基於HITS演算法之去中心化協作推薦平台;Decentralized Collaborative Recommendation Platform Based on HITS Algorithm

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/92514

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/92514

Title:	基於HITS演算法之去中心化協作推薦平台;Decentralized Collaborative Recommendation Platform Based on HITS Algorithm
Authors:	李奕昕;Lee, Yi-Hsin
Contributors:	資訊工程學系
Keywords:	區塊鏈;智能合約;去中心化;推薦演算法;Blockchain;Smart Contract;Decentralized;Recommendation algorithms
Date:	2023-07-18
Issue Date:	2023-10-04 16:03:39 (UTC+8)
Publisher:	國立中央大學
Abstract:	本研究為促進各專業領域的使用者共同協力創作，提出一個整合推薦系統與協作系統的平台，並選擇去中心化的區塊鏈作為整體架構，利用以太坊智能合約做為自動化管理、協助和監督的角色，以其可信任第三方的特性，免去傳統意義上的第三方仲介角色，可省去仲介手續費以及三方在溝通互動上的摩擦。取而代之，在推薦系統上，由於平台用戶紀錄有著較小的造假可能，配合推薦演算法的應用，讓使用者能在高可信度資料的優勢下找尋所需人才；在協作系統上，使用者可迅速著手於團隊專案的創作，同時也為協作的過程留下可追溯紀錄，以兼具效率且彼此信任的條件下，實現平台在推薦以及協作兩大服務項目。此研究專注於協作推薦平台在去中心化基礎架構的實現，為其提供相關功能設計和技術的參考，相較於中心化協作平台上既有的功能框架，提出不一樣的觀點作嘗試，這將使研究成果富有意義，將兩種概念作整合，得以在後續的研究當中實現功能更完備且高效的協作推薦平台。 ;This study proposes a platform that integrates a recommendation system and a collaboration system to promote collaborative creation among users in various professional fields. The platform is built on decentralized blockchain architecture using Ethereum smart contracts as an automated management, assistance, and supervision role. With its trustworthy third-party characteristics, it eliminates the need for traditional third-party intermediaries, saving on intermediary fees and reducing friction in communication or interaction among the three parties. Instead, in the recommendation system, due to the lower possibility of falsification of user records on the platform, combined with the application of recommendation algorithms, users can find the talents they need under the advantage of high credibility data. In the collaboration system, users can quickly start creating team projects while leaving traceable records of the collaboration process, achieve both efficiency and trustworthiness in the platform’s two major service: recommendation and collaboration. This study focuses on the implementation of a decentralized architecture for a collaborative recommendation platform, providing relevant functional design and technical references. Compared to the existing framework on centralized collaboration platforms, this research offers a different perspective for experimentation, making its results meaningful, integrating two concepts to achieve a more complete and efficient collaborative recommendation platform in future work.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	47	View/Open

社群 sharing

Loading...