Harish B's repositories
COVID-19-Mutation-Rate-Analysis-using-Recurrent-Neural-Network-based-LSTM-model
The new coronavirus SARS-CoV-2, also known as COVID-19, has caused a global pandemic. This contagious RNA virus threatened the entire world. More than 7.9 million people have been infected as of June 15, 2021 with 432k deaths. This virus is capable of causing mutations in humans. Understanding the evolution of this virus and determining the risk of the emerging infectious disease requires accurate mutation rate determination. This research investigates the mutation rate of the entire genome sequence based on patient data. The obtained data is analysed independently to detect nucleotide and codon mutations. Furthermore, the mutation rate analysis shows that a large number of Thymine (T) and Adenine (A) nucleotides are mutated when compared to other nucleotides, although the mutation rate of codons is not frequent when compared to nucleotides. A recurrent neural network (RNN) based Long Short-Term Memory (LSTM) model was used to predict the future mutation rate. This model gives very less error it is around 1.03 in both the training and testing sets. The mutation of the 83rd patient has been predicted by this model.
Sleep-Apnea-Detection
Sleep Apnea is a serious sleep disorder in which breathing starts and stops repeatedly during sleep. If this condition is left untreated, it can lead to serious health issues. This project proposes a robust method for detecting sleep apnea and keeping track of the region of interest automatically. The methodology involves converting the captured video into grayscale images, and then apply Gaussian blur and Canny edge detection to detect noise-free edges of the body. Further, the current frame is compared to the previous frame to detect body movements accurately. The proposed method is experimented with different testing conditions to demonstrate its efficiency and accuracy in identifying sleep apnea. This project also introduces a novel algorithm for automatically adjusting the region of interest within the frame. Furthermore, the advantages and limitations of the proposed algorithm are also been discussed.
Anthropomorphic-arm-with-Spherical-Wrist
Robot manipulators are commonly used in most of today's industrial applications. This usage only continues to grow as the technology improves, and as such, increases in relevancy as an area of research. In this report an anthropomorphic robot arm with spherical wrist was simulated in MATLAB. Through utilizing the Denavit-Hartenberg method along with forward and inverse kinematics, the robot’s end effector is able to reach a specified point in space. In today's industrial applications, robot manipulators are extensively employed. The most prevalent form of industrial robot is the anthropomorphic arm with spherical wrist robot. The anthropomorphic arm along with a spherical wrist and can do 'human-like' actions. The ability to operate the robot's shoulder, elbow, and wrist joints is necessary to accomplish these human qualities. As a result, the robotic arm is imbued with industrial expertise, allowing it to do tasks like "pick-and-place." The forward and inverse kinematics have been calculated, and mathematical solutions based on the D-H (Denavit–Hartenberg) parameters have been discussed in this report. The kinematics solution has been verified by solving the manipulator’s motion. Therefore, this report presents an efficient analytical computation procedure of both forward and inverse kinematics, and moreover we have done the simulation of the robot with MATLAB and robotics toolbox.
Decentralized-Finance-Lending-Application
We need to create an application to lend and borrow cryptocurrencies(Neo/Ether/BTC,etc.) The three important entities in this application are the Lender, Borrower and DeFi lending platform. The lender can deposit money in the platform(which will be some smart contracts) similar to how people deposit money in their banks. The Borrower can deposit crypto assets as collateral and obtain cryptocurrency loans with a specified loan maturity period and interest rate. The borrower first deposits crypto collateral. He receives a specified amount of loan after successful deposit of collateral He makes periodic payments to the platform ,which in turn pay the amount to the lender similar to how banks work. After successful payment of the loan , the borrower gets back his collateral and the lender receives his investment amount with a certain interest.
AUTO-COMPLETE-FEATURE-FOR-DATA-STRUCTURE-RELATED-TERMS
This project is an attempt to implement the autocomplete feature for data structure-related terms using the trie data structure. Trie is an efficient information retrieval data structure. Trie data structure helps in bringing down the search complexities to an optimal level, which makes it more efficient for data retrieval. The program code is done in the Java programming language. We have also implemented a GUI with the help of the swing framework, to enhance the user interaction with the program. Apart from auto-suggestions display in the GUI window, the user has the functionality to insert and delete terms. Also, the information about the terms entered by the user will be displayed in the GUI, which has been implemented with the help of the JSoup library for web scrapping.
AutoScope
A hackathon project for detecting damages in vehicles using deep learning and computer vision
Credit_Card_FDS
Credit card FDS using Hidden Markov Model
Defect-Recognition-to-Multi-Camera-Multi-object-Tracking
Solution for the given Problem Statements
Image_Object-and-Text_Detection
Object and Text Detection using Flask
Jinglebell
An AI Assistant which performs the task efficiently on daily basis by voice command.
LearnCrafter
Education Website
Leetcode_Interview_Problems
Study Preparation Kit
MBTI-Indicator-Based-Personality-Type-Prediction
This is a personality prediction project based on MBTI personality type indicator. The web app is available here: https://personality-prediction-t6.herokuapp.com
Metabolic-Pathway-Class-Prediction
In this project we are predicting the metabolic pathway class for a given query compound. The web app can be found at https://metabolic-pathway-prediction.herokuapp.com/
OpenAI-Gym-CartPole-Env-using-DQN
Gym is a toolkit for developing and comparing reinforcement learning algorithms. It supports teaching agents everything from walking to playing games like Pong or Pinball.
Page-Rank-Algorithm
The importance of a website is an inherently subjective issue that depends on the interests, knowledge and attitude of the reader. However, much can still be said objectively about the relative importance of websites. This article describes PageRank, a method of objectively and mechanically evaluating web pages that effectively measures people's interest and attention to them. We compare PageRank to an idealized, random internet user. We show you how to efficiently calculate PageRank for a large number of pages. We show you how to apply PageRank to user search and navigation.
Python-chat-application-using-Socket-Programming
GUI based chat application using Socket programming in python language
QuickJobs
A job portal using twitter streaming API
Retroactive-Data-Structures
Data Structures
TimeTangle
TimeTangle
Turtle-Text
AI Summarizer
Twitter_API_Job_Portal
Quick Jobs an simple job portal application for real time update of latest jobs with Twitter API.