| dc.contributor | 經濟系 | |
| dc.creator (作者) | 陳樹衡 | |
| dc.date (日期) | 2020-10 | |
| dc.date.accessioned | 16-Jul-2025 11:12:26 (UTC+8) | - |
| dc.date.available | 16-Jul-2025 11:12:26 (UTC+8) | - |
| dc.date.issued (上傳時間) | 16-Jul-2025 11:12:26 (UTC+8) | - |
| dc.identifier.uri (URI) | https://nccur.lib.nccu.edu.tw/handle/140.119/158038 | - |
| dc.description.abstract (摘要) | 在大數據經濟下，無所不在的文字、圖像、影音等多面資訊和運用，為人們的經濟行為帶來巨大的變化，同時也迫使經濟學家面對如何處理這些過去所不能獲得、對人類行為記載鉅細靡遺的資料，傳統經濟學理論無法有效處理異質化的個人行為及決策模式，勢必重用計算經濟學始能解決這個問題，然而目然的計算經濟學尚未充份引入大數據分析法來分析模擬出來的個體行為資料。而大數據分析師雖然可以很細緻地說明不同類人的行為，但其結論卻常缺乏（行為）經濟學的理論根據，不能讓吾人對構成這些大數據行為的機制或大數據對資訊整合的機制問題有所了解。如何整合計算經濟學、行為經濟學、及大數據分析，以對大數據的本質提出理論基礎，乃為當今經濟學的重要課題，而代理人基計算經濟學，從個體行為和交互規則編程自底向上來考察整體現象的建模方式，則為此提供了解決方案。資訊彙整（整合）一直是從海耶克以來支持市場經濟的重要論述，本計畫旨在建構大數據代理人基經濟學模型，以探究在大數據經濟的複雜社群網路環境中，資訊如何在個體間、個體和大數據間的互動關係下彙總出來，只有充分理解大數據資訊整合機制，才能真正把握大數據可能帶來的優勢和潛在風險。本計劃利用研究以下主題之方式來探究大數據資訊整合之機制：（1）資訊量和資訊品質對大數據資訊整合機制的影響；（2）網絡謠言在社群網站下的散播、偵測、和控制機制；（3）個人特質和社群網路運算法交互影響下產生意見兩極化或包容化之機制。本研究結果預期將幫助我們理解大數據背後的資訊彙總機制，提出政策意含及參考，規避大數據內含之潛在風險。 | |
| dc.description.abstract (摘要) | The ubiquitous, multifaced information available in the big data economy have changed our economic behaviors and decision making rules. Big data record what people say, do, and even think. The concept of big data is not new, but the accessibility is and must be accepted and utilized in economic research. Classical mathematical models cannot easily handle heterogeneous behaviors and decision-making rules. Simulation models, particularly agent-based models must be applied. However, big data analytics has not been effectively used in agent-based models to understand the simulated big data. On the other hand, although human behavioral patterns can be detected by using big data analytics, big data analysts usually do not have (behavioral) theories to underpin their findings. The integration of agent-based modeling, behavioral economics, and big data will provide us with a way to understand the theory underpinning the behaviors and macro phenomena in the big data economy. This study is to construct an agent-based big data model to understand the mechanism of how information aggregation emerges from the interactions between individual agents and between agents and their social networking environment. Information aggregation mechanism has been argued to support market economy since Friedrich Hayek. Only with the understanding of the information aggregation mechanism in big data can we truly harness big data for social good. We will investigate the information aggregation mechanism from the following three dimensions: (1) the impact of the quantity and quality of information on the valid information aggregation in big data, (2) the mechanisms of rumor spreading, detecting, and control in social networks, and (3) how opinion polarization or accommodation is formed through the interactions of personal characteristics and social media algorithms. Our findings will provide an understanding of the underpinning of big data, from which policy suggestions can be drawn for better information aggregation, rumor control and mitigation of opinion polarization. | |
| dc.format.extent | 116 bytes | - |
| dc.format.mimetype | text/html | - |
| dc.relation (關聯) | 科技部,, MOST106-2410-H004-006-MY2, 107.08-108.07 | |
| dc.subject (關鍵詞) | 代理人基建模; 大數據; 資訊整合; 謠言傳佈; 意見兩極化; 社群網路; 集群智慧; 集體愚昧 | |
| dc.subject (關鍵詞) | agent-based modeling; big data; information aggregation; rumor spreading; opinion polarization; social networks; wisdom of crowds; stupidity of herds | |
| dc.title (題名) | 大數據與經濟行為：一個以代理人基經濟建模為主的整合性架構 | |
| dc.title (題名) | Big Data and Economic Behavior: an Integrating Framework Using Agent-Based Computational Economic Modeling | |
| dc.type (資料類型) | report | |
