Project | 01
Project | QuickDraw
This project was based on the Google's 'Quick, Draw!' Dataset. We built an LSTM & CNN Classifier which was trained on a dataset of 15,000 images of 4 different classes of objects. The LSTM Classifier predicts the doodle that one is trying to draw as they draw it, identifies strokes. The CNN Classifier was added to confirm the given output from the LSTM Classifier to increase accuracy. Click here for project link.
Project | 02
Project | Twitter Sentiment Analyzer
This is an implementation of Opinion Mining, ie. fetching tweets from your Twitter Account using the official Twitter API Tweepy.
It uses TextBlob library that helps us find the polarity of a sentence by tokenizing it and applying Natural Language Techniques. Click here for the project link.
Project | 03
Project | Data Analysis and Transfer using HADOOP
This project is based on setting up the 3-Node(NameNode and 2 DataNodes) cluster using Ambari, SSH-ing into each node without passwords, sending and receiving data, using HIVE, HBASE, SQOOP to operate data interpretation/analysis on a book from the online Gutenberg library.