PORTFOLIO

Project - Analysis of Uber Drive (Nov 20)

The project was based on the trips made by Uber drivers. Different aspects of the trip were analyzed by using different functions in Python.

Skills and Tools

Python Functions, Data Interpretation

Project - Statistical & Probabilistic Analysis of Store Sales, University Survey, & Manufacturing data (Dec 20)

The project involved drawing inferences from 3 datasets, namely - Wholesale Customer Data (Store Sales), University Survey Data & Manufacturing Shingles Data. The concepts of various measures of Descriptive Statistics, Probability and Probability Distributions and various measures of Estimation & Hypothesis Testing are used to analyze these datasets.

Skills and Tools

Descriptive Statistics, Probability & Probability Distributions, Estimation, Hypothesis Testing

Project - Drug Analysis using ANOVA and Principal Component Analysis on College Admissions Data (Jan 21)

The project involved drawing inferences from 2 datasets, namely - Hay Fever Drug Analysis, College Admissions Data. The concepts of Exploratory Data Analysis, Analysis of Variance, and Principal Component Analysis were used to draw inferences from these datasets.

Skills and Tools

ANOVA, PCA, EDA

Project - Bank Customer Segmentation and Insurance Claim Prediction (Mar 21)

The project involved drawing inferences from 2 datasets, namely - Bank Marketing & Insurance. The concepts of Clustering, CART, Random Forest, Artificial Neural Network were used to draw inferences from these datasets. Various performance metrics were used to validate the performance of predictions on Test & Train sets.

Skills and Tools

Clustering, CART, Random Forest, Artificial Neural Networks

Project - Gems Price Prediction & Holiday Package Prediction (Apr 21)

This project was based on 2 datasets : Gems Price Prediction and Holiday Package prediction. In the first dataset, linear regression was applied and the price of gems based on multiple variables was predicted to help company maximize profits. In the second case, logistic regression and linear discriminant analysis were used to predict if the customer would purchase the holiday package to target the relevant customer base.

Skills and Tools

Linear Regression, Logistic Regression, Linear discriminant Analysis

Election Exit Poll Prediction and U.S.A Presidential Speech Analysis using Machine Learning (May 21)

Visible to public

Course Machine Learning

This project is based on 2 case-studies: Vote Prediction and Text Analysis. The first project is to predict which party a citizen is going to vote for on the basis of their age and according to the answers given by the citizens to the questions asked in a survey conducted. The second project is based on the analysis of the inaugural U.S.A. Presidential speeches. One has to draw inferences based on the analysis done on these speeches.

Skills and Tools

Text Mining Analytics, Support Vector Machine - K Nearest Neighbor - Naïve Bayes, Ensemble Techniques, Logistic Regression - Linear Discriminant Analysis

Built a model to Forecast monthly sales of Wine for certain Wine Estate for the next 12 month (Jun 21)

Visible to public

Course Time Series Forecasting

Analyzed historical monthly sales data of a company. Created multiple forecast models for two different products of a particular Wine Estate and recommended the optimum forecasting model to predict monthly sales for the next 12 months along with appropriate lower and upper confidence limits

Skills and Tools

Exploratory Data Analysis for Time Series Data, Exponential Smoothing Models, ARIMA/SARIMA Models, Moving Average Models

Visualizing Insurance Claims using Tableau (Jul 21)

Visible to public

Course Data Visualization using TABLEAU

This project explored the art of problem-solving with the aid of visual analytics. Tableau’s data visualization tools were used to create interactive dashboards to provide high-level insights to the CEO of an Insurance company to drive the company's policymaking

Skills and Tools

Business Intelligence, Tableau, Dashboard Designing

Online retail Orders Analysis (Aug 21)

Visible to public

Course SQL

This project is based on the order management functionality of an online retail store in which you are provided with the “orders” database and you are asked some queries related to it. Answers to these queries will help the company in making data-driven decisions that will impact the overall growth of the online retail store.

Skills and Tools

Joins, Sub Queries, SQL-clauses-statements-conditions, SQLite using DB Browser and MySQL Workbench

Understanding Customers' Buying Patterns for an Automobile Parts Manufacturer (Aug 21)

Visible to public

Course Marketing & Retail Analytics

This project aims to find the underlying buying patterns of the customers of an automobile part manufacturer based on the past 3 years of the Company's transaction data and hence recommend customized marketing strategies for different segments of customers.

Skills and Tools

RFM, Exploratory Data Analysis, Python, KNIME

Recommending ways to increase revenue of a Grocery Store (Aug 21)

Visible to public

Course Marketing & Retail Analytics

The project involves conducting a thorough analysis of Point of Sale (POS) Data for providing recommendations through which a grocery store can increase its revenue by coming up with attractive combo & discount offers for customers.

Skills and Tools

Market Basket Analysis, Exploratory Data Analysis, KNIME, Python

Credit Risk Default Model (Sep 21)

Visible to public

Course Finance and Risk Analytics

The project involved developing a credit risk default model on Indian companies using the performance data of several companies to predict whether a company is going to default on upcoming loan payments.

Skills and Tools

Credit Risk, Loan Default, Finance

Recommending an ideal portfolio considering stocks of large-cap industries (Sep 21)

Visible to public

Course Finance and Risk Analytics

Building a portfolio by analyzing stocks of several large-cap industries from different industry verticles and selecting stocks based upon the risk and return associated with them.

Skills and Tools

Market Risk, Portfolio Optimization, Financial Risk Analytics

Capstone - Social Media Analytics Project to identify likely customers based on digital and social behaviour on social media. (Nov 21)

Domain

An aviation company that sells trips to the customers wants to apply a targeted marketing approach instead of reaching out to each of the customers individually to cut marketing costs. This targeted approach helps in reaching the previously defined audience with a well-established propensity to taking up the offer. The company collaborates with a social networking platform, to learn about the digital and social behaviour of the customers and then uses the inferences to publish digital advertisements on the user pages of the targeted customers who have a high propensity to take up the product. A binary classification model needs to be developed to identify prospective customers.

Skills and Tools

ANOVA, Cluster Analysis, Decision Trees, Discriminant Analysis, Factor Analysis

Conclusion

The dataset was cleaned and treated for missing values and outliers and thereafter split into train and test datasets. Ten different binary classification models were developed and their performance was checked on the test dataset. The best three models were shortlisted for prediction on any new unseen datasets. The three models were shortlisted based on their overall performance, i.e. Precision/Recall/F1-Score. Presence/absence of overfit/underfit of the models was also duly considered prior shortlisting the models.