Search

This paper proposes a novel approach for analysing the structure of Electronic Health Records (EHRs) using deep learning. Different types of EHRs and their respective structures are discussed. The authors then explore different approaches to applying deep learning to the task of analysing EHRs. This includes an overview of deep learning models, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), and generative adversarial networks (GANs). The authors propose a framework for applying deep learning to the task of EHR structure analysis. The framework consists of a pre-processing step to clean the data and a model to classify the EHRs. Finally, the authors provide a case study to demonstrate the efficacy of their proposed approach. Deep learning is a powerful and versatile tool for analysing the structure of an Electronic Health Record (EHR). Deep learning can be used to identify patterns and relationships between EHR data elements such as patient demographics, diagnoses, medications, and laboratory results. This information can then be used to generate predictive models and identify potential areas of improvement within the healthcare system. Deep learning techniques can also be leveraged to gain insights into the interactions between medical entities and healthcare processes, such as disease comorbidities or medication-related adverse events. By understanding the structure of an EHR, deep learning can provide valuable insights into the quality of care and help clinicians make more informed decisions.

Now a days Extreme Learning Machine has gained a lot of interest because of its noteworthy qualities over single hidden-layer feedforward neural networks and the kernel functions. Even if ELM has many advantages, it has some potential shortcomings such as performance sensitivity to the underlying state of the hidden neurons, input weights and the choice of functions of activation. To overcome the limitations of traditional ELM, analysts have devised numerical methods to optimise specific parts of ELM in order to enhance ELM performance for a variety of complicated difficulties and applications. Hence through this study, we intend to study the different algorithms developed for optimizing the ELM to enhance its performance in the aspects of survey criteria such as datasets, algorithm, objectives, training time, accuracy, error rate and the hidden neurons. This study will help other researchers to find out the research issues that lowering the performance of the ELM.

Graphs have proved to be a promising data structure to solve complex problems in various domains. Graphs store data in an associative manner which is analogous to the manner in which humans store memories in the brain. Generathe chatbots lack the ability to recall details revealed by the user in long conversations. To solve this problem, we have used graph-based memory to recall-related conversations from the past. Thus, providing context feature derived from query systems to generative systems such as OpenAI GPT. Using graphs to detect important details from the past reduces the total amount of processing done by the neural network. As there is no need to keep on passingthe entire history of the conversation. Instead, we pass only the last few pairs of utterances and the related details from the graph. This paper deploys this system and also demonstrates the ability to deploy such systems in real-world applications. Through the effective usage of knowledge graphs, the system is able to reduce the time complexity from O(n) to O(1) as compared to similar non-graph based implementations of transfer learning- based conversational agents.

Deepfake algorithms can make forged pictures and videos that people cannot differentiate them from true ones. The suggestion of technology that locates and proves the truth of virtual visual media is as a result essential. Deepfakes generate realistic forged images or videos of targeted persons by swapping their faces another person saying or doing things that are not really done by them and public start trusting in such forged videos, as it is not identifiable with the normal human eye. This paper offers a novel method to locate Deepfakes. We present discussions on challenges, studies, advances and strategies associated with Deepfake along with the method based on long-term recurrent convolution network (LRCN). By reviewing the history of Deepfakes and cutting-edge Deepfake detection strategies, this method will give a new approach towards the Deepfake technology and will help in the advancement of new and strong strategy to deal with an increasing number of Deepfakes.

This research paper focuses on optimization-based extreme learning machine (ELM), in order to enhance generalization ability and learning efficiency of ELM. The study explains new approach which enhances the efficiency of basic ELM algorithm and also reduces the problems of unmodified ELMs of generalization ability by reducing redundant information and low learning efficiency. In most cases, the proposed method can produce the generalization performance on the basis of a few classification and regression problems, and it was found to be competent. The proposed model has a greater potential of reducing the training time, improving the predictive accuracy and robustness. In the future, efforts will be made to attain a higher level of efficiency.

Deepfake algorithms can make forged pictures and videos that people cannot differentiate them from true ones. The suggestion of technology that locate and prove the truth of virtual visual media is as a result essential. Deepfakes generates realistic forged images or videos of targeted persons by swapping their faces another person saying or doing things that are not really done by them and public start trusting in such forged videos, as it is not identifiable with the normal human eye. This paper offers a survey of tools and algorithms used to make deepfakes and, additional significantly, methods to locate deepfakes. We present huge discussions on challenges, studies, advances and strategies associated to deepfake. By reviewing the history of deepfakes and cutting-edge deepfake detection strategies, this gives a comprehensive assessment of deepfake techniques and helps the development of latest and more robust strategies to deal with an increasing number of tough deepfakes.

In recent years widespread rumors and fake news has given rise to many social and political problems. Most of the information today is acquired from digital sources. In Digital media it is difficult to assign accountability to the opinion due to which the data received cannot be authenticated. Lack of constant supervision has motivated the miscreants to spread fake information. Fake news articles that are planted over digital media shares important linguistic features such as immoderate usage of unconfirmed hyperbole and non-verified quotes. It is necessary to invent an automated mechanism to identify fake news and also to minimize its impact by restricting its spread. This survey comprehensively and systematically studies different methodologies in the detection of fake news in digital media. The survey identifies and specifies fundamental theories in Machine Learning, to facilitate and enhance the research of fake news detection. By understanding the different methodologies in fake news studies, we highlight some potential research gaps at the end of this survey.

In today’s digitized era, data is one of the most important entities. Every system generates a lot of data and even it has to deal with bulk data. The efficiency of the system depends on how the data is available and in which form the data is stored. With the advent of newer techniques of storing and processing the data, the medical domain has leaped the form of the electronic patient record (EPR) or electronic health record (EHR). This paper is based on a study of EHR which aims at highlighting its significance and emphasize its importance for medical reforms. In this paper, our sincere efforts were to introduce the concept of EHR with the help of literature that was already in place. Further, we have focussed on the proposed structure of EHR along with the different categories. Even though the EHR system has added a lot of benefits, there were many challenges in its adoption and implementation. We had tried enlisting the challenges for the EHR system and sincerely tried suggesting some measures to mitigate the same. Keeping aside the obstacles in a way of EHRs, numerous existing applications have been discussed further. Finally, we have given some novel and evolving usages of the EHR system which has increased its productivity to a greater extent. Overall our paper tries to give a complete overview of the EHR system which shall be useful in earning a basic understanding of the concept. It shall prove to be good literature for the novices working on the EHR related research.

Agriculture plays an important role across the globe. Nowadays manual plant disease identification has become difficult and time-consuming. Plant disease is an important aspect that reduces plant growth. According to the studies, plant disease has reduced the quality of plants in agriculture. Machine Learning approach helps to identify these plant diseases. Hence diseases like rice blast, stem rot, leaf scald, smut tungro, bacterial blight, brown spot, sheath blight, false, bacterial leaf streak, and fungal diseases, etc. can be identified using machine learning methods like clustering, classification, regression, etc. Different plant images that are affected by diseases are captured as well as wireless sensor network (WSN) also helps in disease identification. These images can be processed, and diseases can be identified. This practice gives rise to plant yield. This survey paper describes the machine learning approach for plant disease identification and a detailed study of various techniques is classified.

Classification and function approximation tasks are used widely in artificial neural network (ANN), and there are a number of algorithms by which a neural network can be trained. ELM is known as extreme learning machine; a novel algorithm has come under the spotlight in the last decade, and the reason for its popularity is a rapid learning rate and a more efficient generalization capability, in comparison with orthodox gradient descent learning algorithms for (SFLN) single-hidden-layer feed-forward neural network. Even though it has many advantages, the single-hidden-layer feed-forward neural network trained using ELM requires a number of hidden neurons. Due to this, it faces poor condition problem because of the randomly created hidden biases and input weights. The solution to this problem is to use an optimization technique, using it to find a set of inputs that are optimal. We have proposed a hybrid learning scheme that uses one such optimization technique which will be used to figure out an optimal set of input-hidden node weights, and then a set of optimal input weights are used to determine the output weights analytically. Few of the techniques that are used for optimization like competitive swarm optimization (CSO) and particle swarm optimization (PSO) have been proposed in the past. The hybrid learning scheme provides (i) help in enhancing the search variety, which encourages us to explore the solution search space better and (ii) improve the structure of ELM to solve the problem of random input. The proposed scheme in addition to other multi-objective optimal algorithms requires smaller amounts of hidden neurons compared to the conventional ELM for achieving higher classification accuracy. The proposed hybrid methodology has the ability to produce good condition SLFNs with an improved accuracy with better generalization.

Stock price forecasting phenomenon has been majorly made on the basis of quantitative information. Over the time, with the advent of technology, stock forecasting used technical analysis to get more accurate predictions. Until recently, studies have demonstrated that sentiment information hidden in corporate reports can be effectively incorporated to predict short-run stock price returns. Soft computing methods, like neural networks, fuzzy models and support vector regression, have shown great results in the forecasting of stock price due to their ability to model complex non-linear systems.In this paper we propose a hybrid method for stock price predication, which is combinational feature from technical analysis and sentiment analysis (SA). The features of sentiment analysis are based on a Point wise Mutual Information (PMI) and we apply neural network and ε-support vector regression models to predict the yearly change in the stock price.