Collaborative recommender system for online jewelry store

The existence of vast amount of data exist in internet nowadays has become a dilemma in the field of electronic commerce. Searching of the desired information has become so inconvenient since there are too many irrelevant information exist all over the shopping platform. One of the most popular s...

Full description

Bibliographic Details
Main Author: Hon, Morris Mao Ning
Format: Final Year Project Report / IMRAD
Language:English
Published: Universiti Malaysia Sarawak, (UNIMAS) 2015
Subjects:
Online Access:http://ir.unimas.my/id/eprint/12194/
http://ir.unimas.my/id/eprint/12194/3/Morris%20Hon.pdf
Description
Summary:The existence of vast amount of data exist in internet nowadays has become a dilemma in the field of electronic commerce. Searching of the desired information has become so inconvenient since there are too many irrelevant information exist all over the shopping platform. One of the most popular solution nowadays to solve this dilemma is recommender system. Recommender systems are now pervasive in user’s lives. They aim to help users in finding items that they would like to buy or consider based on huge amount of data collected. Parsing a huge amount of data to predict user’s preference base on his or her similarity with other group of users is the core of recommender system. One of the famous approach that could be applied to the implementation of recommender system is collaborative filtering approach. The motivation to do this project comes from my eagerness to improve my web developing skill especially in the field of jewelry e-commerce and to get a deep understanding of recommender system. In this project, a prototype of online jewelry selling store with the implementation of a collaborative filtering based recommender system was developed. The algorithm under collaborative filtering approach that been used in this project is called slope one algorithm which basically works by predicting user’s preference based on other user’s rating history on specific items in the system. Finally, the prototype built in this project was evaluated in term of the performance of the recommender system under user’s point of view and the results of evaluation was discussed in details to draw out a conclusion for its further improvement.