Please use this identifier to cite or link to this item:
|Title:||E-Commerce Collaborative Filtering System based on Cloud Computing|
|Authors:||Chan, Wai Ming|
|Department:||Department of Electronic Engineering|
|Supervisor:||Supervisor: Dr. SO, H C; Assessor: Prof. LI, Ping|
|Abstract:||Web applications nowadays are generating billions GB of data every day, the data are not only huge in volume but also variety in different formats and generating in high velocity. Traditional data analysis technique is becoming difficult to handle the Big Data in terms of data storage size and computation speed. By analysing the web user preference data, we can predict the users’ favour and then make recommendation to them. One approach of recommendation system is collaborative filtering. Collaborative filtering is a framework for filtering information based on users’ preferences and it is used to make user-item rating prediction. This project is to analyse existing collaborative filtering methods for the rating prediction and show the capability on the increasing data size of the methods. Next, the project introduces distributed computing manner of collaborative filtering so as to handle scalable data size. The project introduces Stochastic Gradient Decent method on collaborative filtering and implements it on Hadoop MapReduce framework. Finally, the project tries to deploy the work in the cloud platform, Amazon Web Service (AWS), in order to analyse gigabytes of data.|
|Appears in Collections:||Electronic Engineering - Undergraduate Final Year Projects |
Items in Digital CityU Collections are protected by copyright, with all rights reserved, unless otherwise indicated.