我國保險業自2004年導入要保人數位簽章以來，不僅在推動上面臨困難，網路業務也未能順利拓展，然政府業務及金融交易已成功導入電子憑證的各項應用，未來，如何兼顧保險電子商務及數位簽章的應用發展，將是保險業的重大挑戰。本研究的主要目的在於探討保險業對要保人數位簽章的態度與傾向，以創新採用速率的主要影響因素作為理論分析基礎，探討風險資安因素對數位簽章的採用影響，以及金融保險憑證與政府相關憑證的採用傾向。本研究採郵寄問卷調查方式蒐集保險業對數位簽章的看法，總計回收有效問卷為46份。研究結果如次： 一、 產險業與壽險業皆不會因為風險資安因素而採用數位簽章。 二、 產險業配合政府政策而採用憑證的意願偏低，壽險業配合政府政策而採用憑證的意願較高。 三、 產險業傾向不採用網路保險憑證，但傾向採用網路銀行憑證、網路下單憑證及政府相關憑證。 四、 不論是網路保險憑證、網路銀行憑證、網路下單憑證及政府相關憑證，壽險業皆傾向採用。 我國保險業如擬繼續推展要保人數位簽章，本研究建議： 一、 訂定未取得要保書正本簽名或未採用數位簽章所應遵循的規範，包括傳真簽名及其他身分確認機制，以創造有利於網路業務發展的環境。 二、 擴大要保人數位簽章的應用範圍，包括開放保險業銷售中風險的保險商品，以及授權保險業存取政府資料庫管控核保風險等，以利保險業提供差異化的網路投保服務。 三、 原則上，以政府相關憑證為主，金融憑證為輔的方式推動保險業採用數位簽章，如有推動上的困難，則改以保險憑證為主，金融憑證為輔的方式替代，並開放保險憑證及金融憑證可經由異業結盟互為使用。 四、 基於維持保險電子商務市場秩序考量，保險經代人經營電子商務宜納入保險業管理規範，以建立公平的市場競爭環境，俾利保險業經營網路投保業務，消費者享有更多保費折扣優惠。 關鍵字：數位簽章、創新感知屬性、網路投保、保險電子商務 / Since the insurers employed insured digital signatures in 2004, they have not only faced difficulties in promotion but also expanded online businesses hardly. However, the government businesses and the financial transactions have fulfilled various applications of digital certificates successfully. In the future, how to well develop the applications of e-insurance and digital signatures at the same time will become a significant challenge for the insurance industries. The major goal of this study is to delve into the insurers’ postures and bents toward the implementation of insured digital signatures. The analysis theory is based on primary factors of affecting the rate of adoption of an innovation. It supports to explore the influence of operational risks and information securities upon the use of digital signatures, and the tendencies of the use of financial digital certificates and government-related digital certificates. To collect the required data provided by insurers, this study used mail questionnaire method, totaling the valid questionnaire of the recovery as 46s. The findings are listed as follows: 1. Both the P&C insurers and the life insurers do not adopt insured digital signatures due to operational risks and information securities. 2. Regarding the compliance of the government policies, the P&C insurers incline not to adopt digital certificates. However, the life insurers incline to adopt digital certificates. 3. The P&C insurers incline not to adopt insurance digital certificates, but incline to adopt banking digital certificates, stock-dealing digital certificates and government-related digital certificates. 4. No matter what kind of digital certificates are, including insurance digital certificates, banking digital certificates, stock-dealing digital certificates and government-related digital certificates, the life insurers incline to adopt any of them. If the insurance industries would like to continue to advance the development of insured digital signatures, the propositions are listed as follows: 1. In order to create a beneficial environment for the development of online businesses, this study suggests developing directions for the signatures by fax and the other signer authentication mechanism. 2. In order to let the insurers offer differential online insurance services, this study proposes permitting the insurers to sell middle-risk insurance products and access government databases for underwriting risk management. 3. In principle, it is preferable that the insurers employ government-related digital certificates and financial digital certificates. If hardly, the insurers employ insurance digital certificates and financial digital certificates instead. Moreover, both digital certificates have reciprocal usage for each other by means of strategic alliance from different financial businesses. 4. Based on the consideration of keeping e-insurance market order, this study propounds that the e-insurance of brokers and agents are incorporated into directions for the e-insurance of insurers in order to establish a fair market competition environment. It is helpful to insurers’ online businesses and consumers’ preferential premiums. Keywords: Digital Signatures, Rate of Adoption of an Innovation, Online Insurance, e-Insurance

行動時代的到來為社群媒體帶來了發展的春天，原生廣告也隨之應運而生，成為備受社群媒體青睞的新興廣告形式。而在中國市場，WeChat與Weibo這兩大社群媒體自2012年起相繼推出了資訊流廣告服務，並不斷試水新的技術和表現形式。兩者的資訊流廣告在運作機制和表現形式上雖有所雷同，但仍各有特點，其中最主要的差異體現在廣告訊息的按讃者和按讃數量上。 因此，本研究以WeChat和Weibo為研究對象，援引推敲可能性模式理論，希望探究消費者在觀看WeChat和Weibo資訊流廣告時，是否會因為廣告訊息按讚者或按讚數量的不同，在品牌態度和購買意願上產生差異。 本研究採用實驗室實驗法，招募60名國立政治大學非傳播相關科系的大陸籍學生進行實驗。研究結果發現WeChat和Weibo的資訊流廣告訊息接收者在品牌態度和購買意願的形塑歷程上確實存在差異：就WeChat而言，廣告訊息可信度更能正向影響消費者的品牌態度和購買意願；就Weibo而言，感知按讃數量更能正向影響消費者的購買意願，但對品牌態度不構成顯著影響。 / Native advertising has emerged both as an exciting new way for digital marketers to engage with the consumer, and as a new source of advertising revenue for social media. In the China market, news feed ad, a major type of native advertising, has been very popular in WeChat and Weibo since 2012. Visually the native advertisement in these two social media seem pretty similar, but they have enough differences to make them different, particularly in the differences of likers and the number of likes on advertisement. Using the Elaboration Likelihood Model (ELM) adopted by Harkins and Petty, this study focus on WeChat and Weibo and try to clarify the differences of consumer’s information processing, brand attitude and purchase intention between these two social media, which has various social tie strength. Adopting an laboratory experimentation, a total of 60 samples, participated in this study during January, 2017. Finding of this study show the differences of brand attitude and purchase intention towards news feed ad when subjects are exposed to WeChat between Weibo. In WeChat, message credibility indicate more significant influence on the brand attitude and purchase intention towards native advertisement rather than perceived number of likes. Oppositely perceived number of likes positively influence on purchase intention towards native advertisement in Weibo.

在互動敘事(Interactive Storytelling)的應用中，適當的攝影機規劃能幫助我們呈現虛擬世界中所發生的故事。在本研究中，我們設計了一個可以讓使用者設計互動敘事的應用平台－IMStory。針對故事創作，我們提供了一個故事腳本的描述語言及其剖析器，當使用者撰寫好互動故事之後，系統便能夠根據故事情境內容及與使用者互動的結果產生出符合情境的自動攝影機規劃系統。在此系統中，我們建立了一個能夠讓攝影機理解故事腳本的關係模型，建立虛擬攝影機組態與「故事情境」間的對應關係，即時找出更能表達當時情境的拍攝方式與架設位置。最後，我們設計了一個場景劇本進行實驗，在實驗中，我們的系統能夠即時因應場景中角色物件的位置作出即時的攝影機規劃，並且能夠配合互動環境中改變的情境參數而使用適合的攝影模組，產生出符合攝影學法則且賦有敘事能力的鏡頭。透過我們的方法，希望能提供一個更加有敘事能力的互動敘事平台。 / In an 3D interactive storytelling system, virtual camera is the key for us to understand what happens in the virtual world. In this thesis, we have designed an intelligent camera planning system for interactive storytelling, in which story plots are influenced by the user’s actions. Since the story flow cannot be determined in advance, it is essential to utilize real-time camera control to express the story for given spatial relationships between characters and objects. A script language and its parser were designed for our storytelling system (IMStory) to describe events in the story as well as the contexts associated with the events to model the properties of a story for camera control. We have developed a camera planning system by mapping contexts parameters into appropriate camera control parameters. We have implemented the 3D interactive storytelling system and demonstrated its capabilities with several examples. The camera planning system is shown to perform in real time, and the generated shots are more expressive in establishing the relation between camera configurations and story context.

在無線網狀網路(Wireless Mesh Network)中，每個節點須幫助相鄰節點轉送資料及提供使用者網路存取，例如WLAN(IEEE 802.11s)、WMAN(IEEE 802.16)等，皆可利用多跳接方式將資料轉送至通訊閘道器(Gateway)。在無線網狀網路中，常利用密集佈建的方式來解決通訊死角的問題。當網路節點的密度增加時，無線訊號的干擾也會增強，並且各節點的效能會顯著下降。 在本研究中，將利用幾何學概念，解決網路干擾問題，並提出拓樸重建演算法來重建路徑，使網路干擾達到最小化。我們試著最小化節點與節點間的干擾，以提升整體無線網狀網路效能。我們將網路問題轉換成幾何問題，並定義在幾何圖形中線段交錯問題，之後驗證在幾何圖形中是否有線段交錯的現象發生。若發生線段交錯時，則將此線段從幾何圖形中移除，並且利用三角化演算法將此區域線段重新規劃，使相鄰節點間的干擾最小。當網路拓樸建立完成後，我們利用標準差公式將干擾較大的連線移除，使得網路效能提升。上述測試線段交錯及三角化多邊形演算法可在時間複雜度O(n log n)內找到干擾最小的解。最後，我們將利用網路模擬器(Network Simulator)驗證所提出的方法是否能達到預期的系統效能指標。 / In wireless mesh networks, such as WLAN (IEEE 802.11s), WMAN (IEEE 802.16), etc., each node should forward packets of neighboring nodes toward gateway using multi-hop routing mechanism. In wireless mesh network, as the density of network nodes increases, the RF interference will increase and the throughput of each node will drop rapidly. In our research, we use the geometry to resolve the RF interference problem by rebuilding network topology. We try to minimize the interference between neighboring nodes and improve the throughput in wireless mesh network. We transform the network topology problem into geometry problem and define the line intersection problem in geometric graph, then check path intersection in the geometric graph. If line intersection occurs in the graph, we remove the intersection line from the graph and re-plan the region by triangulation algorithm. When the network topology is built up, we use a standard deviation formula to improve network performance by removing longer links. The line intersection algorithm and triangulation algorithm, both of time complexity O(n log n), are used to find the minimal interference solution. At the end of our research, we use network simulator to verify if the proposed methods can help to meet all those performance expectations.

在這份研究中，我們提出一個類分散式神經網路框架，此框架為高頻交易系統研究下之子專案。在系統中，我們透過資料探勘程序發掘財務時間序列中的模式，其中所採用的資料探勘演算法之一即為類神經網路。我們實作一個在分散式平台上訓練類神經網路的框架。我們採用Apache Spark來建立底層的運算叢集，因為它提供高效能的記憶體內運算（in-memory computing）。我們分析一些分散式後向傳導演算法（特別是用來預測財務時間序列的），加以調整，並將其用於我們的框架。我們提供了許多細部的選項，讓使用者在進行類神經網路建模時有很高的彈性。 / In this research, we introduce a distributed framework of artificial neural network (ANN) as a subproject under the research of a high-frequency trading (HFT) system. In the system, ANNs are used in the data mining process for identifying patterns in financial time series. We implement a framework for training ANNs on a distributed computing platform. We adopt Apache Spark to build the base computing cluster because it is capable of high performance in-memory computing. We investigate a number of distributed backpropagation algorithms and techniques, especially ones for time series prediction, and incorporate them into our framework with some modifications. With various options for the details, we provide the user with flexibility in neural network modeling.

市場導向指的是一種辨識市場狀態和顧客需求的企業理論，可以藉由不同的市場情報蒐集來達成。由此，其中的「市場感知能力」和「市場連結能力」能夠幫助企業不斷地修正、改進、創新及再定義原本的市場觀點，並研究市場中所有角色之間的關係。本研究致力於如何以文字探勘、資料探勘及機會發掘等理論創造一個服務的資訊系統來提升這兩個能力。運用本系統可以協助企業在不同市場中找到潛在資源及合作夥伴以共創複合性產品。換句話說，以這種開放關係和不同公司的共創方式更能達到相互曝光、產品與服務互補的效果。而更好的市場導向可以幫助企業間創造出更好的複合式產品並達到利基市場的競爭優勢。 / Market orientation refers to the business philosophy that focuses on identifying and meeting the stated or hidden needs of customers through various approaches of intelligence gathering activities. To this end, two important capabilities, market sensing and market relating, are to enable business to formulate, examine, modify, renovate and redefine their market views and investigate among all players in the market. This study focuses on how to exploit information system as a service to facilitate the acts of the market sensing and market relating capabilities in terms of information technologies of the text mining technique and the chance discovery theory. In addition, this service would help various businesses seek new, necessary and related resources or fellow companies for each other to cooperate to form a complex service product from different markets. In other words, cooperating in the open relationship, different companies attain more possibility to mutually expose their features, product or service, which achieves a marketing for those companies in different markets as well, and better market orientation is also avail business to make better complex product or service for niche market to get competitive advantages.

N世代青年族群是開始以網路為主，資訊傳遞翻轉的新興世代，本研究藉由資訊-動機-行為技能模式（簡稱IMB模式）瞭解影響青年族群正確用藥行為的因子，探討青年族群醫藥素養、個人態度、用藥自我效能、行為障礙感知、醫藥傳播資訊暴露程度及正確用藥行為的發展現況，藉此發現喚起病人或家屬對病人安全及醫療風險認知的切入點，並增加警覺及主動積極性行為，與醫藥人員共同合作。為自己的健康狀態把關。 本研究所採用的方法是網路問卷調查，針對1977年到1997年出生的N世代青年族群蒐集資料，共獲得1,198份有效問卷。 本研究發現青年族群整體的用藥行為偏向經常做到，其中實踐程度由高至低依序為堅持正當藥品取得方式、主動告知身體狀況、看清楚用藥標示和主動諮詢專業。青年族群九成九以上有高醫藥素養，有六成以上的受訪者得到滿分。大部分青年族群偏向「同意」有正向的用藥態度；自我效能整體的認同度偏向於「同意」。受訪者對於用藥行為障礙感知程度整體偏向「不同意」有用藥行為障礙。 多數青年從媒體獲得醫藥資訊的頻率偏向「每月1~2次」為主。整體面向暴露由高至低為專業管道、傳統大眾媒體及新興網路。青年族群前三大醫藥資訊暴露管道依序為Facebook、電視、醫護人員，而BBS、大眾運輸工具看板、海報、傳單、宣傳手冊、產品包裝等是接著依序常用的管道。 醫藥素養只對堅持正當藥品取得方式有統計顯著正向預測力。個人態度和自我效能與用藥行為的四項依變項主動告知身體狀況、看清楚用藥標示、堅持正當藥品取得方式和主動諮詢專業，在相關分析及階層複迴歸分析都顯示有統計顯著正向相關。行為障礙感知與用藥行為的四項依變項主動告知身體狀況、看清楚用藥標示、堅持正當藥品取得方式和主動諮詢專業，在相關分析及階層複迴歸分析都顯示有統計顯著負向相關。 傳統大眾媒體與主動告知身體狀況、看清楚用藥標示和主動諮詢專業，在相關分析顯示有統計顯著正向相關，但階層複迴歸分析都未有統計顯著預測力；堅持正當藥品取得方式在相關分析及階層複迴歸分析都是顯著負向相關。新興網路與主動告知身體狀況和主動諮詢專業，在相關分析及階層複迴歸分析顯示都有統計顯著正向相關；新興網路與看清楚用藥標示，在相關分析顯示有統計顯著正向相關，但階層複迴歸分析未有統計顯著預測力；新興網路與堅持正當藥品取得方式在相關分析及階層複迴歸分析都是顯著負向相關。專業管道與主動告知身體狀況、看清楚用藥標示和主動諮詢專業，在相關分析及階層複迴歸分析顯示都有統計顯著正向相關；專業管道與堅持正當藥品取得方式在相關分析是顯著負向相關，但階層複迴歸分析中是無統計顯著預測力。 本研究的主要貢獻是證實藉由IMB模式的各項因子，都能有效預測部分或全部的正確用藥行為，尤其是個人態度、自我效能及行為障礙感知這三面向，並且呼應行為技能是影響行為的重要關鍵。因此往後如果有醫藥行為要推廣或宣傳時，就可以藉助IMB模式的因子為原型、大綱，相信會有更好的效果產生，也希望有人可以做後續相關研究，證實IMB模式的可行性。 / The Net Generation is the first generation to grow up in the digital world. The goal of this study is to understand the medical behavior of the Net Generation using the information-motivation-behavioral skills (IMB) model and the conditions of health literacy, personal attitudes, medical self-efficacy, the perception of behavioral barriers, the level of media exposure and medical behavior. To emphasize patient safety and medical risks, we tried to find the way to improve the awareness and active behavior to cooperate with medical professionals and to maintain self-health. The 1,198 effective samples were collected from people who were born between 1977 and 1999 through a web questionnaire. The study found that the Net Generation usually had correct medication-taking behavior. The frequency of the behavior from high to low was obtaining medicine through official channels, actively telling self-conditions, clearly reading medication labels, and actively consulting medical professionals. Over 99% of the participants had high health literacy and over 60% got full scores. Most of the participants tended to agree that they had positive personal attitudes and self-efficacy but disagreed they had the perception of behavioral barriers. Most of the participants obtained medical information through media once or twice a month. The exposure to media from high to low was professional channels, traditional mass media, and internet. The exposure to media channels in order were: Facebook, television, medical professionals, BBS, public transportation boards, and advertising brochures. Health literacy could only positively predict actively telling self-conditions. Personal attitudes and self-efficacy could positively predict four dependent variables of correct medication-taking behavior including obtaining medicine through official channels, actively telling self-conditions, clearly reading medication labels, and actively consulting medical professionals. And, the perception of behavioral barriers could negatively predict them. There were some statistically positive correlations between traditional mass media and correct medication-taking behavior including actively telling self-conditions, clearly reading medication labels, and actively consulting medical professionals. However, there was no prediction in hierarchical regression. There was a statistically negative correlation in Pearson’s correlation and hierarchical regression between traditional mass media and obtaining medicine through official channels. There were statistically positive correlations in Pearson’s correlation and hierarchical regression between internet and actively telling self-conditions or actively consulting medical professionals. However, there was a statistically negative correlation between internet and obtaining medicine through official channels. There was a statistically positive correlation between the internet and clearly reading medication labels but there was no prediction in hierarchical regression between them. There were statistically positive correlations in Pearson’s correlation and hierarchical regression between professional channels and actively telling self-conditions, clearly reading medication labels, or actively consulting medical professionals. There was a statistically negative correlation between professional channels and obtaining medicine through official channels but there was no correlation in hierarchical regression. The main contribution of this study was to support the argument that the IMB model could predict correct medication-taking behavior, especially when personal attitudes, self-efficacy and the perception of behavioral barriers are looked at. Therefore, behavioral skills were the key to correctly using medication. If someone wants to do medical behavior promotions, then the concept of the IMB model could be very effective.

This study examines the perception of Mandarin lexical tones by native speakers of Greek who are naïve to Mandarin. Being able to discriminate between Mandarin lexical tones is essential for effective and accurate communication in the language. While there is an increasing number of research studying the perception of Mandarin tone by speakers from various backgrounds, the Greek language has not received much attention. By employing an AXB discrimination task, this study tested the perception of Mandarin tone pairs T1-T2, T1-T4, and T2-T3 by native speakers of Greek. Due to their acoustic similarity, these three pairs are often found to be the most confusing tone contrasts for non-native listeners of Mandarin. Greek speakers (NG) had the highest accuracy for T1-T2 (0.88), and similar accuracy rates for T1-T4 (0.83) and T2-T3 (0.82). Subsequently, the Greek speakers were divided into two groups based on their experience with learning or playing a musical instrument, as this is a factor that has been shown to affect tone perception. Compared to non-musicians (NG1), the group of musicians (NG2) had higher accuracy rates for T1-T2 and T1-T4, and a similar accuracy rate for T2-T3. For NG1, from easiest to hardest, the accuracy rates were: T1-T2(0.93)>T1-T4(0.92)>T2-T3(0.82). For NG2, the accuracy rates were: T2-T3(0.82)≥T1-T2(0.82)>T1-T4(0.74). These findings could improve our understanding in regard to the factors that can affect the perception of Mandarin lexical tone by native Greek speakers, which might contribute to language pedagogy. / 本论文旨在研究以零基础普通话希腊语母语人士对普通话声调范畴的感 知能力。为了进行有效和准确的研究，本文实施了普通话发音辨别实验。虽 然声调范畴感知能力的研究越来越多，但以希腊语母语者为研究对象的研究 并不多见。通过 AXB 发音辨别测试，本论文调查了希腊语为母语的人士对普 通话 T1-T2、T1-T4、T2-T3 三组声调组合的辨识感知。由于声音的相似性， 这三对声调组合对非普通话母语的听众经常被发现是最容易被混淆的。结果 发现，讲希腊语的人士（NG）对这三组声调组合辨识的准确度为：T1-T2 （0.88）是最高的，T1-T4（0.83）和 T2-T3（0.82）是相似的。有研究认为 音乐训练可以影响对声调的感知，所以接着实验参与者根据学习或演奏乐器 的经验又被分成两组。与非音乐背景人群（NG1）相比，有音乐背景人群 （NG2）T1-T2 和 T1-T4 的准确率更高，但对 T2-T3 组合的结果是一样的。若 从最简单到最难排序，有音乐背景人群（NG1）的准确度是：T1-T2 （0.93）>T1-T4（0.92）>T2-T3（0.82）。而无音乐背景人群（NG2），准 确度是：T2-T3（0.82）≥T1-T2（0.82）>T1-T4（0.74）。这些发现可以帮 助理解以希腊语为母语的人对普通话声调感知困难的根源，所以对语言教学 法的相关讨论会有所贡献。

本論文出自於對於康德道德哲學中同時兼具先驗性與實踐性之二重關係的研究興趣，因而於此藉由釐清康德哲學中的崇高情感與道德法則之間的關連是如何被談論的，嘗試以此整理出一些在理解上的關鍵線索。在這條主軸的引導之下，此處採取對其所涉及的相關文獻研讀與理解作為基本的脈絡開展。本論文並依從前批判期與批判期的區分將概略畫分為兩個部分︰ 第一部分︰一開始即企圖從對康德批判期美學談論的背景進行文獻上的考察，從中導引出對於康德美學與其形上學內在本有的關連，以及對絕對者/超感知者對康德方法論上的影響。接著審視《觀察》作為此時期思想實驗的產物，又是如何在康德自身的評判(即其《評著》)居於一個重要轉向的地位，在其中康德將美感與崇高感與德行關連起來的道德情感論述方，何以導向對主體意識能力的考察，而自由─自我意識與道德法則之間的關連也在這當中被觸及。 第二部份︰主要關注於康德於《基礎》與《第二批判》中「對法則的敬重」的兩重性所關涉到的一種特殊主體意識活動─即「道德意識」─的談論，其具體意涵又何以關連到康德道德性之理性奠基的先驗要求︰「終究只有一個理性」。最後以康德在《第三批判》中對反思判斷力的重新修正、予以審美經驗先驗奠基之架構，重新檢視康德如何以審美判斷─特別是「崇高判斷」─的不確定性概念關連到主體意識能力的提升；並藉由數學崇高與構想力、力學崇高與理性使命之間關係的建立與理解，繼而以此去討論，主體之特有的崇高判斷與道德法則之間本有的奠基性關連，並且何以生命情感作為一種道德意識的洞察能力，可以進行一個既是普遍有效、且與自然達到真正和諧關係的「擴延」；而從這當中誕生出來的道德情感，便根源於意志自由而作為道德行為的發生的關鍵推動力、但其自身卻仍不能作為行為之根據。