A cryptocurrency is a network-based digital exchange medium, where the records are secured using strong cryptographic algorithms such as Secure Hash Algorithm 2 (SHA-2) and Message Digest 5 (MD5). It uses blockchain technology to make the transactions secure, transparent, traceable, and immutable. Due to these properties, the cryptocurrencies have gained popularity in almost all the sectors especially in financial sectors. Though, cryptocurrencies are getting recognition form the approval bodies, but still, the uncertainty and dynamism in their prices risk the investments substantially. Cryptocurrency price prediction has become a trending research topic globally. Many machine learning and deep learning algorithms such as Gated Recurrent Unit (GRU), Neural Networks (NN), and Long short-term memory (LSTM) have been used by the researchers to predict and analyze the factors affecting the cryptocurrency prices. In this paper, a LSTM and GRU-based hybrid cryptocurrency prediction scheme is proposed, which focuses on only two cryptocurrencies, namely Litecoin and Monero. The results depict that the proposed scheme accurately predicts the prices with high accuracy, revealing that the scheme can be applicable in various cryptocurrencies price predictions.

A Deep Learning-based Cryptocurrency Price Prediction Scheme for Financial Institutions

The paper presents an approach and case study of a distributed driver monitoring system. The system utilizes smartphone sensors for detecting dangerous states for a driver in a vehicle. We use a mounted smartphone on a vehicle windshield directed towards the driver's face tracked by the front-facing camera. Using information from camera video frames as well as other sensors, we determine drowsiness, distraction, aggressive driving, and high pulse rate dangerous states that can lead to road accidents. We propose a cloud system architecture to capture statistics from vehicle drivers, analyze it and personalize the smartphone application for the driver. The cloud service provides reports on driver trips as well as statistics to developers. This allows to monitor and improve the system by developing modules for personification and taking into account context situation. We identified statistically that the driver eye closeness is related to the light brightness and drowsiness recognition should be adjusted accordingly.

Cloud-Based Driver Monitoring System Using a Smartphone

Blockchain technology is becoming increasingly popular because of its applications in various fields. It gives an edge over the traditional centralized methods as it provides decentralization, immutability, integrity, and anonymity. The most popular application of this technology is cryptocurrencies, which showed a massive rise in their popularity and market capitalization in recent years. Individual investors, big institutions, and corporate firms are investing heavily in it. However, the crypto market is less stable than traditional commodity markets. It can be affected by many technical, sentimental, and legal factors, so it is highly volatile, uncertain, and unpredictable. Plenty of research has been done on various cryptocurrencies to forecast accurate prices, but the majority of these approaches cannot be applied in real-time. Motivated from the aforementioned discussion, in this paper, we propose a deep-learning-based hybrid model (includes Gated Recurrent Units (GRU) and Long Short Term Memory (LSTM)) to predict the price of Litecoin and Zcash with inter-dependency of the parent coin. The proposed model can be used in real-time scenarios and it is well trained and evaluated using standard data sets. Results illustrate that the proposed model forecasts the prices with high accuracy compared to existing models.

/pdf/deep-learning-based-cryptocurrency-price-prediction-scheme-2uakybxh16.pdf

Deep Learning-Based Cryptocurrency Price Prediction Scheme With Inter-Dependent Relations

Deaths and injuries resulting from road traffic crashes remain a serious problem globally and current trends suggest that this will continue to be the case in the foreseeable future. In this paper, we propose a reliable method towards drowsiness detection by analyzing images of the driver’s face. In fact, the shadows caused by wearing glasses and/or bad light conditions may, in particular, decrease the accuracy rate of the blinking detection. In addition, drowsiness symptoms are not just restricted to the frequency of blinking, as most of existing works do, but also include yawning and nodding. Thus, instead of using a single facial feature to predict the drowsiness state, we jointly combine eye closure and yawning by measuring the eye and the mouth aspect ratios, and head pose which is estimated by analyzing the optical flow. Then, we investigate the multilayer perceptron and the K-nearest neighbors as two classification techniques. The proposed method has been tested on the challenging private NTHU-DDD benchmark video dataset, and the preliminary experimental results show the effectiveness of the proposed automated method.

Driver Drowsiness Detection Based on Joint Monitoring of Yawning, Blinking and Nodding

Sentiment analysis is a considerable research field to analyze huge amount of information and specify user opinions on many things and is summarized as the extraction of users’ opinions from the text Like sentiment analysis, Bitcoin which is a digital cryptocurrency also attracts the researchers considerably in the fields of economics, cryptography, and computer science The purpose of this study is to forecast the direction of Bitcoin price by analysing user opinions in social media such as Twitter To our knowledge, this is the very first attempt which estimates the direction of Bitcoin price fluctuations by using deep learning and word embedding models in the state-of-the-art studies For the purpose of estimating the direction of Bitcoin, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long-short term memory networks (LSTMs) are used as deep learning architectures and Word2Vec, GloVe, and FastText are employed as word embedding models in the experiments In order to demonstrate the contibution of our work, experiments are carried out on English Twitter dataset Experiment results show that the usage of FastText model as a word embedding model outperforms other models with 8913% accuracy value to estimate the direction of Bitcoin price

https://www.ijisae.org/IJISAE/article/download/1062/616

Sentiment Analysis Based Direction Prediction in Bitcoin using Deep Learning Algorithms and Word Embedding Models

Investment in cryptocurrency has been in trend from last many years. Bitcoin is one of the most popular and valuable cryptocurrency. Many studies have been done on bitcoin price prediction using various parameters which includes bitcoin factors, social media etc. In this paper, a comparative study of the various parameters affecting bitcoin price prediction is done based on Root Mean Square Error (RMSE) using various deep learning models like Convolutional Neural Network (CNN), Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). We have studied the effect of Gold price on the price of bitcoin.

Deep Learning Approach to Determine the Impact of Socio Economic Factors on Bitcoin Price Prediction

One of the most prevailing problems across the globe nowadays is the booming number of road accidents. Improper and inattentive driving is one of the major cause of road accidents. Driver's drowsiness or lack of concentration is considered as a dominant reason for such mishaps. Research in the field of driver drowsiness monitoring may help to reduce the accidents. This paper therefore proposes a non-intrusive approach for implementing a driver's drowsiness alert system which would detect and monitor the yawning and sleepiness of the driver. The system uses Histogram Oriented Gradient (HOG) feature descriptor for face detection and facial points recognition. Then SVM is used to check whether detected object is face or non-face. It further monitors the Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) of the driver up to a fixed number of frames to check the sleepiness and yawning. Since the drowsiness or tiredness of the driver is also based on the number of hours he or she has been driving, an additional feature of varying the threshold frames for eyes and mouth is included. This makes the system more sensitive towards drowsiness detection. Also, this requires the inclusion of face recognition implementation so that monitoring can be done individually for every driver. Our experimental results shows that our proposed framework perform well.

Real-Time Driver's Drowsiness Monitoring Based on Dynamically Varying Threshold

Identification of relevant features influencing movie reviews using sentiment analysis

Data poisoning attacks are an adversarial machine learning setting in which a malicious entity corrupts training data with the aim to deteriorate model accuracy. The behaviour of the model in such settings has classically been studied with regards to the dataset as a whole. In this thesis, we present an empirical analysis of the effects of a backdoor trigger data poisoning attack on a machine learning model from the perspective of the data. We contribute a new metric for the vulnerability of a sample to a backdoor attack, the poison resistance score, and devise a correlation metric that can be used to compare this to existing sample-level scores. We explore the poison resistance results in the context of long tail theory and show that this score is representative of the prototypicality of a sample. We also ask whether backdoor attacks affect different classes of the data distribution differently and discover that in general they do not, but the details of the attack strategy can cause a strong bias towards certain classes. Finally, we briefly touch upon the effect of model architecture on poison resistance, and find the resistance orderings of the samples to be relatively consistent between architectures.

Understanding Backdoor Poisoning Attacks from the Perspective of the Data Distribution by Isha Gupta Supervised by Hidde Lycklama

Sentiment analysis is the field of computer science that studies people's behaviour or attitude towards objects, events, organizations, and products. Organizations make use of this to improve their products. This paper aims to study people's behaviour towards COVID 19. We have taken data from Twitter to analyse the sentiments. The results show that people don't have positive sentiments towards this coronavirus.

Isha Gupta

Papers

Deep Learning Approach to Determine the Impact of Socio Economic Factors on Bitcoin Price Prediction

Real-Time Driver's Drowsiness Monitoring Based on Dynamically Varying Threshold

Identification of relevant features influencing movie reviews using sentiment analysis

Understanding Backdoor Poisoning Attacks from the Perspective of the Data Distribution by Isha Gupta Supervised by Hidde Lycklama

Sentiment Analysis of COVID-19 Tweets