Intelligent Computing and Networking

In the last decade, self-driving cars have witnessed a meteoric rise in popularity due to exceptional research in the fields of Edge Computing and Artificial Intelligence. Nowadays, autonomous vehicles use elaborate mathematical models in tandem with sophisticated Deep Learning techniques to navigate safely. PID Controllers have been used ubiquitously by researchers for autonomous vehicles. Deep Learning techniques like YOLO allow autonomous vehicles to be able to detect a wide range of objects in their surroundings leading to better responses. In this paper, a PID controller has been implemented to navigate a vehicle in CARLA Simulator. A Custom Traffic Light detection model has also been integrated with the controller to respond to traffic lights in the path of the vehicle.

High dimensional data for classification does create many difficulties for machine learning algorithms. The generalization can be done using ensemble learning methods such as bagging based supervised nonparametric random forest algorithm. In this paper we solve the problem of binary classification for high dimensional data using random forest for polycystic ovary syndrome dataset. We have performed the implementation and provided a detailed visualization of the data for general inference. The training accuracy that we have achieved is 95.6% and validation accuracy over 91.74% respectively.

The polycystic ovary syndrome diagnosis is a problem that can be leveraged using prognostication based learning procedures. Many implementations of PCOS can be seen with Machine Learning but the algorithms have certain limitations in utilizing the processing power graphical processing units. The simple machine learning algorithms can be improved with advanced frameworks using Deep Learning. The Linear Discriminant Analysis is a linear dimensionality reduction algorithm for classification that can be boosted in terms of performance using deep learning with Deep LDA, a transformed version of the traditional LDA. In this result oriented paper we present the Deep LDA implementation with a variation for prognostication of PCOS.

This report is about detecting motorbike riders without a helmet and also the pillion rider with the use of YOLO object detection algorithm. We introduced the updated approach for helmet detection. This approach is an upgradation of YOLO object detection algorithm which detects not only the rider’s helmet but also the helmet of the pillion rider. Primary objective is Detection of helmet of rider and pillion rider in that targeted image and Increase the accuracy of the YOLOv5 algorithm by adding one layer for detection of small details in an image. In this proposed model a new layer has been added for detection of smaller objects having smaller features. This has been done by changing the configuration of YOLOv5 architecture. The helmet detection using this proposed model has been carried out for a dataset containing images with maximum 3 people, with no helmets, 1 helmet each or 2 wearing it.

The Stock Market Prediction and Analysis has always been one of the most challenging tasks (Polamuri and Mohan in A survey on stock market prediction using machine learning techniques, 2019; Parmar et al. in First international conference on secure cyber computing and communication (ICSCCC), pp. 574–576, 2018). The variety of influences and unpredictability beats even the heavyweights to ground when it comes to successfully analyzing Stock Price data. In the proposed System, we have designed and successfully built a Machine Learning model using Long-Short Term Memory (LSTM) algorithm which helps for prediction of stock price data. We have done experimentations for better training, accuracy and results, on used data. The proposed system is also deployed on a web application which helps eliminate/reduce the difficulty of its use for the users. The model also works on the real-time data as we are using Yahoo finance API for getting updated data for model training and prediction. Lastly, The Indian stock market prices are also heavily driven by public sentiments which have for providing a better public opinion upon a particular stock. To help our users tackle this, we have added twitter sentiment analysis as a feature which provides us results in term of percentages of positive and negative sentiments within the tweets in the public domain at present about a particular stock, achieving a better opinion on a particular stock for the users. The resulting model successfully gives us a prediction graphs as an output when given a particular stock on the proposed web application. We obtained least error in prediction, for Asian Paints data for the split of 80:20, using 75 epochs.

In the Internet of Things (IoT) paradigm, boundless solutions have been designed and implemented to do effective and secure communication among it’s smart objects and it’s network. The outcome of effective and secure communication always relies on which IoT protocol has been used at the application layer. Generally IoT devices communicates using various IoT push protocols such as XMPP (Extensible Messaging and Presence Protocol), MQTT (Message Queuing Telemetry Transport), AMQP (Advanced Message Queuing Protocol) among which MQTT protocol is widely used protocol within IoT platform because it requires nominal resources as it’s lightweight and efficient, it also support bi-directional communication among smart objects and cloud and MQTT also guarantees and support reliable message delivery through 3 Quality of Service (QoS) levels. This research paper focuses on key concepts on MQTT protocol architecture, basic security fundamentals such as identity, authentication, authorization and MQTT advance security fundamentals which includes X.509 client certification authentication, OAuth 2.0 and payload encryption.

Social networking sites have given users unprecedented opportunities for the generation and dissemination of content. Block chain Technology as defined the decentralized system for distributed registers which are used to record data transactions on multiple computers. So a variety of social networking sites exist for different purposes, to afford users a range of anonymous and non-anonymous options for self-expression, and the ability to be a part of a virtual community. Sometimes a misinformation, propagated by users and group can create chaos or in some cases, might leads to cases of riots. Therefore, a robust and new system is required to check the information authenticity within the network, to stop the propagation of misinformation. In this paper, propose of block chain based framework is for sharing the information securely at the peer level. In the block chain model, a chain is created by combining blocks of information. I analyze real data by exploiting one of the most well-known DApps sites (decentralized applications), and also compare current technologies in order to get better algorithm or tool to secure our information. such as Facebook.

Emotions can be challenging to describe and interpret, which is why music has been proposed as an art. In recent times, music can be used as a mood regulation mode, to assist someone balance, understand and deal with their emotions better. ‘Euphonia’ is intended at easing that process. The purpose of ‘Euphonia’ is to use real-time facial recognition to acquaint the machine with abilities to recognize and examine human emotions. With this, the machine will be trained to provide the user with suitable songs for that particular mood. Besides this, the machine will also recommend the user with a general playlist pertaining to the user’s likes and dislikes which they can access whenever they wish to. Machine learning concepts and the available datasets have been utilized to classify a vast set of music that is stored using automatic music content analyses. It was implemented using Python, Pandas, OpenCV, and NumPy.

The amount of data generated in today's environment is increasing at an exponential rate. These data are structured, semi-structured, or unstructured in some cases. Processing vast amounts of data is a difficult task. MapReduce is a technique for processing large amounts of data. For storing big data sets, Hadoop data file system (HDFS) is employed. A MapReduce workload is made up of a number of jobs, each of which has multiple map tasks and numerous reduce tasks. In this paper, we propose the Shortest Task Ordering (SJA) algorithm for optimising Mkspan (Makespan) and TCTime (Total completion time) for MapReduce Workloads using dynamic job ordering and slot design. We conducted a comparison analysis on data sets of various sizes. By comparing objective measurements such as Mkspan and TCTime, the experimental results show that performance in terms of speed has improved. Each set of jobs, such as WordCount, CharCount, LineCount, and Anagram, displays comparative improvement in our work. When compared to previous algorithms such as MkJR and MkTctJR, the results demonstrate an improvement in time efficiency, slot usage, and execution speed. In terms of Mkspan and TCTime, the Shortest Task Assigned (SJA) method for job ordering yielded results that were up to 95% better than MkJR. In comparison to the previous algorithms MkSfJR and MkTctSfJR, there is also an increase in slot usage. In terms of Mkspan and TCTime, the Shortest Job Assigned (SJA) algorithm achieved results that were up to 150 percent better than MkSfJR.

Many researchers have recently been inspired by the success of deep learning algorithms in the field of artificial intelligence to improve plant disease detection performance. Deep learning's main goal is to teach computers how to solve real-world problems using data or experience. Detecting diseases is a critical task for farmers. They take shortcuts such as using chemical pesticides, which have negative effects on consumable foods. So, in this paper, we used deep learning algorithms to detect plant diseases. Deep learning is a popular trend in which technological benefits can be imparted to the agricultural field. Detecting plant diseases with deep learning techniques is less expensive than using chemical pesticides. This paper reviews existing techniques and recommends the best technique that farmers can use to identify disease faster and at a lower cost.

The process of establishing democracy in a country is defined by election. Elections might just be a significant event in today’s democracy, however many segments of population throughout the world lack faith in their electoral system, which is a major source of concern for democracy. Even the world’s most powerful democracies, such as the Republic of India, the United States, and Japan, have a flawed legal system. To eliminate these drawbacks of the electoral system, Blockchain Technology is considered as an ultimate solution. With the present surge in sales and use of blockchain technology for a number of purposes, including banking, medical, and identity, a lot of attention has been focused on the legal concerns rather than its practical uses in administration. In this paper, we discuss the concept of Blockchain in a detailed manner making the reader understand the working of blockchain, its characteristics etc. everything from the scratch, as well as how this concept can be implemented as an efficient solution for public voting and how it is more beneficial from the traditional voting methods, with the goal of eliminating the drawbacks of India’s current electoral system while also providing a better, more trustable, safe, and transparent means of public governance. Blockchain is really an emerging technology that promises to improve the resiliency of electronic voting systems. This method offers a way to profit from blockchain’s advantages, such as cryptological foundations and transparency, to achieve an efficient theme for the e-voting system.

Visually handicapped people’s lives are subject to a multitude of unrelenting challenges because they’ve been made bereft of the gift of sight. The proposed solution is a wearable Smart Voice Assistant that is developed to accommodate the needs of the visually impaired to aid them in every aspect of their everyday lives. It takes advantage of recent breakthroughs in the fields of language processing and computer vision to provide a broad spectrum of applications, including emergency response functionality, object recognition, and optical character recognition. It comprises hardware components that provide feedback in the form of sound, haptics, and speech to help with obstacle avoidance. The voice assistant also interacts with a smartphone application to enhance the user’s experience by enabling them to read the messages from their phone, send an SOS message to their closest connections in an emergency, customize the device settings through the mobile application, and find the device with the press of a button if it is misplaced. The proposed solution will enable the user to live a life in relative safety and comfort, which is essential for people suffering from varying levels of visual impairment.

Python libraries are a collection of essential functions that eliminate the need for users to develop code from scratch. Python is a plethora of libraries that serve a range of purposes and it has become a necessity to have sound knowledge of the best ones. Human and machine data production greatly outpaces humans’ ability to absorb, assess, and make complicated decisions based on that data. AI (Artificial Intelligence) is the foundation of all computer learning and the future of all intricate decision making. These technologies are being looked upon as tools and techniques to make this world a better place. It’s application ranges from various fields like healthcare, finance, transport, manufacturing, fraud detection and so on which evidently depicts its potential to transform the future. This paper intends to well verse the readers with the top libraries used to implement concepts of Artificial Intelligence like Machine Learning, Data Science, Deep Learning, Data Visualization and so on. It provides meticulous and unambiguous details about the essential building blocks necessary to execute and perform such ideas. It also includes a comparative analysis of various libraries to provide a detailed understanding and overview of them.

This paper presents a study of Indian rainfall and prediction of the annual rainfall in the state of Maharashtra and Konkan. The decreasing trends in seasonal rainfall and post-monsoon rainfall and increasing occurrence of the deficit rainfall years indicates the probable intensification of water scarcity in many states and sub divisions of India. Rainfall serves a major source of water only when it is conserved, thus a proper analysis and estimation of rainfall globally is of utmost importance. In an agricultural country like India, where the majority of agriculture is rain dependent rainfall prediction can help to understand the uncertainty in rainfall pattern which may affect the overall agricultural produce. The present study consists a descriptive analysis of annual rainfall in India from 1950–2020, this visualization may prove helpful for deciding the right model for prediction. This study is aimed at finding the most apt model for making accurate prediction for the rainfall dataset. Two machine learning model and a neural network model are implemented and their results are compared. The performance of the results was measured with MSE (mean squared error), RMSE (root mean square error), MAE (mean absolute error). The machine learning models showed high level of deviation as the time series data in use was highly inconsistent, on the other hand the neural network showed better efficiency due to the local dependency in the model which helps it to learn and perform better.

Medical care is vital to having a decent existence. Be that as it may, it is undeniably challenging to get an appointment with a specialist for each medical issue and due to the current global pandemic in the form of Coronavirus, the healthcare industry is under immense pressure to meet the ends of patients’ needs. Doctors and nurses are working relentlessly to treat and help the patients in the best possible way and still, they face problems in terms of time management, technical resources, healthcare infrastructure, support staff as well as healthcare personnel. To resolve this problem, we have made a chatbot utilizing Artificial Intelligence (AI) that can analyze the illness and give fundamental insights regarding the infection by looking at the data of a patient who was previously counselled at a health specialist This will also assist in lessening the medical services costs. The chatbot is a product application intended to recreate discussions with human clients through intuitive and customized content. It is in many cases portrayed as the most moving and promising articulations of communication among people and machines utilizing Artificial Intelligence and Natural Language Processing (NLP). The chatbot stores the information in the data set to recognize the sentence and pursue an inquiry choice and answer the corresponding inquiry. Through this paper, we aim to create a fully functional chatbot that will help the patients/users to know about the disease by simply entering the symptoms they possess. Additionally, they can also get information about certain medicine by simply typing the name of the medicine. Another additional feature is the ability of the bot to answer general questions regarding healthcare and wellbeing.

Over the course of this article, a simple machine learning model for the prediction of a football match winner will be discussed. An in-depth analysis and insight of this model is presented further below. And we would go about the process of building it, the relevance of the project will also be mentioned along with its business implications. Finally, the merits and flaws of the project will be discussed along with ways in which it can be improved in future.

According to the WHO i.e., World Health Organization, a target of 10–20 donors per 1,000 persons in any country is required to ensure adequate blood supplies. Traditionally, it is identified and observed that whenever a person has a requirement of blood, they either approach a blood bank or a blood donor with the same blood group. However, it becomes difficult to find a suitable blood donor during the time of emergency requirement of blood. Moreover, availability of the suitable blood group is not guaranteed even in a blood bank. We aim to propose an app that connects the recipient of the blood to its donor in the time of crisis and provides the flexibility of finding the blood banks near them based upon their location. The app can help to increase the possibility for a patient to get a blood donor as the requestor will be connected to all eligible donors sharing the same blood group in the same city. Thus, providing an expanded search space to the person who is in the need of blood. The application makes sure that the important crucial information of the registered users is kept private and confidential before the confirmation from both parties. The application can also be used by organizations such as blood banks or non-profit service organizations that aim to search for blood donors for their blood donation camps and create awareness to a broader mass by creating digital campaigns for their blood donation drives.

As we know, Red Blood Cells are the main part of blood that is responsible for the circulation of blood in the human body. Anemia is a well-known disease that is caused due to the deficiency of healthy red blood cells. Due to anemia, red blood cells are unable to supply oxygen throughout the body. This sickness can be lethal to the human body if not treated promptly. We are using machine learning techniques such as Random Forest, SVM, and others to detect anemia in a patient in this study. We can detect anemia in a patient using machine learning methods. As a result, we intend to create a classification-based ML model in which we provide the essential CBC test values for our model to predict whether a patient is anemic. With the help of machine learning techniques, we are automating the process for detecting anemia in this study work. We compared the statistical analysis of all algorithms we've utilized to predict anemia in this paper.