Speech is the most efficient and convenient way of communication. The learning capabilities of the deep learning architecture can be used to develop the sound classification system to overcome the efficiency issues of the traditional systems. We propose to develop a model that classifies the audio of the speaker.

Big Data faces many challenges with respect to the security while storing them on a cloud server. There is high chance of getting data viewed by the hacker and the server for performing various operations. In order to provide high level of confidentiality and integrity, a new Encryption technique is introduced known as Attribute Based Encryption (ABE), which instead of making use of the receiver’s public key for encryption, uses various attributes. As a future scope, ABE can be combined with Homomorphic Encryption (HE) to provide a secure transfer of the identity. In order to provide more privacy of data, web log Analyzer is used to find out the loopholes and the unauthorized access to the web data

This paper attempts to perform a comparative analysis of brain signals dataset using various machine learning classifiers such as random forest, gradient boosting, support vector machine, extra trees classifier. The comparative analysis is accomplished based on the performance parameters such as accuracy, area under the ROC curve (AUC), specificity, recall, and precision. The key focus of this paper is to exercise the machine learning practices over an Electroencephalogram (EEG) signals dataset provided by Rochester Institute of Technology and to provide meaningful results using the same. EEG signals are usually captivated to diagnose the problems related to the electrical activities of the brain as it tracks and records brain wave patterns to produce a definitive report on seizure activities of the brain. While exercising machine learning practices, various data preprocessing techniques were implemented to attain cleansed and organized data to predict better results and higher accuracy. Section II gives a comprehensive presurvey of existing work performed so far on the same; furthermore, section III sheds light on the dataset used for this research.

In this paper, a framework for localization of multiple co-channel transmitters using phase difference measurements between two antennas mounted on sensors of a sensor network is proposed. To pursue localization, we equip each sensor with two antennas and we use temporal cross-correlations between the received signals of the {two} antennas to extract the phase differences between each antenna pairs, named as phase interferometry measurements (PIMs), provoked by each {transmitters} using tensor decomposition. We calculate Cramer-Rao lower bound of error of localization using PIMs. Our simulation results show that highly accurate estimations can be achieved using PIMs. We also compare the accuracy of our proposed technique with a sensor network that exploits highly directional linear array antennas and show that {our} proposed technique can perform similar to a network that employs very large antenna arrays.

Spreading fake news has become a serious issue in the current social media world. It is broadcasted with dishonest intentions to mislead people. This has caused many unfortunate incidents in different countries. The most recent one was the latest presidential elections where the voters were mis lead to support a leader. Twitter is one of the most popular social media platforms where users look up for real time news. We extracted real time data on multiple domains through twitter and performed analysis. The dataset was preprocessed and user_verified column played a vital role. Multiple machine algorithms were then performed on the extracted features from preprocessed dataset. Logistic Regression and Support Vector Machine had promising results with both above 92% accuracy. Naive Bayes and Long-Short Term memory didn’t achieve desired accuracies. The model can also be applied to images and videos for better detection of fake news.

This paper describes our research work in regards building a data aggregation web service for eSports domain,

With new technologies, increasing mechanization and automation, almost all challenging social and economic issues that people face can be investigated towards a prediction or a solution. Digital technologies have connected the world in a single knot, but the technical knowhow does not contribute to solving these challenges. Encapsulation of technological advancement and social responsibility is highly needed. This calls for cross-regiment and inter-field solutions coupled with social enthusiasm. This survey provides a study on two aspects of social relations within a connected world. It also investigates and reviews solutions and work for advanced interactive technology. The survey draws on major insights from few researchers highlighted in our reference list.

IoT devices (wireless sensors, actuators, computer devices) produce large volume and variety of data and the data produced by the IoT devices are transient. In order to overcome the problem of traditional IoT architecture where data is sent to the cloud for processing, an emerging technology known as fog computing is proposed recently. Fog computing brings storage, computing and control near to the end devices. Fog computing complements the cloud and provide services to the IoT devices. Hence, data used by the IoT devices must be cached at the fog nodes in order to reduce the bandwidth utilization and latency. This chapter discusses the utility of data caching at the fog nodes. Further, various machine learning techniques can be used to reduce the latency by caching the data near to the IoT devices by predicting their future demands. Therefore, this chapter also discusses various machine learning techniques that can be used to extract the accurate data and predict future requests of IoT devices.

Java Server Faces is a framework to build a web application in java. We using Java Server Faces framework to build Student Registration System. We using MVC architecture in our model. In this paper, we given detail explanation about all the components which are helpful to build the application.

A paper exploring one of the applications of Artificial Intelligence in the field of Software Development

Our new idea of using a simple FMCW radar as an electronic walking cane is described. The letter goes beyond the idea by proposing the required architecture architecture, and even proposed antenna type and usable COTS part numbers. The idea was submitted to IEEE AP-S Student design contest in 2020.

— Skin-cancer is the most common type of cancer among all other types of cancers spreading both developed and developing countries. In this paper, a web service is developed in order to help physicians as well as individuals to upload an image and diagnose the particular types of lesion. Computer vision is used to analyse lesions on images by providing computers with somewhat similarity as humans do. For this a Convolution Neural Network (CNN) is used with multi classification on International Skin Imaging Collaboration (ISIC) 2018 dataset with HAM10000 images. This dataset is a meta-data which has various types of images with seven different labels. At first, the model is trained with a larger training set and saved in a zip folder. Secondly, a web service is created where users or a doctor can upload an image for classification. Thirdly, the images uploaded are pre-processed as there is noise, hairs on image. Techniques like resizing, normalisation, thresholding, black-hat filtering and inpainting are used for this purpose. After this, the saved model is called to define whether the uploaded image is benign or malignant. The experimental results reveal that the proposed model is superior in terms of detection and diagnosis accuracy as compared to modern methods.

This paper uses the random forest algorithm model to quantify and predict the monetary policy of the People’s Bank of China under the input of 16 indicators macroeconomic indicators. It is compared with three other machine learning algorithms (CART decision tree, support vector machine and neural network algorithm), discrete selection model and combined prediction model. The results show that the random forest algorithm shows better prediction accuracy in predicting the direction of the central bank’s monetary policy.

In this paper, we demonstrate, for the first time, an isolating bandpass filter with low-loss forward transmission and high reverse isolation by modulating its constituent resonators. To understand the operating principle behind the device, we develop a spectral domain analysis method and show that the same-frequency nonreciprocity is a result of the nonreciprocal frequency conversion to the intermodulation (IM) frequencies by the time-varying resonators. With appropriate modulation frequency, modulation depth, and phase delay, the signal power at the IM frequencies is converted back to the RF frequency and adds up constructively to form a low-loss forward passband, whereas they add up destructively in the reverse direction to create the isolation. To validate the theory, a lumped-element three-pole 0.04-dB ripple isolating filter with a center frequency of 200 MHz and a ripple bandwidth of 30 MHz is designed, simulated, and measured. When modulated with a sinusoidal frequency of 30 MHz, a modulation index of 0.25, and an incremental phase difference of 45°, the filter achieves a forward insertion loss of 1.5 dB and a reverse isolation of 20 dB. The measured nonmodulated and modulated results agree very well with the simulations. Such nonreciprocal filters may find applications in wideband simultaneous transmit and receive radio front ends.

This paper reports an approach to designing compact high efficiency millimeter-wave fundamental oscillators operating above the fmax=2 of the active device. The approach takes full consideration of the nonlinearity of the active device and the finite quality factor of the passive devices to provide an accurate and optimal oscillator design in terms of the output power and efficiency. The 213-GHz single-ended and differential fundamental oscillators in 65-nm CMOS technology are presented to demonstrate the effectiveness of the proposed method. Using a compact capacitive transformer design, the single-ended oscillator achieves 0.79-mW output power per transistor (16 μm) at 1.0-V supply and a peak dc-to-RF efficiency of 8.02% (VDD=0.80 V) within a core area of 0.0101mm2, and the measured phase noise is -93:4 dBc/Hz at 1-MHz offset. The differential oscillator exhibits approximately the same performance. A 213-GHz fundamental voltage-controlled oscillator (VCO) with bulk tuning method is also developed in this work. The measured peak efficiency of the VCO is 6.02% with a tuning rang of 2.3% at 0.6-V supply.

This paper presents the development and implementation of a low-cost smart meter. The smart meter is one of the major components of the distribution network. The proposed low-cost smart meter is compared with already existing smart meters deployed in the market. The comparison metrics involve cost, security, installation, and lifetime. In this study, we have presented the market segment and the beachhead market of the proposed product. The TAM size is calculated for better market analysis by calculating the overall customer size. A complete proposed process flow is explained for cost saving comparison, along with a complete market survey. In this paper, the whole technology and hardware needed for the low-cost smart meter is explained in detail along with the total cost of testing and installation.

The work presented here works on getting higher performances for image recognition task using convolutional neural networks on the MNIST handwritten digits data-set. A range of techniques are compared for improvements with respect to time and accuracy, such as using one-shot Extreme Learning Machines (ELM) in place of the iteratively tuned fully-connected networks for classification, using transfer learning for faster convergence of image classification, and improving the size of data-set and making robust models by image augmentation. The final implementation is hosted on cloud as a web-service for better visualization of the prediction results.

The advancement of Artificial Intelligence & Deep Learning has catalyzed the field of technology. The progression in these fields is exponentially increasing, and the discoveries which were once just an imagination are now changed into reality. The evolution of cars each year has made a lot of difference in people travelling from one place to another. One such reform involving Artificial Intelligence & Deep Learning is the birth of a self-driving car. The future is here where one can reach their destination hassle-free safely without the fear of accidents. This paper introduces a practical model of the self-driving robotics car, which can travel from one position to another on different types of tracks. A Pi-camera module is attached with the help of Raspberry Pi, which sends series of image frames to the Convolutional neural network, which then foretells the car to move in a specific direction, i.e. right, left, forward and reverse direction. The outcome is the robotics car, which travels in the desired direction without any individual effort.