Indian Start-up Ecosystem Funding

What is the funding overview of startups in India. Data Exploration with Python

Elvis Darko

Our Company, XYZ Partners Limited seek to enter the Indian Market. As Data Analysts, we carefully explore funding for startups in India. We ask business questions, use the the available data to answer the questions and present recommendations.

Natural Language Processing :Sentiment Analyis

Fine-tuning already built text classification models from Hugging Face and deploying them with a streamlit app to analyze tweet sentiments

Elvis Darko

As a Data Analyst, I finetune a pre-trained machine learning model to assess if a Twitter post related to vaccinations is positive, neutral, or negative.This solution could help governments and other public health actors monitor public sentiment towards COVID-19 vaccinations and help improve public health policy, vaccine communication strategies, and vaccination programs across the world.

Time Series Analysis and Sales Prediction

What is the sales pattern of Coporacion Favorita? What factors influence sales?
Times Series Forecasting and ML modeling with Python

Elvis Darko

As a Data Analyst for Corporacion Favorita, a large Ecuadorian-based grocery retailer, I perform a times-series analysis on sales from all outlets. Also, I build a model that more accurately predicts the unit sales for thousands of items sold at different Favorita stores.

Customer churn analysis

What are the key indicators of churn?
Data exploration and ML modeling with Python

Elvis Darko

In my role as the Senior Business Analyst of GMC Telecommunications, an imaginary telecommunications company, I was informed that customer retention has got to a record low. My job is to discover what the key contributors are for customers dropping the company's products and services. Again, I am tasked to build a model that predicts the likelihood of a customer churning.

ML Model Deployment : Embedding ML model into interactive GUI

Model deployment forms part of the CRISP-DM framework
To get the most value out of my machine learning models, it is important I seamlessly deploy them into production so a business or client can start using them to make practical decisions

Elvis Darko

As a Senior Data Scientist for GMC Consulting Partners, I make business use of my machine learning models by deploying them into interactive Graphic User Interfaces. I deploy my regression model with a Streamlit App and classification model with a Gradio App. This will allow clients to put my models to productive use for strategic decision making.

ML Model Deployment : Embedding ML model into web app using Fast Api and Docker

Model deployment forms part of the CRISP-DM framework
In this dployment project, I use Fast API and Docker to deploy an ML model predicting sepsis disease status on a web App

Elvis Darko

Sepsis, a life-threatening condition arising from infection, poses a significant global healthcare challenge. By harnessing advanced data analytics techniques and exploring diverse parameters such as vital signs, medical history, and demographic information, I aim to identify early warning signs and risk factors for sepsis development. The focus lies on using the FastAPI and Docker frameworks to create a robust and user-friendly web interface for healthcare professionals to accurately detect and classify sepsis cases and respond effectively to this life-threatening condition

Expresso Customer Churn Prediction

ML model deployment using Streamlit, FastAPI and Docker.

Elvis Darko

Customer attrition is a critical business problem. As Data Scientists consulting for Expresso Telecom, we focuse on developing a machine learning model that accurately predicts customer who are likely to become inactive and not make any transactions for 90 days. By identifying potential churners, management can take proactive measures to retain customers and improve satisfaction.
Also, we build a simple Graphic User Interface using streamlit to deploy and operationalize our best performing ML model. Additionally, we use FastAPI to deploy same model on the web.