugar levels. Such diseases lead to serious health issues such as heart ailments, blood vessel complaints, eye ailments, kidney function disorders, and nerve ailments. Hence, accurate assessment and management of risk factors are crucial for the onset of diabetes. Our proposed approach combines fuzzy logic & machine learning algorithms for diabetes risk prediction. Three machine learning models were trained to classify patients into two categories of diabetes (Type-I and Type-II) based on their clinical dataset collected from Katihar Medical College & Hospital and Suvadhan Lab. The polynomial regression algorithm achieved a score of 0.947, while the support vector regression algorithm with the rbf kernel achieved a score of 0.954, with a linear kernel achieved a score of 0.73. Our proposed approach performed well with respect to the conventional approaches with improved accuracy by identifying the patients at diabetes risk. In future work, we further analyze the relationship between other ignored factors which contribute to diabetes risk.

Numerous face recognition systems employ deep learning techniques to identify individuals in public areas such as shopping malls, airports, and other high-security zones. However, adversarial attacks are susceptible to deep learning-based systems. The adversarial attacks are intentionally generated by the attacker to mislead the systems. These attacks are imperceptible to the human eye. In this paper, we proposed a feature-based masking iterative method (F-MIM) to generate the adversarial images. In this method, we utilize the features of the face to misclassify the models. The proposed approach is based on a black-box attack technique where the attacker does not have the information related to target models. In this black box attack strategy, the face landmark points are modified using the binary masking technique. In the proposed method, we have used the momentum iterative method to increase the transferability of existing attacks. The proposed method is generated using the ArcFace face recognition model that is trained on the Labeled Face in the Wild (LFW) dataset and evaluated the performance of different face recognition models namely ArcFace, MobileFace, MobileNet, CosFace and SphereFace under the dodging and impersonate attack. The F-MIM attack is outperformed in comparison to the existing attacks based on Attack Success Rate evaluation metrics and further improves the transferability.

Hospitals are the most important part of the healthcare system. Statistics show that a significant portion of health budgets are allocated to hospitals. The continuous impact of information technology on hospitals’ performance has led to perfect competition. Accordingly, this study aimed to evaluate the quality indicators of hospital services considering information technology using a hybrid approach of the Kano model, Analytical Hierarchy Process (AHP), and Quality Function Deployment (QFD). In this regard, based on related studies, a total of 18 needs were recognized to evaluate the service quality of a hospital. The statistical population of the study consisted of patients of the hospital and due to the difficulty of access to the patient, a limited sample of 50 patients was selected. After collecting data, the identified needs were classified into three categories called basic, functional, and motivational using the Kano model, and 7 needs were set as basic needs. Then, using the AHP technique, the importance of the basic needs was calculated and considered as the input of the QFD model in the next phase. After providing some solutions based on the literature to meet these 7 needs, solutions were ranked and prioritized using the QFD model. Since the organization had limited resources, the Pareto technique was used to respond to 20% of these strategies and achieve 80% satisfaction. The results of the study showed that the hospitals can achieve 80% satisfaction by implementing the strategies of “holding ethics training courses online” and “creating team spirit and using health information technology in the hospital”, respectively.

The cybersecurity in the modern world has become global, and cyber attacks are becoming more complex and large-scale. In the system of civil and criminal justice, computer forensics helps to ensure the integrity of digital evidence presented in court cases. The purpose of this study is to develop scientifically sound proposals and recommendations for the implementation of tools for forensic research of computer tools and systems in the fight against cybercrime. The relevance of this study is due to the need to implement active ways to protect and combat cybercrime. To achieve the goal of the study, methodological principles and approaches of legal science were used. It is proposed to use computer forensic methods more widely research in the fight against cybercrime.This study identifies the types of computer forensics: forensics database; electronic forensics; malware forensics; criminology of memory; mobile forensics; network forensics. The authors foundlack of a regulatory mechanism to regulate cybersecurity, capture and use of digital evidence and the regulatory framework for international cooperation. To brought need in strengthening international cooperation and in developing appropriate policies and legislative initiatives of security and network and information systems, improvement legislation in the field countering cybercrime.

The rapid emergence and evolution of technology have greatly impacted the way people live their lives. The internet has become a vital part of our daily lives. E-commerce is a type of technology development that enables customers to buy and sell products online. It is a unique form of transaction that connects people from all around the world. Today, many consumers shop for products online and present their products along with their specifications. This is becoming more prevalent. This increases the number of consumers online, which can result in a drop in growth. This is one of the main factors that a company uses to measure its success. Growing business success is revealed with retained customers. Satisfied consumers are the assets for a growing business. The study investigates the factors that influence people's decisions when it comes to buying items online. It shows that the experiences they have while shopping online can affect their decisions. The following statistical tools were applied for this study: percentage analysis, mean score with rank correlation, and t-Test. The results reveal that the factors quality, cost, product variety, uniqueness, and safety payments were highlighted as important indicators of performance and that the companies that do online businesses had to take care of their main goal as per the proposal.

The widespread use of E-commerce websites has drastically increased the need for automatic recommendation systems with machine learning. In recent years, many ML-based recommenders and analysers have been built; however, their scope is limited to using a single filtering technique and processing with clustering-based predictions. This paper aims to provide a systematic year-wise survey and evolution of these existing recommenders and analysers in specific deep learning-based hybrid filtering categories using movie datasets. They are compared to others based on their problem analysis, learning factors, data sets, performance, and limitations. Most contributions are found with collaborative filtering using user or item similarity and deep learning for the IMDB datasets. In this direction, this paper introduces a new and efficient Hybrid Filtering based Recommendation System using Deep Learning (HFRS-DL), which includes multiple layers and stages to provide a better solution for generating recommendations.

The military and political leadership of Ukraine considers the domestic military-industrial complex as an important component of the country's national security and defense strategy and pays special attention to increasing the efficiency of production and scientific and technical activities of defense industry enterprises and organizations. The study represents directions for improving the state regulation for the further development of the military-industrial complex in Ukraine under the conditions of the transition to modern information technologies. Proposals have been made for the formation of the organizational and economic mechanism for state regulation development of the military-industrial complex, aimed at ensuring its innovativeness, stimulating scientific and technical activity, and implementing modern information technologies systematically during the production of weapons, ammunition and military goods.

The article is devoted to the modern development of high technologies and computer technology greatly enhanced the development of automated banking systems of banking sector organizations and allowed the synthesis of information and communication technologies for their formation. The main purpose of the article is to select the main indicators for assessing the level of financial security of the banking system of the state and identify promising areas of its development using forecasting models. In the process of research such analytical functions have been used: polynomial, exponential, power and logarithmic. The authors believe that the information and analytical provision of the financial security of the bank is an information provision that combines, on the one hand, information work, that is, ways, means and methods of collecting the necessary information, and on the other - analytical work, which includes forms and methods of information analysis and processing, which ensures an objective assessment of the situation and the adoption of a balanced management decision. As a result, forecast models were built for each of the indicators and also, it has been found that most indicators of the banking system of Ukraine in 2021-2023 will remain at “unsatisfactory” and “critical” levels. In conclusions it was proposed to introduce measures that would be aimed at improving the reliability and stability of the banking system of Ukraine.

The Covid-19 outbreak has had a severe effect on the world economy. The company's business operations and profitability are damaged during the covid 19 outbreak. This deterioration is not only threatening the company’s survival position but also destroy the investor’s investment return. Therefore, it is vital to establish an effective early prediction technical method to foresee a corporate distress by a Pro-technical measurement to enhance the corporate sustainability. This study applies Altman Z-Score Model to as a Pro-Technology technique to the financial distress prediction of Malaysia’s Government Linked Plantation Companies (GLC-P) over a period of 10 years starting from 2012 to 2021. The significant contribution of the study is that the Z-Score Model provides an advanced indication tool regarding the financial stability of the respective GLC-P companies. The findings indicate that Financial Distress Prediction was dependent via in-time application of leverage, liquidity, activity, and profitability to the Altman Z-Score Model. Profitability and leverage were found to be superior prediction tool to financial distress.

Due to obstruction in photosynthesis, the leaves of the plants get affected by the disease. Powdery mildew is the main disease in cucumber plants which generally occurs in the middle and late stages. Cucumber plant leaves are affected by various diseases, such as powdery mildew, downy mildew and Alternaria leaf spot, which ultimately affect the photosynthesis process; that’s why it is necessary to detect diseases at the right time to prevent the loss of plants. This paper aims to identify and classify diseases of cucumber leaves at the right time using a deep convolutional neural network (DCNN). In this work, the Deep-CNN model based on disease classification is used to enhance the performance of the ResNet50 model. The proposed model generates the most accurate results for cucumber disease detection using data enhancement based on a different data set. The data augmentation method plays an important role in enhancing the characteristics of cucumber leaves. Due to the requirements of the large number of parameters and the expensive computations required to modify standard CNNs, the pytorch library was used in this work which provides a wide range of deep learning algorithms. To assess the model accuracy large quantity of four types of healthy and diseased leaves and specific parameters such as batch size and epochs were compared with various machine learning algorithms such as support vector machine method, self-organizing map, convolutional neural network and proposed method in which the proposed DCNN model gave better results.

Net Asset Value (NAV) has long been a key performance metric for mutual fund investors. Due to the considerable fluctuation in the NAV value, it is risky for investors to make investment decisions. As a result, accurate and reliable NAV forecasts can help investors make better decisions and profit. In this research, we have analysed and compared the NAV prediction performance of our proposed deep learning models, such as N-BEATS and NBSL, with the FLANN model in both univariate and multivariate settings for five Indian mutual funds for forecast periods of 15, 20, 45, 63, 126, and 252 days using RMSE, MAPE, and R2 as evaluation metrics. A large forecast horizon was chosen to assess the model's consistency, reliability, and accuracy. The result reveals that the N-BEATS model outperforms the FLANN and NBSL models in the univariate setting for all datasets and all prediction horizons. In a multivariate setting, the outcome demonstrates that the N-BEATS model outperforms the FLANN model across all datasets and prediction horizons. The result also shows that, as the number of forecast days grew, our suggested models, notably N-BEATS, maintained consistency and attained the highest R2 value throughout the longest forecast duration.

The application of machine learning technologies for cancer detection purposes are rising due to their ever-increasing accuracy. Melanoma is one of the most common types of skin cancer. Detection of melanoma in the early stages can significantly prevent illness and fetal death. The application of innovative machine learning technology is highly relevant and valuable due to medical practitioners' difficulty in early-stage diagnoses. This paper provides an open-source tutorial on the performance of an algorithm that helps to diagnose melanoma by extracting features from dermatoscopic images and their classification. First, we used a Dull-Razor preprocessing method to remove extra details such as hair. Next, histogram adjustments and lighting thresholds were used to increase the contrast and select lesion boundaries. After using a threshold, a binary-classified version of image was obtained, and the boundary of the lesion was determined. As a result, the features from skin tissue were extracted. Finally, a comparative study was conducted between three methods which are Artificial Neural Network (ANN), Support Vector Machine (SVM) and K-Nearest Neighbor (KNN). The results show that ANN could achieve better accuracy (83.5%). In order to mitigate the biases in existing studies, the source code of this research is available at hadi-naghavipour.com/ml to serve aspiring researchers for improvement, correction and learning and provide a guideline for technology manager practitioners.

Most real-time speech signals are frequently disrupted by noise such as traffic, babbling, and background noises, among other things. The goal of speech denoising is to extract the clean speech signal from as many distorted components as possible. For speech denoising, many researchers worked on sparse representation and dictionary learning algorithms. These algorithms, however, have many disadvantages, including being overcomplete, computationally expensive, and susceptible to orthogonality restrictions, as well as a lack of arithmetic precision due to the usage of double-precision. We propose a greedy technique for dictionary learning with sparse representation to overcome these concerns. In this technique, the input signal's singular value decomposition is used to exploit orthogonality, and here the ℓ1-ℓ2 norm is employed to obtain sparsity to learn the dictionary. It improves dictionary learning by overcoming the orthogonality constraint, the three-sigma rule-based number of iterations, and the overcomplete nature. And this technique has resulted in improved performance as well as reduced computing complexity. With a bit-precision of Q7 fixed-point arithmetic, this approach is also used in resource-constrained embedded systems, and the performance is considerably better than other algorithms. The greedy approach outperforms the other two in terms of SNR, Short-Time Objective Intelligibility, and computing time.

An administrator is employed to identify network security breaches in their organizations by using a Network Intrusion Detection and Prevention System (NIDPS), which is presented in this paper that can detect and preventing a wide range of well-known network attacks. It is now more important than ever to recognize different cyber-attacks and network abnormalities that build an effective intrusion detection system plays a crucial role in today's security. NSL-KDD benchmark data set is extensively used in literature, although it was created over a decade ago and will not reflect current network traffic and low-footprint attacks. Canadian Institute of Cyber security introduced a new data set, the CICIDS2017 network data set, which solved the NSL-KDD problem. With our approach, we can apply a variety of machine learning techniques like linear regression, Random Forest and ID3. The efficient IDPS is indeed implemented and tested in a network environment utilizing several machine learning methods. A model that simulates an IDS-IPS system by predicting whether a stream of network data is malicious or benign is our objective. An Enhanced ID3 is proposed in this study to identify abnormalities in network activity and classify them. For benchmark purposes, we also develop an auto encoder network, PCA, and K-Means Clustering. On CICIDS2017, a standard dataset for network intrusion, we apply Self-Taught Learning (STL), which is a deep learning approach. To compare, we looked at things like memory, Recall, Accuracy, and Precision.

With the development of a new stage of the industrial revolution, the importance of digitalization of business development tools is growing. The purpose of this article is to study the applied aspects of digital marketing tools usage for business development in the B2C market. To achieve the purpose and objectives of the study general and special methods are used: comparative analysis of the results of economic and statistical surveys; method of expert assessments by questionnaires using a 5-point Likert scale. The concordance coefficient was used to determine the consistency of the experts' opinions taking into account the related ranks in method of expert assessments. According to the results of the research, it is established that the Ukrainian business of the B2C sector was actively mastering digital marketing tools. The analysis of penetration level of digital technologies in the development of trade business showed the emergence of basic conditions for updating marketing tools to influence the B2C market. There is a rapid coverage rate of multi-purpose use of the Internet among consumers and businesses; gradual growth of digital skills among practitioners; positive dynamics of development of interactive services in the trade sphere. However, the level of use of the retail businesses websites remains low in many spheres of customer service. An important trend of the current development stage of the consumer market is the usage of business Internet platforms designed for mass dissemination of information. Effective marketing channels of interaction with consumers include social media (social networks, blogs or microblogs, websites with multimedia content, knowledge sharing tools), websites, e-shops, and sales via mobile devices. According to the results of expert evaluation, foreground digital technologies, which are able to bring business to a qualitatively new level of interaction with consumers and the provision of trade services have been identified. These are artificial intelligence and cognitive technologies, BigData, Internet of Things (IoT), and cloud computing. The structural and logical scheme of research of digital marketing tools is used for business development which includes two stages is offered. In the first stage, trendwatching, benchmarking and evaluation of internal opportunities for the use of digital marketing tools are performed. In the second stage, three components of digital readiness of business are defined: technological; competence; institutional. The obtained results form the basis of further research to determine the priorities of adaptive digital business behavior for the productive use of existing digital opportunities.

On a global scale, one of the prevalent causes of dementia is Alzheimer’s disease (AD). It will cause a steady deterioration in the individual from the mild stage to the severe stage, and thus impair their capacity to finish any tasks with no aid. The diagnosis is done with the utilization of existing methods which include medical history; neuropsychological testing as well as MRI (Magnetic Resonance Imaging), a lack of sensitivity as well as precision does affect the consistency of efficient procedures. With the deep learning network’s utilization, it is possible to create a framework for detecting specific AD characteristics from the MRI images. While automatic diagnosis is done with the application of diverse machine learning techniques, the existing ones do suffer from certain constraints with regards to accuracy. Thus, this work’s key goal is to increase the classification’s accuracy through the inclusion of a pre-processing approach prior to the deep learning model. The Alzheimer's disease Neuroimaging Initiative (ADNI) database of AD patients was used to develop a deep learning approach for AD identification. In addition, this study will present ideas for Haralick features, feature extraction from Local Binary Pattern (LBP), Artificial Neural Network (ANN), and Visual Geometry Group (VGG)-19 network techniques. The results of the experiments show that the deep learners offered are more effective than other systems already in use.