post-image

BitKhan: Creating my own cryptocurrency with Python

Created a cryptocurrency using the Blockchain Technology with just Python and a few libraries. Built a working model to transfer the currency from one node to another on the decentralized network. The transactions are tamper-proof and uses the same sha256 hash encryption that Bitcoin uses.

Read More
post-image

Speech Recognition with Neural Network

Built an algorithm that understands simple spoken commands from the English language with Convolutional Neural Network. Preprocessing is done to convert the audio files to spectogram to make the training easier for the neural network. Over 60,000 samples were used to train 30 words for an accuracy of 85%.

Read More
post-image

Can you predict Bitcoin market crash with Twitter?

Built a Dash Web App that tracks tweets, analysis its sentiment and plots them on a graph for visual representation. Everything is done live and the graph updates after each incoming tweet. Was able to find a pattern in user sentiments, which could be used as a potential indicator for dramatic market crashes.

Read More
post-image

How the world reacted to Samsung Galaxy S9

Sentiment Analysis on the new Samsung Galaxy S9 YouTube ad by studying the comments to understand the different aspects people are eager to see on Samsung's flagship smartphone. The sentiments for these aspects are rated on a scale of 1-5 (extremely unhappy - extremely happy).

Read More
post-image

'Friends' chatbot with Deep Natural Language Processing

Implemented Deep Natural Language Processing seq2seq model(RNN) in Tensorflow and Python to develop a chatbot trained with conversations between the characters of the Friends TV show. With minimum available data(and from a tv show), this is not the best chatbot ever built but this concept is applicable to many modern day chatbots.

Read More
post-image

Convolutional neural network with The Simpsons

Convolutional Neural Network(CNN) is a type of neural network especially useful for image classification tasks. I applied CNN on thousands of Simpsons images training the classifier to recognise 10 characters from the TV show with an accuracy of more than 90 percent.

Read More
post-image

Unsupervised Machine Learning Simulator - Download

An easy-to-use windows application to simulate the two major Clustering algorithms for unsupervised machine learning, KMeans and Hierarchical Clustering. I'll keep updating versions to make the application more intelligent and usable with different kinds of datasets.

Read More
post-image

Data Scientist vs Data Analyst vs Data Engineer using Word Cloud - Python

The terms Data Scientist, Data Analyst and Data Engineer are often used interchangeably. Although all three are data focused roles, they have subtle differences that separate each other. With even the hiring companies switching between the terms we'll take a different approach understanding each role.

Read More
post-image

Supervised Machine Learning Simulator - Download

An easy-to-use windows application to simulate major Machine Learning algorithms. Right now, it supports very basic algorithms at their default settings and I'll keep updating versions to make it more intelligent and usable with different kinds of datasets.

Read More
post-image

Crofters: The Good, the Bad and the Ugly

This blog discusses my experience as a young startup founder, forming the team, the product cycle, the high points and of course, the fall. We started with three co-founders with the goal of growing fresh greens and veggies inside living rooms. After a turbulent journey of close to two years, we had to shut the company down.

Read More
post-image

Buddy: My WhatsApp Personal Assistant with Python

This is a different kind of chat bot built with Selenium. Selenium is a tool for automating web browsers. Selenium is capable of doing anything that you would be able to do on a browser. This script simulates what you'd normally do to message your friend on WhatsApp.

Read More
post-image

Extract Facebook and Twitter data from any page

You may need data from social media like Facebook and Twitter for a variety of reasons. I for one use it for statistical analysis - to get the reactions on posts from a certain page and make it into a spreadsheet for easy analysis. To be able to extract publicly available data using a python code, you need to register as a developer and then get your app's access tokens.

Read More
post-image

Face Detection with Darknet YOLO

You only look once (YOLO) is a state-of-the-art, real-time object detection system. It comes with a few pre-trained classifiers but I decided to train with my own data to know how well it's made, the potential of Image Recognition in general and its application in real-life situations.

Read More
post-image

Analysis of water situation in chennai - Web Scraping

This post is inspired from a recent forward on social media about rain and water situation in Chennai. This also triggered the curiosity of data monks (a secret group) in Report Bee with questions like?…

  • Where does Chennai get its water? What is its storage capacity?
  • How dependent is Chennai on Monsoons?.....
Read More