分散式環境下主題性歷史消費經驗整理研究

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：48

、訪客IP：18.221.11.166

姓名

陳萬鈿(Wan-Dian Chen) 查詢紙本館藏

畢業系所

資訊工程學系

論文名稱

分散式環境下主題性歷史消費經驗整理研究
(Topic-Based Thematic Consumption Experience Research In Distributed Environment)

相關論文

★ 應用自組織映射圖網路及倒傳遞網路於探勘通信資料庫之潛在用戶	★ 基於社群網路特徵之企業電子郵件分類
★ 行動網路用戶時序行為分析	★ 社群網路中多階層影響力傳播探勘之研究
★ 以點對點技術為基礎之整合性資訊管理及分析系統	★ 在分散式雲端平台上對不同巨量天文應用之資料區域性適用策略研究
★ 應用資料倉儲技術探索點對點網路環境知識之研究	★ 從交易資料庫中以自我推導方式探勘具有多層次FP-tree
★ 建構儲存體容量被動遷徙政策於生命週期管理系統之研究	★ 應用服務探勘於發現複合服務之研究
★ 利用權重字尾樹中頻繁事件序改善入侵偵測系統	★ 有效率的處理在資料倉儲上連續的聚合查詢
★ 入侵偵測系統：使用以函數為基礎的系統呼叫序列	★ 有效率的在資料方體上進行多維度及多層次的關聯規則探勘
★ 在網路學習上的社群關聯及權重之課程建議	★ 在社群網路服務中找出不活躍的使用者

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

網路資訊流通快速，許多傳統商業行為轉變為網路行銷模式，消費者在進行消費決策時，無法有效評估產品品質，往往只能透過網路與親友的評論與實測來評估產品品質，由於在許多網路平台上評論的真實性無法詳細確認，因此本研究透過分析歷史消費評論幫助消費者進行消費決策以及紀錄產品全面且長期表現。
現今有許多平台針對不同產品提供消費者進行產品討論，由於大部份的平台都是中心化管理，並且不同族群喜好不一，使用的平台也不盡相同，因此本研究期望透過整合跨平台資料，針對產品評論進行情感分析並提供使用者一個較為客觀的指標進行決策分析，過往在自然語言處理情感分析方法有關鍵字識別、統計方法、詞彙關聯但這些方法對於深層且複雜多類別的主題並不那麼有效，本研究透過使用大型語言模型針對產品多面向的發展採用較為全面的情感分析。
結合去中心化技術和智能合約能更有效地保證評論的不可竄改性，從而使情感分析指標更加可信，透過智能合約，中心化平台可以避免人為竄改的風險，進一步增強此去中心化應用系統的可靠性。

摘要(英)

In the fast-paced world of internet information flow, many traditional commercial behaviors have shifted towards online marketing models. Consumers, when making purchasing decisions, are often unable to effectively evaluate product quality and typically rely on internet and personal acquaintances′ reviews and hands-on tests. Given that the authenticity of reviews on many online platforms cannot be thoroughly verified, this study aims to assist consumers in making purchasing decisions by analyzing historical consumer reviews.
Currently, many platforms provide spaces for consumer discussions on various products. Most of these platforms are centrally managed, and preferences for them vary across different demographics, leading to the use of diverse platforms. This study seeks to integrate cross-platform data to conduct sentiment analysis on product reviews, offering users a more objective indicator for decision-making analysis. Traditional natural language processing sentiment analysis methods, such as keyword identification, statistical methods, and lexical association, are not as effective for multi-category topics. This research adopts a more comprehensive sentiment analysis approach by utilizing large language models to address the multifaceted development of products.
Decentralized technology and smart contracts can more ensure the immutability of reviews, thereby making sentiment analysis indicators more trustworthy. Through smart contracts, centralized platforms can avoid the risk of human tampering, further enhancing the reliability of this decentralized application system.

關鍵字(中)

★ 區塊鏈
★ 智能合約
★ 大型語言模型
★ 情感分析

關鍵字(英)

★ BlockChain
★ Smart Contract
★ Large Language Model
★ Sentiment Analysis

論文目次

目錄
摘要 i
Abstract ii
誌謝 iii
目錄 iv
圖目錄 vi
表目錄 vii
一、緒論 1
1-1. 研究背景與動機 1
1-2. 研究目的 1
二、文獻探討 3
2-1.區塊鏈 3
2-2.以太坊 3
2-3.星際檔案系統 4
2-4.大型語言模型 5
2-5.QLORA 6
2-6.情感分析 8
2-7.模型效能評估 10
2-7-1.精準度(Precision) 10
2-7-2.召回率(Recall) 10
三、研究方法 11
3-1.系統環境與平台架構 11
3-1-1.可替換情感分析模型架構 11
3-2.評論資料前處理 12
3-3.系統開發流程 13
3-4.評論情感分析 13
3-4-1.相異方法於情感分析方法整理 14
3-4-2. 模型輸入形式範例 15
3-4-3. 層面類別偵測 15
3-5.大型語言模型微調訓練 17
3-6.多層分類任務模型效能評估 19
3-7.智能合約設計 20
3-7-1.User Enroll 21
3-7-2.Upload address 21
3-7-3.Download address 22
四、實驗 23
4-1.實驗規格與環境 23
4-2.實驗資料集 23
4-2-1.前處理資料層面標籤 24
4-3.模型訓練參數 25
4-4.產品長期表現 26
4-5.相異平台偏好 28
4-5-1.A平台情緒指標結果 28
4-5-2.B平台情緒指標結果 29
4-6.智能合約費用 30
五、結論 32
六、參考文獻 35

參考文獻

參考文獻

[1] S. Nakamoto, "Bitcoin: A peer-to-peer electronic cash system," 2008.
[2] L. Lamport, R. Shostak, and M. Pease, "The Byzantine generals problem," in Concurrency: the works of leslie lamport, 2019, pp. 203-226.
[3] G. Wood, "Ethereum: A secure decentralised generalised transaction ledger," Ethereum project yellow paper, vol. 151, no. 2014, pp. 1-32, 2014.
[4] W. Zou et al., "Smart contract development: Challenges and opportunities," IEEE transactions on software engineering, vol. 47, no. 10, pp. 2084-2106, 2019.
[5] A. Gervais, G. O. Karame, K. Wüst, V. Glykantzis, H. Ritzdorf, and S. Capkun, "On the security and performance of proof of work blockchains," in Proceedings of the 2016 ACM SIGSAC conference on computer and communications security, 2016, pp. 3-16.
[6] S. King and S. Nadal, "Ppcoin: Peer-to-peer crypto-currency with proof-of-stake," self-published paper, August, vol. 19, no. 1, 2012.
[7] J. Benet, "Ipfs-content addressed, versioned, p2p file system," arXiv preprint arXiv:1407.3561, 2014.
[8] W. X. Zhao et al., "A survey of large language models," arXiv preprint arXiv:2303.18223, 2023.
[9] S. Khan, M. Naseer, M. Hayat, S. W. Zamir, F. S. Khan, and M. Shah, "Transformers in vision: A survey," ACM computing surveys (CSUR), vol. 54, no. 10s, pp. 1-41, 2022.
[10] E. J. Hu et al., "Lora: Low-rank adaptation of large language models," arXiv preprint arXiv:2106.09685, 2021.
[11] T. Dettmers, A. Pagnoni, A. Holtzman, and L. Zettlemoyer, "Qlora: Efficient finetuning of quantized llms," Advances in Neural Information Processing Systems, vol. 36, 2024.
[12] W. Medhat, A. Hassan, and H. Korashy, "Sentiment analysis algorithms and applications: A survey," Ain Shams engineering journal, vol. 5, no. 4, pp. 1093-1113, 2014.
[13] M. Grinberg, Flask web development. " O′Reilly Media, Inc.", 2018.
[14] K. Bhosale, K. Akbarabbas, J. Deepak, and A. Sankhe, "Blockchain based secure data storage," International Research Journal of Engineering and Technology (IRJET), vol. 6, no. 3, pp. 5058-5061, 2019.
[15] S. U. S. Chebolu, F. Dernoncourt, N. Lipka, and T. Solorio, "Exploring conditional text generation for aspect-based sentiment analysis," arXiv preprint arXiv:2110.02334, 2021.
[16] C. Wu et al., "Multiple-element joint detection for aspect-based sentiment analysis," Knowledge-Based Systems, vol. 223, p. 107073, 2021.
[17] P. Barnaghi, P. Ghaffari, and J. G. Breslin, "Opinion mining and sentiment polarity on twitter and correlation between events and sentiment," in 2016 IEEE second international conference on big data computing service and applications (BigDataService), 2016: IEEE, pp. 52-57.
[18] H. Touvron et al., "Llama 2: Open foundation and fine-tuned chat models," arXiv preprint arXiv:2307.09288, 2023.
[19] W.-M. Lee, "Using the MetaMask crypto-wallet," in Beginning Ethereum Smart Contracts Programming: With Examples in Python, Solidity, and JavaScript: Springer, 2023, pp. 111-144.

指導教授

蔡孟峰(Meng-Feng Tsai)

審核日期

2024-7-22

推文