Augmented Human Research最新文献

Attention-deficit/hyperactivity disorder (ADHD) is one of the most prevalent neuropsychiatric disorders in adolescence and adult, but the origin of this disorder is still under research. The focus of this paper is on classification of resting state functional magnetic resonance imaging (rs-fMRI) of ADHD and healthy controls using deep learning techniques. ADHD-200 dataset includes resting state rs-fMRI images of ADHD, and typically developing controls and deep learning-based techniques such as 2-dimensional convolutional neural network (CNN) algorithm and hybrid 2-dimensional convolutional neural network–long short-term memory (2D CNN–LSTM) were applied on this dataset for the classification of ADHD from typically developing controls. The proposed hybrid system evaluated on the basis of parameters, viz. accuracy, specificity, sensitivity, F1-score, and AUC. In comparison with existing methods, the proposed method achieved significant improvement in analyzing and detection of parameters. By incorporating techniques of deep learning with rs-fMRI, the results built up an adequate and intelligent model to comparatively diagnose ADHD from healthy controls.

Advertising is a way in which a company introduces possible customers to a company’s product/service, the main objective is possibly to convince them to buy their product or use their service. The significance of Advertising is critical for the company, as this alone can make people aware of the company’s product and in doing so can generate a good possibility of it being sold to the customers. It is inevitable for companies to face changes and one such change is the evolution in the way of doing Advertisement. Advertisement is now done with the help of not so newfound helping hand that is Artificial Intelligence and Machine Learning. The answer to the question as to why the change in the process of Advertising is important lies in the before-after statistical observations of companies using this technology. The results themselves are reasonable motivating factors for companies who are yet to acknowledge the change. The serious challenge to this new version of Advertising is to make sure to not allow the usage of it to such a great extent where ordinary person is concerned about his/her privacy. Implementing Advertisements this way, we are quite sure that making laws, enforcing the laws or even having its own governing body can ensure righteous use of deploying this technology. The future of Advertising is going to be even better than before as Artificial Intelligence and Machine Learning will bring more control of Advertising to companies. Summing up, we feel confident that Advertising with Artificial Intelligence and Machine Learning are here for a noticeable and a significant change.

The new pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), originated at Wuhan, Hubei province, China in December 2019, threatening the world and becomes the public health crisis throughout the globe. Due to changing data and behavior of the current epidemic, appropriate pharmacological techniques to cure are getting delayed day by day. The estimated trends of the global and Indian region for COVID-19 epidemic were predicted for the next 21 days till 05/05/2020 on the data recorded till 14/04/2020 in the present work. The main focus of the work was to estimate the trends of COVID-19 outbreak on population, especially the high-risk age group of elderly people (with age 50 years and greater) in the Republic of India. It was observed that this identified age-group could be more prone to SARS-CoV-2 virus infection and chances of death in this age group could be more. The high-risk Indian states/regions were also identified throughout the nation and trends for infection, death, and cured cases were predicted for the next 21 days. The outcome of the present work was presented in terms of suggestions that the proper social and medical care for the identified high-risk age group of elderly people of the Indian population should be required to prevent the COVID-19 community transmission. The work also supported the extension in countrywide proper lockdown, mass testing, and also the strict rules to follow social distancing.

This paper presents a modeling approach for defining a measured data set obtained from an optical sensing circuit based on the use of a fuzzy reasoning system. A simple but effective optical sensor is designed for in vitro determination of glucose concentrations in an aqueous solution. The measured data used in this study include analog voltages that reflect the absorbance values of three wavelengths measured in different concentrations of glucose from an RGB light-emitting diode (LED). The parameters of the fuzzy models are optimized using the bounded-range artificial bee colony (BR-ABC) algorithm to achieve the desired model performance. The results indicate that the optimized fuzzy model demonstrates high performance quality. The minimum mean square error (MSE) obtained from the singleton fuzzy model with the BR-ABC algorithm is 0.00014, which is better than the reported MSE value achieved with the Takagi–Sugeno fuzzy model.

The electrocardiogram (ECG) signal is a method that uses electrodes to record cardiac rates along with sensing minute electrical fluctuations for each cardiac rate. The information is utilized to analyze abrupt cardiac function like arrhythmias and conduction disturbance. The paper proposes strategy classifying ECG signal using various technique. The preprocessing stage includes filtering of input signal via low pass, high pass including Butterworth filter in order to remove clamour of high frequency. From signal, the excess clamour is sliced by Butterworth filter. The peak points are detected by peak detection algorithm, and the signal features are extracted using statistical parameters. At last, extracted feature classification is done via GWO-MSVM, SVM, Adaboost, ANN and Naive Bayes classifier to classify the ECG signal database into normal or abnormal ECG signal. The experimental result indicates the precision of the GWO-MSVM, SVM, Adaboost, ANN and Naive Bayes classifier is 99.9%, 94%, 93%,87.57% and 85.28%. When compared with other classifier, it was determined that precision of GWO-MSVM classifier is high.

Artificial intelligence (AI) is a broad field; this term signifies the application of a machine or computer to construct intelligent behaviour with insignificant human interruption or interference. AI is expressed as the combination of science and engineering for making intelligent computers. The term AI applies to a broad spectrum of matters in medicine and healthcare sectors like robotics, a medical diagnosis which concerns too many different types of diseases, human biology, and medical statistics. AI in medicine and health care is the main focus of this survey. Our goal is to highlight numerous algorithms based on the techniques which rely on artificially intelligent behaviour for detecting many diseases. We then review more precisely regarding AI applications in several categories of diseases such as hereditary diseases, physiological diseases, cancers, and infectious diseases. We have analysed the AI-based algorithms, and results for the same for the diseases included in the categories as mentioned above. Popular AI techniques include machine learning methods, along with the implementation of natural language processing. We have also discussed the impact of big data in the healthcare sector and how it has supported to improve the field of AI. An overview of various artificial intelligent methods is exhibited in this paper alongside the review of relevant important clinical applications.

In Human–Machine Teaming environments, it is important to identify potential performance drops due to cognitive overload. If identified correctly, they can help improve the performance of the human–machine system by offloading some tasks to less cognitively overloaded users. This can help prevent user error that can result in critical failures. Also, it can improve productivity by keeping the human operators at an optimal performance state. This paper explores a new method for identifying user cognitive load by a three-class classification using brain activity data and by applying a convolutional neural network and long short-term memory model. The data collected from a set of cognitive benchmark experiments were used to train the model, which was then tested on two separate datasets consisting of more ecologically valid task environments. We experimented with various models built with different benchmark tasks to explore which benchmark tasks were better suited for the prediction of task shedding events in these compound tasks that are more representative of real-world scenarios. We also show that this method can be extended across-tasks and across-subject pools.

Traffic system shows a great scope of trade with the environment and is directly connected to it. Manual traffic systems are proving to be insufficient due to rapid urbanization. Central monitoring systems are facing scalability issues as they process increasing amounts of data received from hundreds of traffic cameras. Major traffic problems include congestion, safety, pollution (leading to various health issues) and increased need for mobility. A solution to most of them is the construction of newer and safer highways and additional lanes on existing ones, but it proves to be expensive and often not feasible. Cities are limited by space, and construction cannot keep up with ever-growing demand. Hence, a need for an improved system with a minimal manual interface is persisting. One of such methods is introduced and discussed in this paper; smart traffic governance system here used artificial intelligence to regulate and govern the course of transport and automated administration and implementation to make a difference in face of travel scenarios in urban cities suffering from such major traffic issues.

In the current generation, a huge amount of textual documents are generated and there is an urgent need to organize them in a proper structure so that classification can be performed and categories can be properly defined. The key technology for gaining the insights into a text information and organizing that information is known as text classification. The classes are then classified by determining the text types of the content. Based on different machine learning algorithms used in the current paper, the system of text classification is divided into four sections namely text pre-treatment, text representation, implementation of the classifier and classification. In this paper, a BBC news text classification system is designed. In the classifier implementation section, the authors separately chose and compared logistic regression, random forest and K-nearest neighbour as our classification algorithms. Then, these classifiers were tested, analysed and compared with each other and finally got a conclusion. The experimental conclusion shows that BBC news text classification model gets satisfying results on the basis of algorithms tested on the data set. The authors decided to show the comparison based on five parameters namely precision, accuracy, F1-score, support and confusion matrix. The classifier which gets the highest among all these parameters is termed as the best machine learning algorithm for the BBC news data set.

The task of predicting a movie genre from its poster can be very challenging owing to the high variability of movie posters. A novel approach for the generation of a multi-label movie genre prediction from its poster using neural networks that employ knowledge transfer learning has been proposed in this paper. This approach works on two fronts; one is aimed at creating a large, diverse and balanced dataset for movie genre prediction. The second front involves reframing the problem to simpler single-label multi-class classification and generating a multi-label multi-class prediction on a given movie poster as input. The experimental evaluation suggests that our approach generates a remarkable accuracy which is a result of a larger, evenly distributed dataset, simplifying the problem to a single-label multi-class classification problem and because of the use of knowledge transfer learning to extract higher-level feature.