Implemented the insert and search module of R-Tree for various dimensions of data and report a comparative analysis. The implementation follows the algorithm proposed in the paper "R-Tree: A Dynamic Indexing Structure for Spatial Searching"
Implemented k-Nearest Neighbour search algorithm for R-Trees. It first creates an R-Tree for a set of points, and then uses the algorithm proposed in the paper "Nearest Neighbor Queries" to retrieve k nearest neighbours.
Implemented an Image and Content based search engine for a news filtering. Scrapped news articles from NDTV and Indian Express websites and generated tf-idf vectors for the texts. Images, represented as histograms were stored using R-Tree indexing scheme. Given a query keyword, query image and k, the search engine filters the documents having the query keyword and from the images of the filtered articles, k-Nearest Neighbour search algorithm of Assignment 2 was used to search the images and display them.
Presentation and Analysis on the paper "A critique of ANSI SQL isolation levels"
Comparison of the query time for various database engines. Comapred MySql Database with innodbandcolumnstore` engines, MariaDB and Monet DB. Created a Fact table and 3 Dimension tables, inserted identical values in them and then compared the query time for 5 queries.
Implemented several queries for MongoDB using mongoshell and pymongo to insert and return the records in a collection. We have used a json file of twitter records as the database entries.
Implemented a simple Blockchain to calculate the SGPA of a student for a particular semester. Used truffle for compiling and Ganache as the personal Ethereum Blockchain. Also implemented a javascript file using web3js to interact with Ganache.