Data from Twitter is crawled using the Tweepy library, and is stored in the CSV file, Collected and Indexed over 2 GB of tweet data using big data technology such as Hadoop MapReduce, and Lucene to Index data that is later used in spring for building user Interface and used tf-idf page rank algorithm, Built an interactive web interface(Similar to google search engine) using angular+ spring with the simple search box, When the user gives a query it will retrieve the tweet data and display results in the dynamic web page.

The project aims at discovering a pattern and then providing a prediction of how students will perform based on certain parameters. Precisely, the idea is to assess a student's performance based on estimating the proportion of grades scored from their psychographic prediction if a student will pass or fail given a vector of attribute values for a student. We have used supervised learning algorithms like Random Forest, Decision Trees, Support Vector Machines, Gaussian Naive Bayes, and Logistic Regression. worked to pre-process data by implementing data mining techniques

This Project explores Uniform Cost Search, and the A Star algorithm with Misplaced Tile Heuristic and the Manhattan Distance Heuristic. There were puzzles of varying difficulty given to implement. The easiest one being the goal state itself and the hardest one being impossible to solve. However, not much difference was found for the easier puzzles. But as the difficulty level increased, the heuristics performed better with respect to time and space complexity. It was found out from the comparison of the three algorithms that A Star with Manhattan Distance Heuristic performed the best, followed by Misplaced Tile Heuristic, followed by Uniform Cost Search. Though both the heuristics performed better than Uniform Cost Search(heuristic hardcoded to 0), the Manhattan Distance Heuristic performed the better. The language used to develop the algorithms

Used a dataset of nearly 28,500 credit card transactions and multiple unsupervised anomaly detection algorithms-Local Outlier factor, Isolation Forest Identified transactions with a high probability of being credit card fraud.Furthermore, Using metrics such as precision, recall, and F1 scores we will investigate why the classification accuracy for this algorithm can be misleading

• Designed and Developed a fully functional grocery Ecom site using Angular and bootstrap for the front end, Users can register, login/log out, add items to a cart and place an order and it is connected to google cloud's firebase and it can be hosted in Google's firebase for visibility. All the user data is stored in the Fire store database and deployed webpage in google firebase for real-time use.