Pinned Repositories
Automation-and-Web-Scraping
A bunch of python scripts to automate mundane stuff like adding files to GitHub repositories, downloading stock prices from NASDAQ index, paying online bills and scraping historical prices of particular stock. Used following python libraries: BeautifulSoup, Mechanize, urllib and subprocess.
CausalMediationAnalysis
Code for the paper "Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Bias"
Election-Forecasting
Used logistic regression on polling data to construct US presidential elections predictions for 2012. Instead of plotting the data, the script makes use of maps and plotting libraries to plot the predictions for intuitive understanding of the reader's real needs.
Game-Programming
A set of small games and animations which are developed using Unity Engine and C#.
Soccer-Player-Analysis
This script employs machine learning methods to train a model for clustering players with similar style of play which makes it easy to perform analysis. Once players are grouped, it is possible to identify which group can coach the other group on those attributes which the other the group lacks. It also computes which factors affect the overall rating of the players and helps the individual player to hone such skills to stay on top of the game. It also makes the job of team coach easier to decide on the position of a new player by comparing his/her graph with that of already clustered group.
Task-Manager
A simple web application designed using HTML, CSS, Flask, Jinja2 and Python for managing day-to-day tasks. Allows users to create new tasks, update and delete the existing tasks.
Twitter-data-analysis
It is a script deploying twitter API connect libraries which can be used for finding out most trending topics, hastags, locality based topics, topic based tweets, movie reviews and many more. It also has frequency distribution of the words in the tweets which can be used to check whether it is potentially sensitive or not. This can further be extended to predicate the exit polls of elections, ratings for movie by deploying a deep learning techniques by analyzing the tweet diction.
Understanding-the-motive-behind-Voting
In 2006, three researchers Allan Gerber, Donald Green and Christopher Larimer carried out a massive field experiment in Michigan, USA to test the hypothesis that one of the reasons people voted was due to social or extrinsic pressure. Here both logistic regression and classification trees were used to analyse the data.
Website-Crawling
A simple python based program which crawls a website with login authentication and automates some form filling processes which needs human intervention. Its uses mechanize, browser, BeautifulSoup and urllib2 modules to enhance form-filling and handling html tags.
social-insights
A search engine to query social media insights with political theme
Anirudh-Muthukumar's Repositories
Anirudh-Muthukumar/Soccer-Player-Analysis
This script employs machine learning methods to train a model for clustering players with similar style of play which makes it easy to perform analysis. Once players are grouped, it is possible to identify which group can coach the other group on those attributes which the other the group lacks. It also computes which factors affect the overall rating of the players and helps the individual player to hone such skills to stay on top of the game. It also makes the job of team coach easier to decide on the position of a new player by comparing his/her graph with that of already clustered group.
Anirudh-Muthukumar/Election-Forecasting
Used logistic regression on polling data to construct US presidential elections predictions for 2012. Instead of plotting the data, the script makes use of maps and plotting libraries to plot the predictions for intuitive understanding of the reader's real needs.
Anirudh-Muthukumar/Game-Programming
A set of small games and animations which are developed using Unity Engine and C#.
Anirudh-Muthukumar/Twitter-data-analysis
It is a script deploying twitter API connect libraries which can be used for finding out most trending topics, hastags, locality based topics, topic based tweets, movie reviews and many more. It also has frequency distribution of the words in the tweets which can be used to check whether it is potentially sensitive or not. This can further be extended to predicate the exit polls of elections, ratings for movie by deploying a deep learning techniques by analyzing the tweet diction.
Anirudh-Muthukumar/Understanding-the-motive-behind-Voting
In 2006, three researchers Allan Gerber, Donald Green and Christopher Larimer carried out a massive field experiment in Michigan, USA to test the hypothesis that one of the reasons people voted was due to social or extrinsic pressure. Here both logistic regression and classification trees were used to analyse the data.
Anirudh-Muthukumar/Automation-and-Web-Scraping
A bunch of python scripts to automate mundane stuff like adding files to GitHub repositories, downloading stock prices from NASDAQ index, paying online bills and scraping historical prices of particular stock. Used following python libraries: BeautifulSoup, Mechanize, urllib and subprocess.
Anirudh-Muthukumar/Lamprey-Detection-Using-CNN
Sea lampreys have become a major pest in the North American Great Lakes. They are considered as an invasive species that prey on many species of commercial value such as lake trouts, fishes and perches. To tackle this problem, we introduce LampNet, a Convolutional Neural Network (CNN) based classifier to detect lampreys. Our evaluations show that LampNet can identify and detect lampreys with an F1-score of 0.97 and an overall ROC-AUC score of 0.995. This framework could be extended to replace the human operating the fish elevator to prevent the lamprey from entering the water bodies.
Anirudh-Muthukumar/Machine-Learning
Octave programs for computing some machine learning methods.
Anirudh-Muthukumar/Stockzer
Investment has become essential for everyone who looks to have a peaceful retired life. While thinking of investments, equity markets have an edge over bonds, commodity, real estate and oil because of the lucrative returns it has yielded over the past few decades. This script is used to pick stocks positive with certain indicators such as stocks coasting 52-week low/high, stocks targeted by smart money and current trend in its volume. Future scope of this project would be to predict the exact entry and exit prices for the stocks, making it a cake walk for the traders and investors.
Anirudh-Muthukumar/Task-Manager
A simple web application designed using HTML, CSS, Flask, Jinja2 and Python for managing day-to-day tasks. Allows users to create new tasks, update and delete the existing tasks.
Anirudh-Muthukumar/CausalMediationAnalysis
Code for the paper "Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Bias"
Anirudh-Muthukumar/Blockchain
Implemented a class framework to create a functional blockchain that combines ideas of Proof-of-Work, transactions, blocks, blochains and hashable merkle tree. The blockchain can be accessed via API calls more information of which can be found in the readme file.
Anirudh-Muthukumar/Causal-Mediation-Analysis
Anirudh-Muthukumar/EOY-Stock-Return-Predictor
A cluster-specific prediction model to predict whether or not stock returns for the month of December will be positive or not, based on the stock returns for the first 11 months of the year. Though this model achieves a modest accuracy of about 57.88%, this is fairly a good recommender for any investor basing his/her decision only on the stock returns for the previous months of the calendar year.
Anirudh-Muthukumar/github-slideshow
A robot powered training repository :robot:
Anirudh-Muthukumar/ios-style-Calculator
ios style Calculator App for Android
Anirudh-Muthukumar/Leetcode-Solutions
Accepted solutions for leetcode problems in python https://skdoosh-leetcode.herokuapp.com
Anirudh-Muthukumar/NLI-for-Facebook-Comments
A Native Language Identifier used to predict the nativity of the person from his/her facebook comments. It is applied for regional languages in India like Hindi, Bengali, Kannada, Malayalam, Tamil and Telugu.
Anirudh-Muthukumar/opencv
Open Source Computer Vision Library
Anirudh-Muthukumar/Python-Code
A gist of all my python work using libraries such as numpy, itertools, math, statistics, datetime, calendar, collections, functools, string and re.
Anirudh-Muthukumar/relate
Performant database access in Scala
Anirudh-Muthukumar/scrapers
Lots and lots of web scrapers
Anirudh-Muthukumar/snorkel
A system for quickly generating training data with weak supervision
Anirudh-Muthukumar/social-insights
A search engine to query social media insights with political theme
Anirudh-Muthukumar/Sudoku-Solver
(http://skdoosh-sudoku-solver.herokuapp.com/) A web application which uses backtracking paradigm to solve the Sudoku problem.
Anirudh-Muthukumar/summer-with-dsc
Summer with DSC UMass is an initiative to get students into open source through mentorship especially for those who lost their internships for summer 2020 due to the pandemic
Anirudh-Muthukumar/Top-20-Backtracking-Algorithms
Top backtracking algorithms using python.
Anirudh-Muthukumar/Top-Dynamic-Programming-Problems
Solutions for common data structure problems that can be solved using dynamic programming.
Anirudh-Muthukumar/Web-App-for-Broadband-Management-System
Anirudh-Muthukumar/Wheel-Balancer
Its protoype of an application which can be used for wheel balancing in automation industry. In the present era, demand for automobiles especially heavy vehicles has been increasing drastically to satisfy the growing needs of market. This application which is built using Jswing can be used for balancing wheels with ease. It is compatible with RaspberryPi 3 with added keyboard functionalities which makes it easier for the technicians to access the application even from a distance.