Title: 投資組合推薦之群眾智慧機制
A Collective Intelligence Mechanism for Investment Portfolio Recommendation
Authors: 黃柏勳
Huang, Bo-Syun
Li, Yung-Ming
Keywords: 社群網路;文字探勘;投資組合創建;群眾智慧;決策支援系統;social network;text mining;investment portfolio creation;collective intelligence;decision support system
Issue Date: 2017
Abstract: 在過去,人們將閒置的資金存放於銀行來獲取一些額外的收益,隨著全球負利率政策的趨勢,將錢放在銀行不但不能獲利還必須繳納一定的金錢當作資金的保管費,也因此,人們現在傾向於進行投資來賺錢。另一方面,隨著社群網路的興起,越來越多的人喜歡在網路上表達他們對於投資的想法與策略。近年來,像這樣的資訊已經逐漸被用來作為創造投資理財組合的手段。因此本研究致力於提出一個基於群眾智慧的股票組合推薦系統,藉由蒐集大量平台的評論來進行語意分析,並且針對評論者的知識與影響力作為可信度的權重調配,最後產生推薦的投資組合,讓理財初學者也能夠進行投資決策,而不需花費大量的時間研讀財金的知識。
In the past, people hold their unused cash at central banks and hope to receive a small amount of interest in return. But with negative interest rates, central banks charge a fee instead. From European Central Bank to Japan’s central bank, negative interest rates policy has become a trend around the world. Because of that, people tend to spend their money on investment instead of saving. On the other hand, increasing numbers of people are using social media to express their personal experiences on investment. Recently, these plentiful user-generated data sources have been utilized promisingly by investors for portfolio creation. In this research, we propose a social intelligence mechanism that can extract and consolidate the reviews expressed via social networks and derive a portfolio to help users make decisions on investment by analyzing the reviewers' knowledge and authority and their opinion sentiment toward the investment target. The experimental results obtained using eToro.com show that the proposed mechanism outperforms other benchmark approaches in market.
URI: http://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070453429
Appears in Collections:Thesis