近年來，智慧型手機成長快速，根據全球市場研究機構TRENDFORCE調查顯示，2013年第四季全球智慧型手機出貨季成長6.5%，達2億六千五百萬支，與去年同期相比成長32.2%。其中小米科技在競爭激烈的市場達成了全球市佔率3%的成績，同時被預估2014年手機出貨上看4,000萬支，取得空前成功。基於此，本研究試圖從使用者參與、口碑行銷與傳遞，以及被廣泛稱為「飢餓行銷」的限量銷售的手法等三個層面對小米所採取的創新作法進行探討，試圖瞭解小米如何在創立至今四年多的時間達成其餘競爭廠商望其相背的成就。 研究結果發現，小米科技發展初期，尚未發行手機硬體前，口碑的形成源自消費者於論壇討論區大量參與MIUI開發與改進所引起關於服務品質、滿意度以及忠誠度的提升。在手機產品上市後，小米科技除了以較競爭廠商更具性能價格比優勢的手機產品吸引消費者外，亦持續透過論壇與服務經營消費者關係，增進了消費者對於品牌口碑傳遞的意願。本研究透過問卷與訪談發現，在小米口碑傳遞內容部分多為其產品之高性能價格比。對於性能價格比優勢，小米科技則是透過限量的搶購模式有效延長產品話題熱度與生命週期，利用常態性關鍵零組件價格走勢降低製造成本，並以手機產品普及後所帶來之服務收入作為其主要營運獲利模式。 本研究亦建議後續研究可針對限量銷售模式、使用者參與程度與口碑內容的影響等方向進一步進行探討，進而對行銷實務上有所貢獻。 / In recent years, a rapid growth in smart-phone usage is observed. According to a study by market research agency, Trendforce, global smart-phone shipment grew at a pace of 6.5% in fourth quarter of 2013 and reached 260 million units. Chinese brand, Xiaomi, alone captured 3% of the global market share during the most competitive period of smart-phone business. It is expected that its 2014 shipment volume will reach 40 million units. Due to this remarkable achievement, this study will analyze the brand from the perspective of end user behavior, word of mouth marketing and the so called "hunger marketing" where limited editions are used, to explore the success of Xiaomi's strategy and its position in the global market within a four year span. The research result indicates that in the early stage of Xiaomi's establishment, prior to their involvement in hardware, the awareness were already establishing from forum discussion of service quality, consumer satisfaction and loyalty, and the development of MIUI. When the hardware eventually launched, Xiaomi's selling point came not only from its low cost strategy, but they continued to participate in forum studies in consumer service level and naturally created brand viral. Consumer market research shows that leading awareness focuses on Xiaomi's high performance/cost ratio. To build on this advantage, Xiaomi uses a limited volume strategy to create brand viral and prolongs product life cycle. It takes advantage of the declining price trend of key components to lower manufacturing costs and income from the software services to enhance operating profit. Further study is recommended on the benefits and results of limited volume strategy, user participation and word of mouth marketing.

從2007年至今，兩個系統產品華碩的Eee PC及蘋果的iPhone震撼市場，Eee PC更被稱做為破壞式創新。這兩個產品的發展及演化是非常值得研究，本研究欲探討： 1.企業如何打破既有傳統硬體製造商思惟，將硬體、軟體及服務整合為一個系統產品； 2.企業如何與使用者成功的共創軟體。 本研究經過次級資料收集、訪談、個案分析後，研究發現: 1.企業開發創新系統產品時，其創新組織特色為： (1)一個跳脫出主流產品獨立之開發組織(團隊)； (2)將團隊獨立在主流外之後，給予跨部門團隊一個利於互動的開放環境； (3)更獲得高層的支持與參與，以助於產品創新。 2.企業在開發創新系統產品過程中： (1)使用者資訊的取得來自於產品開發者，其具有使用者及開發者的雙重身分； (2)先從定義產品才決定功能； (3)必須創造原型才能讓使用者親眼見到、摸到產品，以獲得使用者回饋改善產品。 然而，華碩在Eee PC上對於共創的努力成果不如蘋果豐碩，乃肇因兩個案之使用者社群特性不同，分別為集體協作社群及消費者生產社群。因此企業必須制訂出利於適合使用者社群之回饋模式、必須與開發人員建立良好的溝通並將過程透明。本研究亦發現，若欲成功與使用者共創，企業必須建構出優於既有平台之解決方案，以打破使用者習慣，才能有效共創。 / From the year of 2007 to the present-day, two systematic products, Eee PC of ASUS and iPhone of Apple, appear in the IT market and shock the world as best examples of destructive-innovation. It is thought worthwhile to take advantage of our understanding of the development and evolution of these two products in the past three decades to investigate: 1. How an enterprise is able to break through the old ways of thinking of traditional hardware makers, and able to integrate hardware, software and services into an innovative system-product; 2. How to create a platform for successful user- developer co-creating. The research method of this paper includes literature review, interview, and case analysis. The research results reveal that: 1. In order to develop an innovative system-product, an independent task force comprise staffs of different expertise should be set up outside the mainstream organization of the enterprise; and the independent task force should be given strong support and participation from the top level leadership of the enterprise; 2. In order to create an environment of user-developer co-creating, the function of the innovative system-product should be designed according to the demand of users; and the information of the users’requirement comes from the party of developer in the course of development; 3. In order to obtain feedback from the users for further improvement, a prototype of the innovative system-product should be made for the users to perform a trial run. When Asus unveiled their Eee PC in 2007, they did try very hard to integrate services and software, and to develop software market through user co-creating, just as the way Apple had done it. But the outcome was and is still not as successful as that of Apple. Both Asus and Apple tried user co-creating, but the outcome is different. The difference of outcome stems from the different characteristics of user’s society, i.e. Mass Collaboration Community, and consumers production society. It is therefore that an enterprise should set up a feedback model suitable for user’s society. The research of this paper also reveals that a successful user co-creating should be based upon a platform which is superior to the existing one, and that the habit of users can be modified (changed) in compliance with the new development of the user co-creating environment.

在知識經濟時代下，企業競爭優勢之來源不再是有形資產，而是企業內部的制度、人力資源、創新等無形資產，即為智慧資本，因此外部資訊使用者所關心的不再只是企業所擁有的有形資產，更包括企業所擁有的智慧資本，但現行會計報表仍偏重衡量與報導有形資產之部分，逐漸喪失正確傳達企業價值之功能，遂興起改革之聲。 目前國內外許多會計學者已積極探討企業價值之來源—智慧資本，希望藉由正確釐清智慧資本之本質，以制訂切合時宜之會計制度，以助企業內部管理其智慧資本，並適切地對外進行智慧資本外部報導，提供有助於外部資訊使用者決策之資訊。智慧資本外部報導之內容應為何呢？為目前學術界積極探討的主題之一，已有多篇相關的學術文章發表，但其多為大原則及大方向之探索性研究，對於智慧資本報導所應包含內容未有較多具體著墨；故本研究以使用者觀點為依據，希望藉由探討外部資訊使用者認為企業報導哪些智慧資本項目有助於其評價企業，以作為企業智慧資本外部報導之參考依據。 本研究利用問卷調查直接蒐集外部資訊使用者與企業之看法，再經由多種統計分析後，發現以下結果： 一、影響外部資訊使用者對企業評價的重要智慧資本有「研發及產品化能力」、「增加客戶購買意願之能力」、「完善的人力資源管理」、「提昇銷售多樣產品之能力」、「拓展新客戶之能力」、「經營團隊品質」、「風險評估與管理能力」、「員工專業能力」等八個因素構面。 二、影響外部資訊使用者對企業評價的智慧資本項目會因產業不同而有所不同。 三、不論是資訊電子業或金融服務業，企業認為影響其評價之智慧資本項目與外部資訊使用者之看法有所差異。 / Intellectual capital — the brain of a company’s employees, their know-how, the processes and customer knowledge that they create — is the new source of competitive advantage in the information age. The source of corporate competitive advantage is no longer the tangible assets of companies. The users of external business reports do not only need information of tangible assets of the firm, but also the details of intellectual capital a firm owns. Our conventional reporting framework only highlights the measurement and reporting of tangible assets. Consequently, the current financial statements do not capture or communicate the true value of the company. Increasing number of users have been issuing calls for the reform of the traditional financial reporting and disclosure requirements to meet the growing need for information on intellectual capital. Numerous papers have been issued on intellectual capital. The scholars have covered such areas as the definition of intellectual capital, the measurement of intellectual capital, and the reform of the conventional reporting framework. What should the corporate intellectual capital report contain? This study seeks to determine what type of intellectual capital information users will need to value firms correctly. We used questionnaire to obtain data from users and firms directly and applied statistics analysis to the data. We found that: 1、Intellectual capital details which can assist users in valuing firms are: the ability on R&D and production, the ability to increase consumer spending, the ability of human resource management, the ability to sell different kinds of products, the ability to increase the number of clients, the quality of managers, the ability to assess and control risk, and the professional skills of employees. 2、Intellectual capital details which affect users valuation of firms are different for each industry. 3、Different Intellectual capital information is used in valuing the firm by the inside and outside users of financial statements.

現今視頻網站佔電腦使用者使用網路時數極大比例，在世界各地亦有越來越取代電視之勢，然而隨著視頻網站之風行，相關著作權問題亦隨之而生。特別是網路上不少行為常介於合理使用模糊空間，關於法律條文與案例事實之合致，常引發爭論。網路之精神著重分享，然而著作權之保護則屬於國家保護之一種排他權，其與分享之精神，某些時候會有所衝突，如何在網路時代運用著作權保護創作人，卻又不扼殺創意、公益使用，應有探討必要。是以本文將針對視頻網站上開相關問題，作一基本探討，並對外國與臺灣在視頻網站相關立法與案例作整理介紹，特別是美國法立法與相關案例，與我國法制、實務判決比較，並分析檢討後，提出具體看法，或可做為我國未來立法走向以及實務判決參考。 本文架構上，第一章為緒論，內容為研究動機與論文大致架構；第二章則就視頻網站現況與商務模式探討，期能鳥瞰全貌；第三章與第四章分別就視頻網站使用者與業者之責任探討，均以美國法與我國法為主，輔以中國大陸法規做比較與得失分析；第五章則論及新科技演進與著作權法制、法院判決之因應；第六張結論，最末章為參考文獻。

在過去傳統劇場展演的藝術型態，因爲欠缺互動，易形成台上的表演與台下觀眾互動不頻繁的隔閡。近年來，結合科技與藝術的數位互動展演成為一種創新的表演形態。但在開發互動展演會有技術難度，完成腳本後可能會因為缺乏彈性和擴充性，重複進行一樣的表演，或是再由專業的劇場與程式相關人員反覆修改需求。本研究為解決上述遇到的問題，研發應用在數位展演上的圖型化領域專屬語言編輯器，讓終端使用者可以藉由階段性的操作完成腳本，也可根據不同需求更新背景或角色函式庫。另外，透過實體與虛擬角色的骨架資料繫結的方式，也解決在整合虛擬與實體平台的互動中，實體裝置收到感測器資料，卻無法直接解析取得裝置名稱的問題。於系統上本研究增加使用者自主控制元件和創作腳本，使虛實互動的創作更加彈性。 / In traditional performance, there is little interaction between actors and their audience. Recently, due to the advance of information and communication technology, art and culture are combined with technology mediated interaction between actors and their audience to create a new style of digital interactive performance. However, there are still many technical barriers in developing digital interactive performance. Essentially, the scripts for such performance are usually fixed, or with limited programmability. As a result, one performance will be repeated played or requires a lot of efforts to modify it. This thesis aims to address the issue of programmable scripts with a visual editor and supporting execution environment. In particular, we focus on helping performance director to develop programmable scripts that enable the flexible interaction between physical characters and their virtual counterparts. Our editing tool allows end-users to write scripts using drag-and-drop elements we developed on top of the Blockly framework. Besides detailed descriptions of the design and implantation of our tool, preliminary evaluation results are also presented.

館藏評鑑是圖書館館藏發展中重要的一環，圖書館透過館藏評鑑了解館藏的現況優劣，包括內容品質、使用情形、使用者的需求差異等層面，作為修正採訪策略或讀者服務的參考。然而基於人力、成本等因素，館藏評鑑常成為圖書館的負擔，對於資源不足的中小型、基層圖書館而言，更是艱難的工作，所以尋求一套相對簡易的方法，以節省人力物力的方式進行館藏的評估，是有其需求的。 因此，本研究嘗試以「品質機能展開(Quality function deployment，QFD)」為基礎，發展一套半自動化的「使用需求導向館藏評鑑系統」，希望以系統化和新的不同評鑑模式來輔助館藏評鑑。本系統所應用之資料來源為使用者問卷調查以及圖書館自動化系統的各種紀錄，透過品質機能展開，以量化的方法將兩者做矩陣式的對應和計算並求得結果，作為輔助館藏評鑑時的參考。 本研究以台北市木柵高工圖書館為對象，研究方法及流程依次為：問卷調查、系統實作分析、訪談。首先以問卷調查法訪問學校圖書館使用者，取得使用者之需求；再將這些需求的數值輸入系統中，與圖書館自動化系統的紀錄合併進行品質機能展開計算；最後，將系統運算分析出來的結果，用訪談的方式訪問木柵高工圖書館員，檢視本系統之有效性與館員對本系統之看法。 研究之發現有以下幾點：1.應用品質機能展開所開發之系統，確實可應用於館藏評鑑，但較適合觀察長期趨勢；2. 不同使用者族群對圖書館館藏重視之層面可能有差異，此差異在本系統中會對結果造成影響，影響幅度由需求差異大小決定；3.無法量化計算的質性使用者需求，仍然受到使用者重視，在館藏評鑑中亦相當重要；4.本研究所開發之系統，有助於資源不足之基層圖書館館藏評鑑作業的推動。 / Library collection evaluation is one of important part of collection development, to understand the goodness or weakness of collection. It includes quality of content, analyzing of use, user needs, etc. But because of lake of staff and funds, collection evaluation becomes a burden to many libraries. To small libraries, it is more difficult. So it is necessary to develop a simple method to save the cost and labors for libraries. This study applies the Quality Function Deployment to develop a half-automatic library collection evaluation system, hoping to help collection evaluation works in a systematic, efficient way. The system in this study uses two kinds of data for computing: questionnaire survey of user needs and library automation system records. By applying the Quality Function Deployment, the system can combine above data and records, and generate analysis of collection and user needs in graphics. Finally, it can help decision making according to this information. This study takes Taipei Muzha vocational high school as the object, the method and procedures would be: questionnaire survey, system developing and analyzing, interview with librarians. The study obtains following conclusions: 1.Applying QFD, this system can help collection evaluation in practical, but it is more suitable to observe long-term trends than short-term; 2. Different group of users concern different needs in using library collections, those differences will affect the result of this collection evaluation system; 3. Some user needs can not be quantitated for system progressing, part of those needs are important in users’ opinions, so in a practical collection evaluation work, libraries can not ignore them; 4. This collection evaluation system can help some small-scale, basic-level, insufficient-resource libraries to accomplish their collection evaluation work.

資訊管理文獻中，對於電子商務環境的顧客資訊滿意度衡量模式甚少探討。目前文獻中的使用者資訊滿意度(UIS)，以及使用者自建系統滿意度(EUCS)兩種衡量模式，主要是適用於「傳統資料處理環境」或是「使用者自建系統環境」。因此，本研究將發展一個適用於「數位產品網路行銷環境」的顧客資訊滿意度衡量模式。首先，本研究探討了網站顧客資訊滿意度(WCIS)的概念性定義，作者從文獻中歸納出初步的顧客資訊滿意度衡量構面與問項，並透過訪談、焦點群體、先導研究等方法來加以補充與調整。再者，本研究也說明了量表問項產生過程、資料蒐集方法、以及純化測量的步驟。作者並運用兩個配額樣本來進行探索性因素分析以及驗證性因素分析，其中嚴謹地檢驗了顧客資訊滿意度衡量模式的信度、內容效度、效標關聯效度、收歛效度、區別效度，以及法理效度。最後，本研究探討了顧客資訊滿意度衡量模式在實務界與學術界的應用方式，並討論了本研究所面臨的若干限制，同時提出一些未來可以進一步研究的方向。作者希望本研究所提出的顧客資訊滿意度衡量模式，未來可以被其他研究人員用來發展網路行銷或電子商務理論。 CHAPTER 1. INTRODUCTION….………………………………………….….…1 CHAPTER 2. DOMAIN OF WEB CUSTOMER INFORMATION SATISFACTION ……………………………………………………………………..4 2.1 The Focus of This Study…………………………………………….…..…….4 2.2 The Impact of E-commerce on the Business Process of DPSPs….……….…..5 2.3 Instruments for Measuring User Information Satisfaction (UIS) and End-User Computing Satisfaction (EUCS)…….……………..……………….6 2.4 The Conceptual Definition of Web Customer Information Satisfaction ….…..9 2.5 The Theoretical Framework for Assessing WCIS……………………….…..12 2.6 Service Quality versus Customer Satisfaction………………………….……14 CHAPTER 3. GENERATION OF SCALE ITEMS……….……………………...16 3.1 Generation of Initial Item List……………………………………………...16 3.2 Pilot Study…………………………………………………………………....17 CHAPTER 4. SCALE PURIFICATION AND EXPLORATORY FACTOR ANALYSIS…...…………………………………………………………..19 4.1 Sample and Procedure……………………………………………………….19 4.2 Item Analysis and Reliability Estimates……………………………………..20 4.3 Identifying the Factor Structure of the WCIS Construct…..…………………21 4.4 Reliability…………………………………………………………………….23 4.5 Content Validity………………………………………………………………25 4.6 Criterion-Related Validity……………………………………………………25 4.7 Reliability and Criterion-Related Validity by Type of Web Site……………..26 4.8 Discriminant and Convergent Validity……………………………………….27 4.9 Nomological Validity………………………………………………………...28 CHAPTER 5. THE CONFIRMATORY FACTOR ANALYSIS OF THE WCIS INSTRUMENT………………………………………………...……………30 5.1 Need for Confirmatory Analysis………………………………………….….30 5.2 Methods……………………………………………………………………....33 5.3 Data Collection for Confirmatory Analysis………………………………….41 5.4 Alternative Models…………………………………………………..……….43 5.5 Criteria for Comparing Model-Data Fit…..………………………………….46 5.6 Checks for Statistical Assumptions………………………………………….49 5.7 Estimation Method…………………………………………………………...50 5.8 Results………………………………………………………………………..50 5.9 Assessment of Reliability and Validity………………………………………55 5.10 The Measurement of Service Quality……………………………………….60 5.10.1 The Development of SERVQUAL and IS-adapted SERVQUAL………61 5.10.2 Refinement of an EC-adapted SERVQUAL…………………………...66 5.11 Research Findings for Confirmatory Analysis……………………………...72 5.11.1 Findings for Question One……………………………………………72 5.11.1 Findings for Question Two……………………………………………74 5.12 Comparison of Underlying Dimensions Between UIS, EUCS and WCIS…………………………………………………………………..76 CHAPTER 6. IMPLICATIONS………….………………………………………..77 6.1 Implications for Practice……………………………………………….....….77 6.2 Implications for Research.……………………………………………………79 CHAPTER 7. CONCLUSION……..………………………………………………81 REFERENCE……………...………………………………………………………..83 GLOSSARY…………………………………………………………………………95 APPENDIX A Measurement of Web Customer Information Satisfaction – Forty-Three Items Used in the Pilot Study…………………………………………...97 APPENDIX B Observed Correlation Matrix of WCIS Instrument in Confirmatory Analysis..……………………………………………………………....99 APPENDIX C Observed Correlation Matrix of Initial EC-SERVQUAL Instrument…………….……………………...………………………………………100 APPENDIX D The LISREL Program for WCIS Model 1……..…………………...101 APPENDIX E The LISREL Program for WCIS Model 2.……….………………...102 APPENDIX F The LISREL Program for WCIS Model 3...………………………...104 APPENDIX G The LISREL Program for WCIS Model 4..………………………...105 APPENDIX H The LISREL Program for EC-SERVQUAL Model 1.…………….106 APPENDIX I The LISREL Program for EC-SERVQUAL Model 2.…………….107 APPENDIX J The LISREL Program for The Structural Model Between WCIS and EC-SERVQUAL Measures………………………………..………………...108 ABOUT THE AUTHOR…………………………………………………………...110 / MIS literature has not addressed the measurement of web customer information satisfaction (WCIS) in electronic commerce. Current models for measuring user information satisfaction (UIS) and end-user computing satisfaction (EUCS) are perceived as inapplicable as they are targeted primarily towards either conventional data processing or the end-user computing environment. This study develops a comprehensive model and instrument for measuring customer information satisfaction for web sites that market digital products and services. This paper first discusses the concepts and definitions of WCIS construct from the literature. The researcher summarizes his findings in a theoretical framework. Based on this framework, the researcher develops a measurement instrument to measure web customer information satisfaction. The procedures used in generating items, collecting data, and purifying a multiple-item scale are described. The researcher has carefully examined evidences of reliability, content validity, criterion-related validity, convergent validity, discriminant validity, and nomological validity by analyzing data from two quota samples. The potential applications for practitioners and researchers are then explored. Finally, the researcher concludes this study by discussing limitations and potential future research. The researcher hopes that the proposed WCIS instrument with good reliability and validity can be used by other researchers to develop and test Internet marketing and EC theories in the future.

Osborne & Gaebler在所著的「新政府運動」(Reinventing Government)一書中，曾引述戴明(W. D. Deming)博士所倡導的「全面品質管理」(Total Quality Management)概念指出：員工的參與才能有效的解決問題，因為員工最了解制度，而且最清楚癥結所在。這樣的提示，雖早被管理者重視且廣泛地運用在組織活動的設計範疇中，不過，就參與效能的評估與改善，仍有不少值得再予強調的課題。本研究既係依此觀點，藉由檢視使用者參與行政管理資訊系統發展過程之現象，提供一個思索的空間。 　　整體研究架構，以檢證Davis所提出的技術接受性模式為基準，同時透過此一模式了解使用者參與在系統使用行為上造成模糊結果的原因，進而提醒行政管理者，在系統發展或執行過程，勿陷入以效率掛帥的功能主義典範窠臼，對於溝通、平等與尊重等新人性主義典範所揭示的概念應予適度關注。也就是說，在民主的工作環境下，應將本身視為組織生活的目的，而非僅有工具性價值；參與效能的評估不僅須針對「結果」而且應顧及「過程」。 　　在上述要旨下，關於研究實體部分，經就文獻分析、實證調查結果的發現與筆者實務經驗，提出以下結論與建議： 　　（一）技術接受性模式，尚可適度詮釋國內行政管理資訊系統使用者的使用行為。惟其適用的分析情境、態度與行為關係之假定，仍有再作深入探究之必要。 　　（二）參與及系統使用行為關係之研究，不僅可透過置入中介變項的權變研究觀點來了解，分析使用者在這二行為間的心理歷程，對於改善參與效能而言，亦是一種可行的途徑。 　　（三）使用者對系統有用性及簡易性的認知，是影響系統使用態度、意向及行為的二大要素。其中有用性的認知更具決定性地位。 　　（四）使用者參與效能的模糊結果，除了源於理論欠紮實、研究模式過於簡化，以及研究方法上的瑕疵以外，研究上未仔細區辨系統使用情境之差異亦為主因。不過，參與究非萬靈丹，無定論的結果，並不代表即是一種錯誤。 　　（五）參與效能的改善，除了管理當局的支持與專家協助以外，有賴從加強系統發展者與使用者的教育訓練著手。 　　（六）自主性的系統使用情境下，使用行為（頻率）始能較正確地詮釋使用者接受程度。 　　（七）未來應重視以系統發展者與使用者互動作為分析單元的研究。 　　至於在行政管理實務上，本研究所欲表達的主要意念是：員工的參與及評估的步驟是缺一不可的。因為，沒有評估的參與，則參與無法發揮實際的效果，充其量僅具象徵性的價值；而僅有評估卻無參與活動，這樣的評估除了可能是膚淺、不切實際以外，在實務上存在的機率亦不大。公部門的決策及管理階層，面對高度層級節制的組織文化，能否擺脫「官大學問大」心態，而以恢宏的胸襟與氣度，善用與因應來自員工的意見，塑造具有參與效能的情境，將是成功地推展各項事務最基本的前提要件。

除了文字訊息，張貼相片也是臉書使用者常用的功能，這些上傳的照片種類繁多，可能是自拍照、風景照、或食物照等等，本論文的研究以影像分析為出發點，探討相片內容跟發佈者間之關係，希望藉由相片獲得的資訊，輔助分析使用者樣貌。 本研究共收集32位受測者上傳至臉書的相片，利用電腦視覺技術分析圖像內容，如人臉偵測、環境識別、找出影像上視覺顯著的區域等，藉由這些工具所提供的資訊，將照片加註標籤，以及進行自動分類，並以此兩個層次的資訊做為特徵向量，利用階層式演算法進行使用者分群，再根據實驗結果去分析每一群的行為特性。 透過此研究，可對使用者進行初步分類、瞭解不同的使用者樣貌，並嘗試回應相關問題，如使用者所張貼之相片種類統計、不同性別使用者的上傳行為、 依據上傳圖像內容，進行使用者樣貌分類等，深化我們對於臉書相片上傳行為的理解。 / Apart from text messages, photo posting is a popular function of Facebook. The uploaded photos are of various nature, including selfie, outdoor scenes, and food. In this thesis, we employ state-of-the-art computer vision techniques to analyze image content and establish the relationship between user profile and the type of photos posted. We collected photos from 32 Facebook users. We then applied techniques such as face detection, scene understanding and saliency map identification to gather information for automatic image tagging and classification. Grouping of users can be achieved either by tag statistics or photo classes. Characteristics of each group can be further investigated based on the results of hierarchical clustering. We wish to identify profiles of different users and respond to questions such as the type of photos most frequently posted, gender differentiation in photo posting behavior and user classification according to image content, which will promote our understanding of photo uploading activities on Facebook.

手機的導航功能是一般人習以為常的生活工具，到陌生地點的路線規劃變成越來越簡單的事，不過目前多數的服務都以地圖方式呈現，資訊陳列和互動設計皆十分視覺導向，造成以聽覺操作手機的視障者覺得難以上手且不友善。統計發現，台灣國內的視障者幾乎天天都有外出需求，卻沒有一款針對他們所設計的導航服務，導致需要外出與獨立行動的視障族群必須透過更繁瑣的準備過程才得以安心的出門。再者，搭配現有導航服務的GPS定位並不夠準確，無法提供視障者所需的近身資訊，東南西北方位與百餘單位的公尺數對明眼人來說都可能是一項挑戰，視障者更是難以消化。因此，本研究透過情境訪查、隨身觀察與專家訪談等質化研究方式，深入探索視障者需求，並了解他們與一般人於行動上的異同。藉由新興的iBeacon微定位技術提供非視覺化的線索，將曲折的旅程分割成一段一段的路線，並在設計過程中融入視障者定向訓練的所知所學，讓他們能以原有技能在街道上安心行走，提升獨立出門的安全性。 / Using mobile apps to help route finding is very common for most of us. However, the visually impaired smartphone user does not benefit from this technology. A tremendous demand for independence and mobility remains unsolved. Unlike us, the visually impaired use their smartphone by auditory sense. But those apps with touch screen are not specially designed for their usage pattern. According to our preliminary study, current navigation service always comes with bad information display and unfriendly interaction design. All of these lead to a very frustrating navigation experience. Through the qualitative research methods like contextual inquiry, shadowing, and expert interview, we’d like to explore how different they are from us on the matter of traveling, and reveal the real need of the visually impaired for mobile navigation. Based on our research finding, we want to propose a navigation app specially made for the visually impaired smartphone user. In addition to GPS base service, we provide hyper-local, meaningful multi-sensory notifications to our user through micro location technology of iBeacon. And we integrate the knowledge they learned from Orientation and Mobility training in the navigation process to create a better walking experience and encourage them to live with independency. The app itself is not only a way finding tool but also an information and experience sharing platform for the visually impaired community. Combining the power of design and technology, we can enhance their cognition to the living environment, and improved the independence & mobility for our users as well. In the end, the visual impaired users will be motivated to explore the world on their own and enjoy the colorful world.