nitinjosephrepo

Changing-Trends-in-Online-Petfood

Use Pytrend which is an Unofficial API for Google Trends to visualize global changes in online purchasing preference in last 3 months

Feature-Selection-Exploratory-Data-Analysis

Effort here is to identify important features in relation to our target variable using multiple Correlation methods based on data type. Feature selection is important for ML models to avoid 'curse of dimensionality' but for this dataset we will be using it build our intution that benefits our later EDA effort

Google-Causal-Impact-Inferring-the-effect-of-an-event

The Causal Impact model lets you examine ecommerce and marketing time series data to understand whether changes have led to a statistically significant performance improvement. Here's how to use PyCausalImpact to analyse changes in marketing activity or in this case on Boeing stock price

Improving-Conversion-Rate-With-Decision-Trees

Use Decison Tree on Bank Marketing Dataset to Identify Customer attributes that are drivers of Conversion. We will then interpret these trained decison tree models by visualizing them using the graphviz package in python.

Items-Frequently-Bought-Together

To answer which items are frequently bought together we will be using Apriori & FPgrowth Algorithm

Marketing-Mix-Modeling-MMM-Using-Robyn

A New GeneratUsing Robyn aims to reduce human bias in the modeling process, esp. by automating modelers decisions like adstocking, saturation, trend & seasonality as well as model validation. Moreover, the budget allocator & calibration enable actionability and causality of the results

Mining-Insights-From-Customer-Reviews

Use GSDMM Package for Topic Modeling on Yelp Review Corpora, GSDMM works well with short sentences found in reviews.

Practical-Time-Series-Analysis

Practical Time-Series Analysis, published by Packt

Using-NLP-Models-To-Improve-SEO

This Repo contains my projects that I have used to improve SEO and Content Strategy

Using-Prophet-to-Forecast-Bikeshare-Demand

Prophet time series forecasting model was developed by Facebook and is a powerful tool for predicting future events. Here's how to use it to forecast & understand your Data

nitinjosephrepo/Marketing-Mix-Modeling-MMM-Using-Robyn

A New GeneratUsing Robyn aims to reduce human bias in the modeling process, esp. by automating modelers decisions like adstocking, saturation, trend & seasonality as well as model validation. Moreover, the budget allocator & calibration enable actionability and causality of the results

nitinjosephrepo/Google-Causal-Impact-Inferring-the-effect-of-an-event

The Causal Impact model lets you examine ecommerce and marketing time series data to understand whether changes have led to a statistically significant performance improvement. Here's how to use PyCausalImpact to analyse changes in marketing activity or in this case on Boeing stock price

nitinjosephrepo/Feature-Selection-Exploratory-Data-Analysis

Effort here is to identify important features in relation to our target variable using multiple Correlation methods based on data type. Feature selection is important for ML models to avoid 'curse of dimensionality' but for this dataset we will be using it build our intution that benefits our later EDA effort

nitinjosephrepo/Improving-Conversion-Rate-With-Decision-Trees

Use Decison Tree on Bank Marketing Dataset to Identify Customer attributes that are drivers of Conversion. We will then interpret these trained decison tree models by visualizing them using the graphviz package in python.

nitinjosephrepo/Items-Frequently-Bought-Together

To answer which items are frequently bought together we will be using Apriori & FPgrowth Algorithm

nitinjosephrepo/Practical-Time-Series-Analysis

Practical Time-Series Analysis, published by Packt

nitinjosephrepo/Using-NLP-Models-To-Improve-SEO

This Repo contains my projects that I have used to improve SEO and Content Strategy

nitinjosephrepo/About-Me

Config files for my GitHub profile.

nitinjosephrepo/Abstractive-Summarization-With-GoogleT5

Repo contains my efforts to use both extractive and abstractive summarization techniques that can help assist with SEO task like Meta Description generation

nitinjosephrepo/Analyzing-Online-Shopper-Purchase-Intention

Analyze online shoppers' purchase intentions using Logistic Regression, K-means clustering & A/B Testing

nitinjosephrepo/Changing-Trends-in-Online-Petfood

Use Pytrend which is an Unofficial API for Google Trends to visualize global changes in online purchasing preference in last 3 months

nitinjosephrepo/Mining-Insights-From-Customer-Reviews

Use GSDMM Package for Topic Modeling on Yelp Review Corpora, GSDMM works well with short sentences found in reviews.

nitinjosephrepo/Using-Prophet-to-Forecast-Bikeshare-Demand

Prophet time series forecasting model was developed by Facebook and is a powerful tool for predicting future events. Here's how to use it to forecast & understand your Data

nitinjosephrepo/Building-Ecommerce-RFM-Model-With-Segementation

The “RFM” in RFM analysis stands for recency, frequency and monetary value. RFM analysis is a way to use data based on existing customer behavior to predict how a new customer is likely to act in the future.

nitinjosephrepo/Conjoint-Analysis-to-uncover-consumer-preferences

Conjoint analysis is a data informed approach to understanding what consumers prefer about a product

nitinjosephrepo/Data-Driven-Customer-Segementation

For any marketing campaign, it's critical to understand different behaviours, types and interests of Customers. Especially in targeted marketing, understanding and categorizing customers is an essential step for effective marketing strategies.

nitinjosephrepo/Data-Prepration-For-ML

Most machine learning algorithms require data to be formatted in a very specific way, so datasets generally require preparation before they can yield useful insights. This repository is to document my study notes as I work through steps that I have personally found most challenging.

nitinjosephrepo/Dyad

nitinjosephrepo/Estimating-Customer-Lifetime-Value-BG-NBD-Model

BG|NBD Model uses binomial probability to determine Customer Life Time Value and the likelihood of which customers are 'alive'

nitinjosephrepo/Logistic-Regression

Repo contains my personal Machine Learning projects with emphasis on explanability and Insights that are relevant for various stake holders in a business.

nitinjosephrepo/loveofdata.github.io

nitinjosephrepo/Measuring-Incrementality-With-GeoLift

Measure the true incremental value of your marketing campaigns with GeoLift

nitinjosephrepo/Multilabel-Segementation-Using-Xgboost

For this Multi-class classification dataset we will be using Xgboost, multi:softprob objective which is a standard alternative to binary:logistic when the dataset includes multiple classes. It computes the probabilities of classification and chooses the highest one.

nitinjosephrepo/Multiple-Linear-Regression-to-Predict-Price-of-Used-Cars

Along with predicting the price of used car we utilize Recursive Feature Elimination from Statsmodel to identify features that have most impact on resale price

nitinjosephrepo/Natural-Language-Processing-Projects

nitinjosephrepo/nitinjosephrepo

nitinjosephrepo/Predicting-Customer-Churn-and-Factors-Responsible

Predicting which factors are responsible for Telco Customer Churn

nitinjosephrepo/Predicting-Customer-Lifetime-Value

In Marketing, the CLV is one of the key metrics to have and monitor. The CLV measures customer's total worth to the business over the course of their lifetime relationship to a business. This metric is especially important to keep track of for acquiring new customers.

nitinjosephrepo/Predicting-Defaulters-Using-Artificial-Neural-Network

ANN or Deep Learning can be utilized for many areas of marketing. Using Neural network models by BrainMaker, Microsoft increased its direct mail response rate my 4.9% to 8.2%. This helped Microsoft bring same amount of revenue for 35% les cost.

nitinjosephrepo/Time-Series-Analysis