Illa Pavan Kumar

Bio: Illa Pavan Kumar is an academic researcher from VNR Vignana Jyothi Institute of Engineering and Technology. The author has contributed to research in topics: Data security & Machine learning. The author has an hindex of 2, co-authored 2 publications receiving 10 citations.

Dominant Color Palette Extraction by K-Means Clustering Algorithm and Reconstruction of Image

Abstract: In our present age, there are numerous applications and devices for converting a standard picture into a picture, which is built from its predominant shading palettes. ‘predominant colors’ are the hues in the picture, in which the pixels of the picture contain the particular shading. These hues are then removed and another picture is then framed from this picture. However, this is just a speculation of the issue. In this undertaking, we dive into much more proficient techniques with a specific end goal to remove the prevailing shading palettes from the picture. In this manner, utilizing a grouping calculation and a significantly more ideal instatement technique, effective shading palette is extracted. The grouping calculation being utilized here is K-implies bunching calculations which is much more effective calculation than the other bunching calculations out there. The bunching calculation groups the pixels in bunches in view of their shading. The palettes separated from the picture are then utilized and another picture is reproduced.

Server Security in Cloud Computing Using Block-Chaining Technique

Abstract: Data security is a key concern in cloud computing . With the increase in attempts to hack a virtual cloud server in order to access the data stored there privately, more and more measures are being taken to ensure a reliable and secure cloud. Enhancements are made on traditional cryptography algorithms so that they are more suited to the distributed, virtual and multi-tenant environment of the cloud. Block-chaining is the newest technique in the fields of computing. It is basically a growing list of records, called as “blocks”, which are linked and secured using cryptography. In this project, we are aiming to implement an encryption algorithm which works on the principles of block-chains, to ensure the security and integrity of a cloud-based system. The infallible nature of block-chains is what makes it perfect for cloud server security.

Medical Image Classification for Disease Prediction with the AID of Machine Learning Approach

Abstract: The need and requirements of machine learning are increasing day by day in the field of medical image classification. This research paper illustrates the relevance of machine learning in making successful detection of medical diseases by classifying the medical images in an organised manner. In this context, a brief evolution of machine learning has been provided here. Besides the growth of machine learning in the field of healthcare is presented neatly in this research paper. The literature part clearly describes the previous work in the respective field and the mechanism of machine learning. In parallel to it, a brief description of how the research has been conducted is also presented in this research paper. In this context, the survey conduction among 67 people is mentioned here. Besides analysis and interpretation of the data which are gathered from the survey is present here. The research paper also talks about the flaws of machine learning which have misguided several medical diagnosis processes in the recent past. The dynamics of ML and the relevant results were presented in the discussion and finding parts. At last, a conclusive summarisation has been presented here to provide an overall idea about the image classification process by integrating ML in the modern medical treatment process.

A Recent Survey on Internet of Things (IoT) Communication Protocols

Abstract: Internet of Things (IoT) consists of sensors embed with physical objects that are connected to the Internet and able to establish the communication between them without human intervene applications are industry, transportation, healthcare, robotics, smart agriculture, etc. The communication technology plays a crucial role in IoT to transfer the data from one place to another place through Internet. This paper presents various communication protocols, namely Zigbee, Bluetooth, Near Field Communication (NFC), LoRA, etc. Later, it provides the difference between different communication protocols. Finally, the overall discussion about the communication protocols in IoT.

Storage Optimization Using File Compression Techniques for Big Data

Abstract: The world is surrounded by technology. There are lots of devices everywhere around us. It is impossible to imagine our lives without technology, as we have got dependent on it for most of our work. One of the primary functions for which we use technology or computers especially is to store and transfer data from a host system or network to another one having similar credentials. The restriction in the capacity of computers means that there’s restriction on amount of data which can be stored or has to transport. So, in order to tackle this problem, computer scientists came up with data compression algorithms. A file compression system’s objective is to build an efficient software which can help to reduce the size of user files to smaller bytes so that it can easily be transferred over a slower Internet connection and it takes less space on the disk. Data compression or the diminishing of rate of bit includes encoding data utilizing less number of bits as compared to the first portrayal. Compression can be of two writes lossless and lossy. The first one decreases bits by recognizing and disposing of measurable excesses, and due to this reason, no data is lost or every info is retained. The latter type lessens record estimate by expelling pointless or less vital data. This paper proposed a file compression system for big data as system utility software, and the users would also be able to use it on the desktop and lossless compression takes place in this work.

Loan Default Prediction Using Machine Learning Techniques

Abstract: AbstractLoans are a very fundamental source of any bank’s revenue, so they work tirelessly to make sure that they only give loans to customers who will not default on the monthly payments. They pay a lot of attention to this issue and use various ways to detect and predict the default behaviors of their customers. However, a lot of the time, because of human error, they may fail to see some key information. This paper proposes a better approach using machine learning approaches like KNN, decision tree, SVM and logistic regression to predict defaulters. The accuracy of these methods will also be tested using metrics like log loss, Jaccard similarity coefficient and F1 Score. These metrics are compared to determine the accuracy of prediction. This can help banks conserve their manpower and fiscal resources by reducing the number of steps they have to take in order to check if somebody is eligible for a loan.KeywordsMachine learningLoan predictionBankingCredit risk managementPredictorClassifiersPython

AI-Based Medical Voice Assistant During Covid-19

Abstract: AbstractPresently, healthcare is the most important domain across the globe. Amidst this pandemic, all the doctors are much occupied and extremely busy. Getting proper treatment from doctor is getting difficult due to unavailability of doctors. Moreover, if a person feels uneasy and does not know what exactly the problem is, they really do not know which doctor they should take an appointment with. So, in this chapter intend to provide a web application as an all-in-one solution with an integrated voice assistant, which can help you with the process of disease diagnosis. For those doctors and patients who may need a second brain to make sure the diagnosis is correct, this is a platform with an AI-based medical voice assistant. It can work verbally with the doctor/patient and can assist him/her with the diagnostic process. This voice assistant should be able to select a patient’s illness on a confidence score to support diagnosis operations. Such a software can help both physicians and patients. Depending on the predicted disease/illness, the doctor may give the patient an e-prescription using the help of a voice assistant. The patient can also order medication using a voice assistant. The archives are stored in an orderly fashion, so users do not have to worry about losing them. This application is exclusively made to assist doctors and patients with intention of giving medical care in an interactive, lifesaving and resource saving manner. This application is a minor step in achieving a bigger target of completely digitalizing our medical care..KeywordsAICovid-19PredictionNLPEMR’s