本研究提供異質系統貢獻度管理模型給WeOS網路服務平台。藉由智能合約將跨系統間的使用者身份進行統一整合，並根據每位使用者行為與評價計算出的貢獻度，來幫助系統用戶選擇值得信任的服務提供者。過往的全域化聲望模型存在公平性不足的問題，每個評論者對於同樣的受評者有著不同的評價標準，難以定義出適用於所有人的聲望模型，因此本研究的貢獻度管理模型，以個人化方式呈現，讓用戶能將所見的貢獻度最大程度的符合自身的觀點而不被他人過度影響。為了解決節點的運算資源與儲存空間的大量消耗，本研究透過WeOS網路平台進行分散式運算，將使用者在網路中產生的回饋資訊分配到多個節點上運算，藉此提升更新的效率。而實驗結果顯示，增加節點能夠減少單一節點負擔避免系統過載，且有效的加速更新時間。 ;The WeOS is a web app platform based on P2P(Peer-to-Peer) structure network, allowing users to become service providers and use the services developed by others. With IT technology advanced, fraud is often occured in community networks. This requires a set of criteria to identify trusted users, while users need multiple and secure web services. The introduction of blockchain technology as a platform for auxiliary applications can prevent tampering of activity records and also provide different network services pipeline. But it will lead to cross-system integration problems.
This research provides credit management model for heterogeneous system to the WeOS, a network services platform. Through integrates user identities on cross-system with smart contract and evaluates the credits through each of user’s actions and reputations, the system users will thus have trustworthy service providers. The past global reputation models may not be fair enough. According to each rater has their own scoring criteria for the same critics, it’s hard to define a reputation model that satisfies everyone. Therefore, this research proposes a credit management model to present personalization so that the users can see the credits to their own point of view without being affected deeply by others. According to each rater has their own scoring criteria for the same critics, it’s hard to define a reputation model that satisfies everyone. Therefore, this research proposes a credit management model to present personalization so that the users can see the credits to their own point of view without being affected deeply by others. Personalized reputation model will consume plenty of computing resources and storage space. This research proposes to adopt distribute computing on the WeOS. While the experimental result show that increasing number of nodes can prevent a single node from overloading and accelerate the update time.
