Capstone

Link to browser code

Link to DuckDuckGo browser code

How can we implement an in-build javascript classification model into a mobile browser to accelerate mobile web pages?

Table of Contents

• Introduction
• Project Overview
• Project Setup
• Project Instructions
• Project Results
• License

Introduction

This project is part of a senior year Capstone project at the New York University. The project is in collaboration with the ComNets. The goal of the project is to implement a machine learning model into the Kiwi browser to accelerate mobile web pages.

Project Overview

The project is divided into three parts:

1. Converting a machine learning model (sklearn classification and clustering) into C++ adaptable format (our choice is ONNX)
2. Implementing a machine learning model into the Kiwi browser
3. Testing the performance of the Kiwi browser with and without the machine learning model
4. Testing the performance of the Kiwi browser with and other existing tools that accelerate mobile web pages

Project Setup

Prerequisites

Check each of the phase directories for the prerequisites. The prerequisites are different for each phase.

Installation

1. Clone the repo

git clone

Usage

Check each of the phase directories for the usage.

Project Instructions

Project Results

License

Distributed under the MIT License. See LICENSE for more information.