• 學位論文

A study of relational bonding strategies across virtual and physical channel

The goal of relationship marketing focuses on development of loyalty and maintenance of long-term relationship based on cooperation and partnership with key customers. Although studies have discussed the types of relational bonds that enhance customer loyalty, empirical research that focuses on the application of relational bonds in benefit and barrier is relatively sparse. In recent years, e-commerce is a blooming business, and Web 2.0 which increases interaction between customers and firms on Internet is released. Beside, many companies operate in virtual and physical environment simultaneously, and the issue of multiple channels management is more and more important. We attempt to explain the relationships among relational bonds, two drivers, and customer loyalty in virtual and physical channel. 關係行銷的目的在於與重要顧客建立忠誠度並維持長久合作的關係。在過去的文獻中，大多將重點放在不同的關係結合策略如何的提升顧客忠誠度，而將焦點放在關係結合策略中如何增加利益及障礙來留住顧客的研究是很少見的。近年來，Web2.0的普及，以及企業多重通路管理議題越來越受到重視。因此，本研究想解釋在虛擬與實體通路，關係結合策略、利益與障礙和顧客忠誠度之間的關係。

• 學位論文

Using Trust Prediction Mechanism to Support Trust-Based Recommendations

隨著網際網路的興起與蓬勃發展，使用者能夠便利地從網際網路中搜尋物品，但是大量的物品資訊卻也使得使用者難以消化，使用推薦技術便是克服這個資訊超載問題的一項解決之道。推薦技術透過建議使用者可能喜歡或感興趣的商品來加快搜尋商品的速度，對企業而言，也能夠藉以促進更多的消費與電子商務網站的使用率。近年來由於Web 2.0技術的發展，以信任關係為基礎的推薦技術正逐漸興起，透過信任關係的分析，參考信任值高的使用者的意見來進行評分預測。然而這類方法在面對信任網路稀疏性的問題時，必須要透過信任傳遞的方法來擴充信任網路，其主要的限制在於在信任網路中沒有任何路徑相連的兩個使用者永遠沒辦法產生任何的信任關係，這將限制了這類方法的效能。為此，我們建立了一套信任預測機制來預測所有還沒出現過的信任關係的信任值，亦即是利用結構性預測器在信任網路中所產生的變數以及機器學習演算法來建立一個信任預測模型，我們採用了Epinions.com所提供的數據為實驗的資料集合，並與傳統的協同過濾推薦技術與採用信任傳遞之推薦技術相比，結果顯示我們的方法的確能夠提供比較高的預測涵蓋範圍(coverage)，同時，在比較相同預測目標的情況下，我們的方法也能提供比較高的預測準確度(prediction accuracy)。

• 學位論文

Incremental Clustering: An Example of Legislative Interpellation

The Parliamentary Library of Legislative Yuan website provides a fair and objective channel for the public to track daily activities of the Legislative Yuan and legislators’ inquiries. However the increased information content cause information overloading problem. To mitigate this program, this study proposed an incremental clustering mechanism to renew the information regularly and transform information from text to statics. This study first initiates a basic categorical structure by two-stage clustering algorithm. Then the incremental clustering method is applied to group related documents corresponding to the same topic into clusters and designates these clusters into existing category or create a new category. Experimental results show the effectiveness of that the proposed incremental clustering method, which enables the management of hierarchical categorical structure on legislative interpellation. With this results, people can track the legislative activities using the information from the Parliamentary Library of Legislative Yuan to recognize the interpellations in each category.

• 學位論文

實踐策略，改變制度：工研院與創新前瞻機制的個案分析

本論文從一個成熟的組織場域出發，根據制度興業的理論觀點，同時採用策略實踐的分析方式，探討核心行動者（興業家）從事改變的動機以及如何運用策略，改變其所處的制度環境。基於提問的是「為何」以及「如何」的解釋性研究問題，引導本文植基於自然探究法的研究精神，實徵上，透過縱向長期的研究途徑，執行歷史導向式的單一個案研究，蒐集大量的初級與次級資料，探討工研院在1987年到2009年間，推動創新前瞻計畫，跳脫傳統科專計畫場域，促成台灣科技專案體系轉型的歷程。本論文發展出一個奠基於行動構面之「湊組–連介–溝通」策略模型與時間構面之「創新期–擴散期–正當期」所交織而成的分析架構，而此分析架構被用來描述與分析本文的個案故事。最後，從理論層次討論制度興業的動機、策略以及制度化的演化模式。

• 學位論文

快速變動的網路市場的服務創新機會-以中國瑞可利為例

本個案研究主要是透過訪談整合文獻與資料彙整，針對日本RECRUIT公司在中國事業推進上，面臨到迅速變遷的中國網路市場，日本RECRUIT公司的兩大支柱人力資源領域、情報產業前進中國市場面臨許多的挑戰，無論是實體對中國市場的生澀，抑或是網路虛擬世界的快速競爭，都是日本RECRUIT公司的課題。本個案研究以RECRUIT中國三大核心事業中的結婚情報誌ZEXY與胡椒蓓蓓Hot Pepper兩間公司各自在網路事業中面臨的挑戰與困難， RECRUIT是台灣艾思網絡公司的投資公司，由於艾思網絡在網路市場的經驗，故RECRUIT委託台灣艾思網路進行顧問諮詢，分析網路競爭者的競合關係，以觀察市場狀況提出策略規畫，研究的重心，首先中國婚慶產業與飲食優惠券的網路市場分析、日本RECRUIT的核心價值、中國RECRUIT的網路事業挑戰…等等。重新分析網路市場定位，進行策略規畫，以檢視組織核心，佈署相關的資源，建構網路服務優勢地位。

• 學位論文

以預先分群為基礎之情境式文件分群

Document clustering is an intentional act that should reflect individuals’ preferences with regard to the semantic coherency or relevant categorization of documents. Thus, an effective document clustering must consider individual preferences and needs to support contextual document categorization. In order to support context-aware document clustering, Yang and Wei (2007) proposed a Context-Aware Document Clustering (CAC) technique that takes into consideration a user’s categorization context represented as a list of anchoring terms and a statistical-based thesaurus constructed via exploiting the Web, and then generates a set of document clusters from this particular preferential context of the user. However, the thesaurus construction in the CAC technique is time-consuming and requires a lot of computational resources. Moreover, the thesaurus is built based on Web pages belong to various domains and may result in including too general terms for anchoring term expansion. In response to the limitations of the CAC technique, we proposed a new approach for context-aware document clustering. Specifically, those anchoring terms provided by a target user is first applied to form some partial clusters which are then adopted to consolidate a set of representative features for clustering the document corpus. Due to the nature of preclustering partial documents in the corpus, we called the proposed approach a preclustering-based context-aware document clustering technique. Our empirical evaluation results suggest that the proposed technique achieves better clustering effectiveness than its benchmark techniques (i.e., the traditional content-based approach and the CAC approach).

• 學位論文

考慮商品異質性改善協同過濾推薦系統

在今日電子商務與知識經濟盛行的環境中，網路上大量流通的資訊成為我們主要 的資訊來源管道，但資訊使用者卻也面臨了資訊超載的挑戰，因此必須尋求可以 協助過濾與挑選商品的篩選機制。同時，線上商家為了提高顧客滿意度與忠誠度， 也需要可以協助加強顧客關係維護的機制。推薦系統便是因應資訊超載與加強顧 客關係維護而產生的一種解決之道，有了推薦系統的服務，資訊提供者便可以適 時地提供合適的產品給使用者，同時使用者也可以免去因面臨過多商品資訊而無 法做出決策的困境。在眾多已發展的推薦技術中，協同過濾推薦機制 (collaborative filtering approach)是其中最成功且最廣受採用的推薦技術， 但是這項推薦技術卻因為沒有考慮到受推薦的商品彼此間的差異性，亦即是當計 算使用者相似度以及進行使用者喜好預測時，所有的商品評分都被視為是同等重 要的，如此一來可能導致參考不可靠的使用者意見，以致無法達到理想的推薦效 果。因此，在本研究中，我們提出一個考慮個別商品內容差異性的協同過濾推薦 技術(item-based content-weighted collaborative filtering approach)，藉 以解決傳統協同過濾技術不考慮商品異質性的缺點。實驗結果顯示本研究所提出 的方法的確可以在考慮內容差異性的情況下，提供比其他兩種協同過濾方式更高 的預測準確度(prediction accuracy)，同時，為了進一步提升可預測的涵蓋範 圍(coverage)，我們也提出一個結合傳統協同過濾技術的混合方法(hybrid approach)，其實驗結果也顯示所提出的混合方法，可以達到與傳統協同過濾技 術相同的可預測的涵蓋範圍，而仍然有較佳的推薦效果。

• 學位論文

產品經理在組織內跨部門影響策略探討

近年來，產品經理的概念從原本運用於消費性產品之產業延伸至許多高科技產業、醫療產業等，因產品經理可以專注於產品之經營，故許多產業也開始採用產品經理之制度；另外，也越來越多學生、社會新鮮人對於產品經理的工作感興趣，希望畢業後投入相關的工作。但產品經理制度於不同產業皆會有不相同之定義，要完全的了解不同公司之產品經理之任務並不容易。 本研究由產品經理需跨部門溝通協調的角色為出發，探討在與跨部門其他人員溝通與發揮影響力時，是否採用特定的影響策略來取得其協助與支持。本研究透過主要權力之兩大方向、三個構面來討論，其主要權力為產品經理之個人權力，個人權力之兩大方向為參考權與專家權，而三個構面就包括外交型策略、強迫型策略(此兩項屬參考權)與專業能力(專家權)，探討產品經理在影響跨部門人員時採用之策略。本研究透過個案公司不同部門數位產品經理之實際訪談，了解在個案公司中產品經理之相關制度、任務，並了解產品經理應具備之能力；再者，分析並討論其產品經理應用不同影響策略之情境，最後歸納成一產品經理跨部門影響系統與影響產品經理採用影響策略之相關因素。 本研究透過訪談分析整理後得到，就在三個構面而言，產品經理對於產品之專業能力是最具有影響力的，在影響策略的應用方面，有90%以上產品經理之跨部門溝通都是採用外交型策略，即可達到其他部門協助與支援之目的；另外，產品經理採用強硬型之影響策略，會因一些其他部門特別要求而採用或因禮貌性而先訴諸上級，也會因產品經理溝通之成效未達一定程度而採用。

• 學位論文

Co-opetition Network and Organization Performance: The Impact of Network Structure in MLB Players Trading

In the past, the social network analysis has been used to explore the relationship between network structure and organization performance in alliance/cooperation network for many years. In this research, we extended the research to the co-opetition network, where the players in this network cooperate and compete with each other at the same time. The major research object is to examine if the network index that have positive influence on organizations performance in alliance/cooperation network could also have positively influence in co-opetition network. In this research, we selected baseball players trading network as our research target because the nature of player trading with other team is cooperation, but still, the teams have to compete with each other and win more games as possible. Besides of network index, we also consider team payroll’s effect on team performance and its interaction with independent variables. The result of this research indicated not only the payroll of a baseball team has positive influence on team performance, but also influence the relationships between network index and team performance. The result indicated that in this co-opetition network, high payroll teams should involve in trading more frequently, while the low payroll team should lower the frequency of trading players. If we take payroll as different capital, the result could help us to understand co-opetition behavior of firms in different industry.

• 學位論文

The Enhancement of Relation Extraction in Discovering Business Ecosystems for a Technology Monitoring Service

The fast evolution of technology has made an intensive competition environment, the company which can update information rapidly and precisely will take the competitive advantage. However, the information explosion made us have to read all the information. It is time consuming to keep information from various sources updated, and could potentially delay the reaction to the changing technology and business environments. This thesis proposes a technology monitoring service to monitor the evolution of business ecosystem through extracting relations from various sources of information automatically. In order to construct business ecosystem that consist of companies, products and technologies. This study enhances the relation extraction methodology to extract mutual interaction from multi name entity types. The service of the proposed system helps companies forecast the technology trends and market status through business ecosystem and other information extracted through text mining. To reduce the monitoring cost of companies, the system can extract information that companies interested in automatically and effectively catch the information related to the evolution of business ecosystem in real time.