結合內容導向與腦波分析為基礎的文章推薦系統

A Document Recommendation System based on Content-based Filtering and Brainwave Analysis

蘇柏銘(Po-Ming Su)

指導教授 ： 陳灯能

國立屏東科技大學/管理學院/資訊管理系所/碩士(2014年)

https://doi.org/10.6346/NPUST.2014.00093

摘要

推薦系統是一種基於使用者紀錄進行資料分析，藉此分析結果並主動式資訊推薦，推薦系統在提昇個人化資訊服務品質上扮演重要的角色。推薦系統的設計著重於將資料庫中使用者的相關記錄進行分析，也因此衍生出內容導向、協同導向等不同演算法為基礎的推薦系統，本研究則是嘗試將使用者的腦波專注力納入推薦系統的演算法設計之中，首先以實驗法收集受測者腦波訊號與其興趣偏好資料，並利用類神經網路分析兩者之間的關聯模型，進而以此關聯模型為核心開發結合腦波與內容導向資訊過濾為基礎的文章推薦系統，最後以實驗法驗證本推薦系統的推薦精準度。研究結果證明腦波能夠有效利用在推薦系統的設計上。

關鍵字

推薦系統 ； 內容導向 ； 內容導向 ； 腦波 ； 資訊過濾

並列摘要

Recommendation system is based on history similar with viewed history or Purchased history to find out what is user demand. There are three kind of recommendation system: Content-based、Collaborative-based and Mixing-based. Brainwave is a kind of biological signal which has been divided into α、β、δ、θ. Device of brainwave is not expensive anymore that could be used to study. In theory of education indicate attention and interest has a Strong relationship. Base on the theory the first purpose of this study is figure out the relationship between the Attention of brainwave and user’s interest in the web site of online reading by neural network algorithms. Then use this relationship to predict items in user history and only adopt items which be predicted to category of interested to conduct Content-based recommended. Expect this way could exclude invalid factor of user history.

並列關鍵字

Recommendation system ； Content filtering ； neural network ； Brainwave

參考文獻

5. 網路家庭國際資訊股份有限公司，“一○○年度年報. ”，2011.

7. 朱璿瑾., 江政祐., and 劉寧漢，“運用腦波識別專注狀態,”，資訊科技國際期刊 ( 7:2), 2013, pp. 14-23.

1. Balabanovic, M., and Shoham, Y. “Fab: Content-based, collaborative recommendation,” Communications of the Acm (40:3), 1997, Mar, pp 66-72.

2. Blair, A. 2010. “Information Overload, Then and Now.”

3. Brusilovsky, P., Kobsa, A., and Nejdl, W. The adaptive web: methods and strategies of web personalization, ed. Springer,2007.