Identify and classify toxic online comments using NLP and Machine-Learning algorithms

scikit_learn

Employee attrition prediction with an interactive environment

Predict rain the next day using daily observations of weather aspects in Australia regions for 10 years

Fake News analysis and prediction in R Script. Naive Bayes, Random Forest, SVM, NNET, ROC, Confusion Matrix, Accuracy, F1 score.

This employee attrition analysis project for XYZ Company identifies key factors driving turnover, like compensation and growth. Using data science, we built a model to predict attrition risks and provide actionable insights to improve employee retention and workforce stability.

ROC curve for any dataset, with 8 classification Algorithm in One.

Drawing Roc Curve

The code loops through the list of models and plots respective roc curves in a single plot

The objective of this project is to predict the probability of loanee or borrower defaulting on a vehicle loan in the first EMI (Equated Monthly Instalments) on the due date.

EC60091-Machine Intelligence & Expertise Systems Course,Autumn-2019

This project aims to build a machine learning model using K-Nearest Neighbor, LogisticRegression, RandomForestClassifier to classify whether or not a person has heart disease based upon his medical attributes. (accuracy achieved : 88.52%)

Rsoftware - TRATAMIENTO DE DATOS QUÍMICO-FORENSES PARA LA DISCRIMINACIÓN DE FLUIDOS BIOLÓGICOS EN MATERIALES SUPERABSORBENTES

Here we will be firstly analysing the how different threshold values effect the area under the Curve in a Receiver Operating charcteristic(ROC) curve. And at last we will show how to define a function in python to calculate the most optimal threshold value for the logistic Regression.

Strategizing to maximize Customer Retention in Telecom Company

Insurance Cross Sell Opportunity Forecast through machine learning algorithm

Poland Bankruptcy Prediction (2009) This project aims to predict whether a Polish company went bankrupt in 2009 based on its financial data. The dataset contains several features derived from companies' balance sheets, and the goal is to build models that can identify bankruptcy effectively — despite the challenge of high class imbalance.

Explorartory Data Analysis with visualization. Use the supervised machine learning models to predict the customer segmentation. Use unsupervised learning model K Nearest Neighbors to create new clusters. Create Personas of the new clusters.

Analiza indicatorilor macroeconomici precum pretul, numarul de actiuni, venit, profit, cash flow etc pentru firmele din industria materiilor prime.

Work from various classes