You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In this project, I explore different methods for detecting credit card fraud transactions; including using the Catboost algorithm with undersampling & oversampling methods, and using an almost new approach, by using deep learning and autoencoder.
Predict if a transaction is a fraud transaction or not, also, dealing with unbalanced data and finding the pattern using correlation between the features.
Assess credit risk of applicants using supervised machine learning. Several different machine learning techniques such as SMOTE, SMOTEENN, RANDOM FOREST, EASY ENSEMBLE were applied, the models were assessed using accuracy score, precision and accuracy to choose the best technique that applies to this type of problem.
It's Technocolabs Software Data Scientist Internship Project (1st Dec 2021 - 15th Jan 2022). In this project the team was instructed to analysis big data of Spotify users and to perform Statistical and Exploratory Data Analysis and Model Development for Predicting Listener Behavior.
The project was intended to detect fraudulent transactions from a highly imbalanced dataset.To solve the imbalance dataset problem random undersampling techniques were used.
Toxicity detection on imbalanced social media data. Focused on 2 main topics of toxicity detection: Class imbalance problem and detecting toxic comments.
Under-sampling based consensus clustering is applied on the three best clustering algorithms found after applying several Clustering Algorithms like K-means, K-modes, K-prototypes , K-means and fuzzy K-means on the majority class data of the IMBALANCED colon dataset to produce a BALANCED dataset.
This project is based on supervised machine learning where you will be predicting whether a credit card transaction is original transaction or fraud transaction based on various parameters. This is a classification problem.
In this project, I worked on a classification problem using an imbalanced dataset which predicts ecological footprints. The aim of the project was not necessarily to build a classification model but to investigate the different methods of correcting an imbalanced dataset in order not to build a biased classifier