[ICML 2024] Outlier-Efficient Hopfield Layers for Large Transformer-Based Models

Successfully established a machine learning model which can predict whether any given water sample is potable or not, based on its set of various properties, to a considerably high level of accuracy.

This case study involves helping X Education, an education company, improve its lead conversion rate by building a logistic regression model to assign lead scores. The aim is to identify potential leads with the highest chances of converting to paying customers and handling future problems to achieve a target conversion rate of 80%.

We harness the power of machine learning and data analysis to real challenges in the copper industry. Our documentation covers data preprocessing, feature engineering, classification, regression, and model selection. Discover how we've optimized predictive capabilities for manufacturing solutions.

"Music Album Popularity Prediction" is a project focused on building a model to forecast the success of music albums. By analyzing streaming data, social media engagement, and other relevant factors, the project aims to predict the popularity of albums across various genres and artists.

Exploratory data analysis to propose innovative application ideas related to food

Customer Profiling for AeroFit Treadmill Products: Unveiling Target Audience Characteristics

X Education Organization wants to identify if a customer registered on their website for enquiry is a potential customer or not. Using past data to build a machine learning algorithm

Checking If a customer will default on the loan, if the bank provides them a loan. We will analyze the past patterns of payments by performing Exploratory Data Analysis.

Our group project aimed to evaluate three predictive machine learning classification models to anticipate whether website visitors engage in transactions. This is done by analysing different attributes of website visitors including duration spent on different web pages, click rates, and bounce rates.

Classification model that will help the bank improve its services so that customers do not renounce their credit cards

Perform Exploratory Data Analysis(EDA) on loan applications to understand how various client attributes (like marital status, education, occupation, etc.) influence the tendency of default.

In this repository, you'll find all the code solutions I've completed during my TECHNOHACKS EDUTECH internship as a data analyst intern.