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.

Testing for COVID-19 has been unable to keep up with the demand. Further, the false negative rate is projected to be as high as 30% and test results can take some time to obtain. X-ray machines are widely available and provide images for diagnosis quickly. This paper explores how useful chest X-ray images can be in diagnosing COVID-19 disease. We have obtained 135 chest X-rays of COVID-19 and 320 chest X-rays of viral and bacterial pneumonia. A pre-trained deep convolutional neural network, Resnet50 was tuned on 102 COVID-19 cases and 102 other pneumonia cases in a 10-fold cross validation. The results were an overall accuracy of 89.2% with a COVID-19 true positive rate of 0.8039 and an AUC of 0.95. Pre-trained Resnet50 and VGG16 plus our own small CNN were tuned or trained on a balanced set of COVID-19 and pneumonia chest X-rays. An ensemble of the three types of CNN classifiers was applied to a test set of 33 unseen COVID-19 and 218 pneumonia cases. The overall accuracy was 91.24% with the true positive rate for COVID-19 of 0.7879 with 6.88% false positives for a true negative rate of 0.9312 and AUC of 0.94. This preliminary study has flaws, most critically a lack of information about where in the disease process the COVID-19 cases were and the small data set size. More COVID-19 case images at good resolution will enable a better answer to the question of how useful chest X-rays can be for diagnosing COVID-19.

In this article, we present the first demonstration of distributed and symmetrical all-band quasi-absorptive filters that can be designed to arbitrarily high orders. The proposed quasi-absorptive filter consists of a bandpass section (reflective-type coupled-line filter) and absorptive sections (a matched resistor in series with a shorted quarter-wavelength transmission line). Through a detailed analysis, we show that the absorptive sections not only eliminate out-of-band reflections but also determine the passband bandwidth (BW). As such, the bandpass section mainly determines the out-of-band roll-off and the order of the filter can be arbitrarily increased without affecting the filter BW by cascading more bandpass sections. A set of 2.45-GHz one-, two-, and three-pole quasi-absorptive microstrip bandpass filters are designed and measured. The filters show simultaneous input and output absorption across both the passband and the stopband. Measurement results agree very well with the simulation and validate the proposed design concept.

A document by Raja Mukhopadhyay . Click on the document to view its contents.

Visual navigation tasks in real-world environments often require both self-motion and place recognition feedback. While deep reinforcement learning has shown success in solving these perception and decision-making problems in an end-to-end manner, these algorithms require large amounts of experience to learn navigation policies from high-dimensional data, which is generally impractical for real robots due to sample complexity. In this paper, we address these problems with two main contributions. We first leverage place recognition and deep learning techniques combined with goal destination feedback to generate compact, bimodal image representations that can then be used to effectively learn control policies from a small amount of experience. Second, we present an interactive framework, CityLearn, that enables for the first time training and deployment of navigation algorithms across city-sized, realistic environments with extreme visual appearance changes. CityLearn features more than 10 benchmark datasets, often used in visual place recognition and autonomous driving research, including over 100 recorded traversals across 60 cities around the world. We evaluate our approach on two CityLearn environments, training our navigation policy on a single traversal. Results show our method can be over 2 orders of magnitude faster than when using raw images, and can also generalize across extreme visual changes including day to night and summer to winter transitions.

This paper provides a comparative analysis between Deep Q Network (DQN) and Double Deep Q Network (DDQN) algorithms based on their hit rate, out of which DDQN proved to be better for Breakout game. DQN is chosen over Basic Q learning because it understands policy learning using its neural network which is good for complex environment and DDQN is chosen as it solves overestimation problem (agent always choses non-optimal action for any state just because it has maximum Q-value) occurring in basic Q-learning.

Social network analysis has been an essential topic with broad content sharing from social media. Defining the directed links in social media determine the flow of information and indicates the user's influence. Due to the enormous data and unstructured nature of sharing information, there are several challenges caused while handling data. Graph Analytics proves to be an essential tool for addressing problems such as building networks from unstructured data, inferring information from the system, and analyzing the community structure of a network. The proposed approach aims to determine the influencers on Twitter data, based on the follower's count as well as the retweet count. Several graph-based algorithms are implemented on the data collected to find the influencer as well as communities in the network.

A project on DDI extraction from medical text using a Convolutional Neural Network Model.

A document by Matthias Quester . Click on the document to view its contents.

The given literature focuses on developing a Smart Factory model based on Cloud and Edge computing used to develop Transportation Management System(TMS) using a iFogSim wrapper. Cloud computing identifies data centres for users and offer computer system services on-demand, including data storage and processing power, without direct active user management. In the smart industry, several devices are connected together across the internet, where vast volumes of data are collected during the entire process of output. Thus, to handle this data smart factory based on cloud and edge computing is used. The intelligent cloud-based factory offers some facility like large scale analysis of data. Concepts like fog and edge computing play a significant role in extending data storage and network capacities in the cloud that addresses some challenges, such as over-full bandwidth and latency. The literature also focuses on the implementation of TMS using the iFogSim Simulator. The simulator provides efficient execution of TMS by showing the amount of resources used which gives an idea regarding optimum use of resources. All types of data related to TMS is obtained at cloud by using smart factory like object location, time taken and energy consumption. To implement the TMS we have created a topology which displays various devices connected to the cloud which gives necessary information regarding the ongoing transportation simulation.

Time-periodicity and non-linearity pose a challenge to the precise input impedance modeling of single-phase power converters. In this study, a precise input impedance model with measurability of the single-phase voltage source rectifier (VSR), which considers the frequency-coupling effect (FCE), is established. Meanwhile, it is revealed that the rectifier input impedance is dependent of the grid impedance. In the proposed modeling approach, only Laplace transform and frequencyshifting operation are required, which avoids the complicated convolution calculation in the frequency domain. In addition, the influence of grid impedance on the input impedance is studied. Simulations are conducted to verify the effectiveness of the proposed method.

Layered two-dimensional (2D) GO films are integrated with silicon-on-insulator (SOI) nanowire waveguides to experimentally demonstrate an enhanced Kerr nonlinearity, observed through self-phase modulation (SPM). The GO films are integrated with SOI nanowires using a large-area, transfer-free, layer-by-layer coating method that yields precise control of the film thickness. The film placement and coating length are controlled by opening windows in the silica cladding of the SOI nanowires. Owing to the strong mode overlap between the SOI nanowires and the highly nonlinear GO films, the Kerr nonlinearity of the hybrid waveguides is significantly enhanced. Detailed SPM measurements using picosecond optical pulses show significant spectral broadening enhancement for SOI nanowires coated with 2.2-mm-long films of 1−3 layers of GO, and 0.4-mm-long films with 5−20 layers of GO. By fitting the experimental results with theory, the dependence of GO’s n2 on layer number and pulse energy is obtained, showing interesting physical insights and trends of the layered GO films from 2D monolayers to quasi bulk-like behavior. Finally, we show that by coating SOI nanowires with GO films the effective nonlinear parameter of SOI nanowires is increased 16 fold, with the effective nonlinear figure of merit (FOM) increasing by about 20 times to FOM > 5. These results reveal the strong potential of using layered GO films to improve the Kerr nonlinear optical performance of silicon photonic devices.

Social media is a platform in which tons and tons of text are generated each and every day. The data is so large that cannot be easily understood, so this has paved a path to a new field in the information technology which is natural language processing. In this paper, the text data which is used for the classification is tweets that determines the state of the person according of the sentiments which is positive, negative and neutral. Emotions are the way of expression of the person’s feelings which has a high influence on the decision making tasks. Here we have proposed the text representation, Term Frequency Inverse Document Frequency (tfidf), Keras embedding along with the machine learning and deep learning algorithms for the purpose of the classification of the sentiments, out of which Logistics Regression machine learning based methods out performs well when the features is taken in the limited amount as the features increases Support Vector Machine (SVM) which is also one of the machine learning algorithm out performs well making a benchmark accuracy for this dataset as the 75.8%. For the research purpose the dataset has been made publically available.

To assess the reliability of dc energy measurement equipment on-board trains, a setup was developed to characterize current transducers under realistic operating conditions. The operating principle is based on a current ratio measurement technique. The reference sensor is a high-precision zero-flux current transducer in combination with a broadband high-precision current shunt. The influence of ac distortion on this reference sensor was found to be within a few parts in 106 using an initial version of the setup, in which ac distortion was applied through a separate winding. A revised version of the setup employs a programmable electronic load to apply dynamic currents up to 600 A. The setup was used to characterize a 100 μohm high-current shunt resistor. The effect of dissipative heating on the dc transresistance error was around 0.03 %, with a settling time of about half an hour. The short-term intrinsic current dependence was also around 0.03 %. The effect of ac distortion was within a few parts in 106. The intrinsic current dependence and the onset of the heating effect were also observed when exposing the sensor to a dynamic current profile that was recorded during a trip between two successive underground train stations on Metro de Madrid. These results demonstrate that the setup described in this paper is very effective for characterizing dc current sensors for practical railway applications. Future work will concentrate on even more demanding current signals, such as chopped signals, and on other types of sensors and measurement systems.

Bank safety deposit box is one of the secure storage for the clients to store their valuables such as important property or business paper, gold or even money but is in some case scenario traditional safe deposit are not safe enough, there have been many robberies from this safe box because of the code lock can be hacked and there no alarming system in this kind of safe boxes. In this paper I proposed a smart secure safe box with facial reorganization unlock feature using raspberry pi and uno Arduino which can used to secure your documents or valuables in a bank or at your home. In this I have used raspberry pi and pi camera to perform the face unlock and to send security alert through email. Secondly, I have used Arduino and hall sensor to trigger an alarm when the door is open unauthorized or kept open.

Developing Chatbot Wrapper for Online Shopping: A Case Study of Using Generic Mobile Messaging System. A conversational UI chatbot using Telegram API was implemented.

Here I have mentioned the essential on the home IoT frameworks and difficulties related to them. After that, I have portrayed about essential requirements of the IoT gadgets and sensors. Additionally, I have referenced about the procedures and methodologies which I have utilized for key less door entry. The reader can get the clear idea regarding the venture by reviewing the wiring diagram and other screenshots.

Civic technology is a fast-developing segment that holds huge potential for a new generation of startups. A recent survey report on civic technology noted that the sector saw $430 million in investment in just the last two years. It’s not just a new market ripe with opportunity it’s crucial to our democracy. Crowdsourcing has proven to be an effective supplementary mechanism for public engagement in city government in order to use mutual knowledge in online communities to address such issues as a means of engaging people in urban design. Government needs new alternatives – alternatives of modern, superior tools and services that are offered at reasonable rates. An effective and easy-to-use civic technology platform enables wide participation. Response to, and a ‘conversation’ with, the users is very crucial for engagement, as is a feeling of being part of a society. These findings can contribute to the future design of civic technology platforms. In this research, we are trying to introduce a crowdsourcing platform, which will be helpful to people who are facing problems in their everyday practice because of the government services. This platform will gather the information from the trending twitter tweets for last month or so and try to identify which challenges public is confronting. Twitter for crowdsourcing as it is a simple social platform for questions and for the people who see the tweet to get an instant answer. These problems will be analyzed based on their significance which then will be made open to public for its solutions. The findings demonstrate how crowdsourcing tends to boost community engagement, enhances citizens ‘ views of their town and thus tends us find ways to enhance the city’s competitiveness, which faces some serious problems. Using of topic modeling with Latent Dirichlet Allocation (LDA) algorithm helped get categorized civic technology topics which was then validated by simple classification algorithm. While working on this research, we encountered some issues regarding to the tools that were available which we have discussed in the ‘Counter arguments’ section.