無 / The main purpose of this paper is to find some statistical properties of social networks to enhance further studies on detailed properties of such networks. Except the degree distribution index, we define the efficiency index to achieve our goal. In general, our efficiency index could be applied to social networks with rapid decay in information transmission. Besides, it has several advantages and could help us to analyze the statistical and topological properties of the networks. Furthermore, we find that there exists an optimal process to improve efficiency of the networks and give an explanation about the reason why social planner always chooses the policy of adding one link to any real networks to improve efficiency.

社會網路分析（social network analysis）將社會關係以網路形式表示，從原本純粹分析社會互動的工具，到近年來被廣泛被應用在社會學、組織研究、資訊科學、生物學、語言學等各種領域，藉由引入數學圖學理論與與日益精進的電腦處理能力，使得社會網路分析能從有別於以往的角度找出個體間行動的規律；而詞義相似度（word similarity）是資訊檢索等技術發展的基礎課題之一，近年來對詞義相似度的量測有許多方法的提出。 本研究針對英語字詞利用社會網路分析這樣的工具，藉由提出不同的網路建構方式，以語料庫為資料來源，設定網路節點與連結關係，以共現網路（co-occurrence networks）為基礎，經由改變產生與篩選的條件，觀察以社會網路分析已有的性質或指標做調整，是否可以對詞義相似度提供另一種量測方式；同時以目前詞義相似度研究上已有同義詞標準評比對前述產生的網路與所計算的性質做驗證，並進一步探討使用社會網路分析在詞義相似度研究上的適用性。

當面對總體蘊藏資訊量極為龐大，單筆資訊則較為零碎的資料來源時，社會網路分析同時兼具微觀及巨觀分析特點的方式，提供了一獨特的切入分析角度。目前在社會網路分析領域中，針對網路性質所發展的分析指標等相關理論，大多以1-mode網路為主要的探討模式，對於2-mode網路模式下的著墨則較為稀少。在本研究論文中，以政府人事異動資料庫為主要資料來源，建構出同時具有人物以及職務兩種不同類型節點的2-mode網路，並選擇以適用於2-mode網路模式下的連結預測理論為主要核心，建置出職務接替人選預測系統，其後透過配合不同的實驗模式設計進行接替人選的預測行動，希望能以此來探討單一職位對於其未來接替人選的考量上，受到其歷任人員職務歷程的影響程度。 實驗數據結果顯示，本研究所建置出的接替人選預測系統，對於不同的職務分類層級以及針對不同部門間的職務預測，均會產生不同的預測成效，而這些成果均可適切反映出因應於升遷法制規範或是部份部門於專業知識上的需求條件，因而使得其在對於職位接替人選的考量上，較易受到歷任人選於職務歷程方面的影響。 / When coping with the information source which can store quite high information load in total while one piece of that tends to be fragmentary, Social Network Analysis provides a unique viewpoint to contain analyzing characteristics from both of microcosmic and macrocosmic perspectives. To date, when it comes to theories related to analysis indicator established from different analyzing characteristics in the field of Social Network Analysis, effort is mainly made to explore 1-mode network. By contrast, little emphasis is put on 2-mode network. In this study, the database of government’s personnel change is adopted as the major information source. This study then establishes a 2-mode network with two different types of node, that is, personnel and position. Choosing Link Prediction Theory as the major core with its applicability of 2-mode network, the present study sets up a prediction system of position taking-over candidate. Then, in accordance with different designs of experimental model, the prediction is conducted in an attempt to investigate whether the consideration of future taking-over candidate for a certain position would be influenced by position courses of all past personnel in this position. According to the experimental data, the prediction system of position taking-over candidate established in this study shows different prediction efficiency when predicting different position layers and positions from different sectors. These results all appropriately reflect that the consideration of position taking-over candidate is more easily influenced by position courses of past personnel due to the rules of promotion and the conditions of professional knowledge in some sectors.

小世界網路是目前研究網路結構中一個比較新穎的主題，研究在龐大的網路族群中之各節點間互相連結分佈的現象，例如所有人口的人際關係或者是網際網路中各個網站的連結等等。在此理論被提出之後，讓許多研究領域對於網路結構激發出了更多新的想法和應用，像是網路的路由技術和社群網路的關係聯繫等等。然而我們想知道在如此的架構下，是否能對各個節點作出區域化的分群現象，讓我們能夠獲得更多的應用價值。 在本研究論文中，我們提出了以社會關係與網路屬性為分群及區域化之依據，最後再將此兩種出發點不同的方法結合起來使用。利用這三種方法個別產生的不同結果，探討在什麼樣的資料來源或是應用需求下，使用哪一個方法來分群會最貼切的達到不同的需求，並且就我們的實驗對象做實例說明及適用的應用領域分析。 在實驗方面，我們以政府人事資料庫做為測試對象。實驗結果顯示，本研究所提出的三種區域分群方法針對不同的應用需求下都有不錯的成果。這些方法可以廣泛的應用在各領域的小世界網路之中，針對小世界的網路結構做區域化的劃分，以期使用者能夠從中獲得更多的思考方向以及應用加值。

在網路架構下的 first-price 買賣，加入了中介，這裡建構買家二階段的決策，並且探討買家的 連結數、 valuation 、連結成本、 bid 之間的關係。在本文中，中介對想要向中介連結的買家收取費用，接著找出最適的費用以及與其他變數之間的關係，接著在一些額外的假設下找出其配對方法。最後在不同買家 valuation 分配下發現變數間的關係是一致的。 / In buyer-seller network, on empirical premise, each buyer and seller must have a relationship, that is "link," to exchange a good. Each buyer should choose whether to link or not and choose how many links he or she should have. In this paper, we introduce an intermediary in the buyer-seller network. Buyers can construct links by themselves, or they can exchange through an intermediary. On the other hand, the intermediary also charges an entrance fee for its service. We then discuss the intermediaries' behavior and define the allocation rule. We find the intermediary allocation rule achieves maximal profit for the intermediary but this rule cannot reach the maximal total surplus. We also characterize the relationship between link cost, buyers' valuation, the optimal number of links, and the payoff of the buyer and intermediary.

在傳統資訊系統開發過程中，需求分析通常被視為非常關鍵的步驟。但在Web-based的資訊系統開發中，需求分析卻很少被提及。這篇論文提要主要強調傳統資訊系統與Web資訊系統的差異，並提出一個社群導向系統的需求擷取架構。這個架構將需求擷取分成三個階段：初步分析、關鍵使用者需求分析以及一般使用者的反應。目前Web的開發技術以及社群設計的方法，通常只做初步分析，在系統開發過程中並不直接蒐集使用者的需求。本文提出的三階段架構強調關鍵使用者的重要性，並建議採用社會網路分析作為辦識關鍵使用者的工具。 關鍵使用者區分為資訊、溝通、交易以及娛樂四大類別。初步預試這四種不同類別的關鍵使用者是否會產生較多該類別的資訊需求，以及是否產生較多的需求總數。結果發現只有資訊的關鍵使用者產生較多的資訊需求以及總需求，而其他類別的關鍵使用者與產生需求數量上的關係都不顯著。而另一項有趣的發現是：20％資訊關鍵使用者可以產生大約80％的需求總數，與Pareto規則的預測相似。資訊關鍵使用者的意見是否可以代表全體的意見，預試的結果顯示80％以上的需求重要性評估，關鍵使用者與全體的意見沒有顯著差異。 實地研究採用則探究兩個線上社群，一為關係導向社群，另一則為興趣社群，用以探索使用者角色，涉入程度以及需求知覺間的關係。多變量共變數分析的結果顯示：去除涉入程度的影響後，使用者角色會顯著影響使用者對於需求的知覺，雖然影響的方式以及程度並不相同。高涉入的使用者對於需求的敏銳度，普遍比低涉入者需求要高，並不因為需求種類而有所不同，最後並討論管理的意涵以及日後的研究方向。 / Although the requirement analysis is generally considered a critical stage in traditional information systems development (ISD), but it does not get enough attention in most Web-based information systems development (WISD) or the emerging community-oriented design. The thesis highlighted the difference between ISD and WISD, and proposed a three-stage model of user requirements elicitation for community-oriented design. This model divides the requirements definition in three stages: initial analysis, key user requirements elicitation, and regular user responses. Most current WIS and community design methodologies consider only initial analysis or attempt to build common system architecture, and neglect actual users’ requirements. Key user input is emphasized in this model, and social network analysis (SNA) is proposed as a tool for identifying key users. The pilot study empirically tested the relationship between the number of key users and that of elicited requirements. The study applied SNA to identify key users (defined as their influence) in “information”, “purchase”, “communication” or “entertainment” networks, and then elicited their requirements of two WIS. The preliminary results demonstrated that the number of key users in “information” dimension was significantly correlated with the numbers of elicited “information” requirements and overall requirements. However, the number of key users in “purchase”, “communication” and “entertainment” dimension had no significant relationships with the number of the elicited requirements. The requirements collected from 20 percent of “key users” accounted for approximate 80 percent of total requirements, similar to the results predicted by Pareto’s rule. In addition, the representativeness of requirements from key users opinions was also tested. Two online communities were designed to explore the relationship between user roles, user involvement and users’ perception towards requirements. And the MANCOVA results showed that user role (with user involvement as a blocking variable) has a major impact on an individual’s perception towards requirements, though the difference varies in a certain way. User involvement has also a determining effect on a user’s perception toward each type of requirements. Managerial implications were also discussed.

我們從自然語言領域中借用Thesaurus模型作為字彙關聯的基礎，陸續加入Folksonomy概念、Social Network Service指標的蒐集以及domain-specific ontology來建構Tag-Thesaurus模型，用來解決使用一般tagging system資源彙整能力不足的問題。首先我們對將要實驗的領域選取初始字彙，並利用這些字彙建構Tag-Thesaurus模型。接著將預先準備的這些字彙釋放到社會網路服務平台的tagging system中，透過社會網路服務平台中的tagging system來蒐集使用者對於資源的平面分類資訊，利用這些資訊來對Tag-Thesaurus模型持續地擴充。透過這樣的Tag-Thesaurus模型，我們將可以獲得較佳的資源彙整。domain-specific ontology的加入將可以強化由上而下的資源彙整。而Social Network Service當中的其他資訊，如FOAF[16]或是個人的偏好等，將可以提昇個人化資源彙整的能力。這樣的結合方式不僅是ontology應用的示範，我們更希望透過這樣的混合式模型，使得Web 2.0這樣子廣泛蒐集眾人智慧的概念能夠成為跨入語意網的橋樑。 / We aggregate various resources through the Tag-Thesaurus Model. There are three parts in Tag-Thesaurus model, the Folksonomy formal model, indices collection on Social Network Service, and lightweight domain-specific ontology. The Folksnomy model reconstruct relationships between tags, and we can aggregate resources by tags. The indices collection on Social Network Service help us to decide which resource are more important. Finally, the lightweight domain-specific ontology provide the standard interface to describe the relationships between tags.

在人們的日常生活中，推薦是很普遍的一種社會行為，它使人們不必親自去體驗所有的事物，可透過別人的經驗來得知一件事情或商品的好或壞。隨著科技的快速發展與網際網路的普及，電子商務已逐漸的融入社會，成為人類生活中不可或缺的一部分。然而在網路上過量的資訊，使得個人在資訊的使用與搜尋上面臨極大的挑戰，更加刺激了對於推薦資訊的需求，因此許多推薦技術相繼提出，推薦系統也應運而生，不僅使得推薦的範圍擴大了，推薦的型態也更為豐富多元；同時，在近年電子商務的發展中，對於個人化與顧客導向服務的愈益重視，使得推薦系統逐漸成為一種必要的線上服務。 在眾多的推薦技術之中，協同過濾推薦方法是最成功且最常被採用的推薦技術之一，許多台灣的拍賣平台上也都有採用類似概念的推薦系統，像是Yahoo!拍賣、露天拍賣上的評價機制均屬此類。然而，現行的拍賣評價機制都沒有採用社會網路的技術，本研究希望透過協同過濾與社會網路的結合，讓評價機制更趨於完備。 本研究以台灣最大的拍賣網站Yahoo!為例，蒐集了44萬筆交易記錄，並以推薦網(ReferralWeb)系統的矩陣方法為基礎，找出人與商品的關係、商品與類別的關係、人與人的關係，建立起一個社會網路，讓使用者可查詢特定領域的專家，並與之交易。除此之外，也可直接詢問專家關於商品的資訊或購買技巧。透過這樣的機制，希望能降低消費者在購買商品時所產生的交易糾紛，讓人們在網路上的購物體驗能變得更好。 / Nowadays, recommendation is a common social behavior between people. People can evaluate things or commodities from others’ experience and opinions instead of their own experiences. Along with the development of technology and Internet today, E-commerce has become an indispensable part of human life. However, due to the overloaded information, people face a fantastic challenge when accessing and searching on the Internet. Therefore, many methods of recommendation were proposed, and systems of recommendation are to come with the tide of fashion. In addition, the development of E-commerce emphasized on personalization and customer-oriented services more in recent years, which make recommendation system becomes a necessary on-line service gradually. Collaborative Filtering is the most successful and adopted one in numerous recommendation methods. There are many auction platforms in Taiwan also use recommendation systems, such like "Yahoo Auction", "Ruten Auction", etc. However, the previous mentioned recommendation mechanisms haven’t used Social Network technology; this study will propose an recommendation system which combines Collaborative Filtering and Social Network technology. This research collects 440,000 transaction data from the Yahoo auction platform, which is the biggest auction website in Taiwan. Based on the matrix method of ReferralWeb system(Shah, 1997), this research would like to build up the matrix of relationships between Person-Commodity, Commodity-Category, and Person-Person. Based on the three matrixes, finally builds up a Social Network. In the Social Network, users can enquire experts refer to the specific category of commodity, and then refer to the shops which the experts like or directly ask them the commodity information and purchase skill. Relying on the mechanism proposed by this research, our goals are to reduce the transaction disputes arising from consumers purchase commodities, and to let people have better experiences in on-line shopping.

對於大量零碎的資料，社會網路分析提供了一個可能的方向，可以更容易 的對個體與整體進行觀察，而得以發掘其中隱藏的資訊。本論文嘗試以我 國政府官員異動資料為一零碎資料的代表，利用社會網路分析做為工具， 提出網路模型以幫助我們分析與觀察其中所隱藏的資訊。利用官員的異動 時間記錄建立網路，實驗的模型共可分為人物共同異動網路、人物共事時 間網路，職務升遷網路，人物共同異動網路部分，將每一人物視為不同節 點，節點間的連結則為兩人共同異動的次數，輔以職務間相關性的關係對 資料中的雜訊進行過濾，人物共事時間網路則以兩人在相關單位共事的時 間長度做為其節點間的連結，職務網路改為以職務做為節點，利用人物的 職務升遷，在前後任職的職務節點間建立連結，對於所建立的各種網路， 再利用社會網路分析中一些性質或指標進行分析。實驗結果顯示本研究所 設計的模型對於觀察人物間相互關係以及不同單位所產生的群體間的關係 有所幫助，並提供對於社會網路分析在類似情境下應用的可能性與限制的 了解。

近年來，由於網際網路的興起，網際網路已成為新聞媒體重要的傳播管道之一，許多新聞網站如雨後春筍般的成立，而讀者也樂於使用這類更加便利、高互動性的新聞網站。但是媒體使用網路作為傳播管道，同時也面臨在傳統傳播模式所未遭遇的新挑戰，網路新聞媒體被迫需要創造獨特的內容吸引使用者，也需發展具黏性的社群經營服務，才能與其他具有類似社群互動機制的Web 2.0網站一較長短，留住廣大的使用者群。 本研究嘗試利用新聞為日常生活人們獲得資訊不可或缺管道的獨特優勢，提出一套有效利用新聞使用社群集體智慧(Collective Intelligence）機制，能夠自動化依據使用者顯隱性回饋，針對每篇新聞分析出分歧度、熱門度、話題性三個社群資訊，並以上述三個社群資訊挑選出合適的焦點新聞，以此促進新聞社群使用者對於焦點新聞的討論與互動，進而提昇新聞傳播的效益與新聞社群的凝聚力。實驗結果證實，本研究所提出的機制確實能夠探勘出滿足大多數使用者關注焦點新聞資訊的需求，並且對於輔助記者掌握讀者對於新聞資訊需求及促進新聞社群經營方面都有很大的助益。 / In recent years, due to the rise of the information and communication technologies, the internet has become one of most important communication channel for Journalism. A long with drastically flourished on-line Journalism, models of readers’ information reception changed while they are enjoyed more convenient and interactive websites providing instant information. At the same time, while mass media utilize internet as communication channel, it has also brought unprecedented challenge to traditional communication. On-line Journalism has not only need to create unique content (information) to attract readers; but it also need to develop a more engaging community management services to interact with other communities with similar mechanisms of Web 2.0 sites to retain user’s attention. This study attempts to exam the proposed on-line journalism system for University Press community, which could automatically analyze readers’ community dataset of University newspaper; including opinion deviation indicators, popularity indicator, and topicality indicator of each news (information). This system selects targeted news (information) according to above indicators to promote discussion and interactivity within readers’ community in hope to promote efficiency of news (information) communication and engagement within readers’ community. Experiment results reveal this proposed mechanism could satisfy most readers’ need for headline news; as well as assist Journalists’ understanding on their readers’ need while promoting on-line journalism social networking management.