rish332 Goto Github PK

analyzing-us-accidents-data-using-pyspark-and-databricks

I have analysed the accidents dataset related to United States of America. To handle the dataset requirements of million+ records, I decided to analyse the dataset using DataBricks which is a cloud-based tool that makes use of PySpark to process the data.

fetch-analysis-challenge

identifying-risky-bank-loans-using-c5.0-decision-trees

Illustration of C5.0 decision trees algorithm which is used to identify risky bank loans. C5.0 is used because of the benefits it provides which is, that it is quite opinionated towards pruning.

identifying-risky-bank-loans-using-random-forest

Illustration of Random Forest model on credits data for identifying risky bank loans. Random forest combines the decision trees and train them on different set of observations. By using this I will compare this with the C5.0 algorithm and determine which one is the best among them depending on the accuracy.

predictive-analysis

Performed predictive analysis using linear and logistic regression where I compared which algorithm worked best. Later, I also managed to perform boosting algorithm to improve the efficiency.

real-estate-price-prediction-using-python

A real estate company wants to predict the prices for the future based on the data available. This way it will help them adjust their budget, change the way they work so as to earn profit and also help them decide where to invest in. As a Data Scientist, I will explore the data available, perform preprocessing, check for manipulation and finally apply and evaluate machine learning algorithms to see which algorithm provides the best result.

sentiment-analysis-using-n-gram-and-naive-bayes

The focus is on the sentiment analysis, and by using the ratings 0 or 1 with negative or positive sentiment. GOAL - Finally, would able to conclude which movie, product or a restaurant is good based on the sentiments. Additionally, doing the n-gram analysis while trying to answer “Whether n-gram analysis is important in every text analysis?”

sql-project

Performing normalisation and analysing the different datasets to gather insights for decision making

text-to-speech

web-scrapping-using-python

Basic template to scrape the web using python.