customizable

Traffic Signs Image Classification

A Multi-class classification of traffic sign images using various ML algorithms aimed at categorization of high impact classes with an accuracy of 93%.

Toolkit: Classification(SVM, Naive Bayes, Random Forest, KNN, Multi-Layer Perceptron, LDA), Python, scikit-learn, EDA, PCA

Learn more

customizable

Predicting Game success

Applying machine learning models to predict new games’ success rate in the current market, along with sentiment analysis of user reviews represented in an interactive visualization.

Toolkit: Regression(Multiple Linear Regression, Poisson), Python, Sentiment Analysis, HTML
Quick View Project Poster

Learn more

customizable

Face Recognition using PCA

Face recognition of subjects by computing top eigen faces from images shot with different expressions.

Toolkit: Python, scikit-learn, PCA, Classification

Learn more

customizable

Image Compression

Performing image compression of multiple image sizes using K-means clustering algorithm implemented from scratch

Toolkit: Python, Clustering (K-Means)

Learn more

customizable

Spectral Clustering

Experimenting Spectral clustering algorithm with a synthetic dataset.

Toolkit: Python, Spectral Clustering, scikit-learn
Spectral Clustering for two rings

customizable

Comparing Classifiers

A comparison of a several classifiers in scikit-learn on synthetic datasets to illustrate the nature of decision boundaries of different classifiers.

Toolkit: Python, Classification(Naive Bayes, Decision Trees,Random Forest, Ada Boost, Neural Nets, SVM, QDA), scikit-learn

Learn more

customizable

ISOMAP for Dimensionality reduction

Implementing the ISOMAP algorithm to obtain a two-dimensional embedding for images corresponding to different poses of the same person and comparing its results with PCA.

Toolkit: Python, Non-linear Dimensionality reduction

Learn more

customizable

Clustering digits

Applying clustering techniques on hand-written images of digits.

Toolkit: Python, Clustering (KMeans)

Learn more

customizable

Principal Component Analysis

Applying PCA on multiple data sets to identify trends and directions.

Toolkit: Python, Principal Component Analysis
Analysis on Leaf data set
Food consumption patterns in European Countries
PCA on hand-written digits images

Learn more

customizable

Regularization

Applying Regularization techniques on a Housing dataset.

Toolkit: Python, Regression, Regularization (Ridge, Lasso)

Learn more