ML / DL Algorithms implemented from scratch. Developed with only numpy as dependency. Machine Learning Algorithms such as Support Vector Machine, Linear Regression, Artificial Neural Networks and other data transformation algorithms are implemented. Project is released as a python package and can be download from Python Package Installer.

Build a machine learning model to predict if a credit card application will get approved.

I have created this project as a part of virtual internship programme in data Science.

An innovative system for filtering and categorizing movie reviews

Exploratory Data Analysis Part-1

Processing of data gaps, coding of categorical features, data scaling.

The sixth project from a Data Scientist with Python track by DataCamp

This is a project to detect anomalies in pump sensor data using One-Class Support Vector Machines (SVM). The data is preprocessed by dropping columns with missing values and scaled using MinMaxScaler. The one-class SVM classifier is trained and used to predict anomalies in the data, which are then saved in a new file "results.csv".

Customer churn prediction using deep learning

Credit Card Approval Prediction based on users' historic data.

pipelines chains together multiple steps so that the output of each step is used as input to the next step

The main objective of this project is to design and implement a robust data preprocessing system that addresses common challenges such as missing values, outliers, inconsistent formatting, and noise. By performing effective data preprocessing, the project aims to enhance the quality, reliability, and usefulness of the data for machine learning.

Performing kmeans clustering and also providing elbow plot

Data Science Project: Comparing 3 Deep Learning Methods (CNN, LSTM, and Transfer Learning).

A Book Recommendation System that utilizes Python libraries such as numpy, pandas, seaborn, and matplotlib to recommend books based on user input.

[ Analyzing the existing customer data and getting valuable insights about the purchase pattern ] | K-Means clustering | silhouette score | minmaxscalar |

The feature engineering techniques discussed are - dimensionality reduction(pca), scaling(standard scaler, normalizer, minmaxscaler), categorical encoding(one hot/dummy), binning, clustering, feature selection. These are techniques performed on a dataset consisting of Californian House Prices.

The Bike Sharing Company wants to understand the independent variables on their past data to analyze and create a machine learning model to understand the demand of the bike and accordingly plan a business strategy.