中小企業一向是經濟成長的主力，然而受限其資源的短缺，往往無法提供顧客更好的商品或服務。雖然文獻指出中小企業可以透過資訊科技與夥伴和顧客合作來克服它先天的弱勢並提供更好的服務，但仍有許多模糊未解的問題。例如，哪一種中小企業較願意導入以資訊科技為基礎的協同合作模式(IT-enabled Collaboration)？而這樣的合作模式是否能真正的提升他們的績效表現呢？本研究以資源基礎理論和服務導向邏輯理論(service-dominant logic)為基礎，將資源分為operant資源、operand資源、資源互補性及資源相似性，試圖找出在何種資源模式下，中小企業會偏向使用以資訊科技為基礎的協同合作模式。研究方法採用個案研究，並以宜蘭縣枕頭山休閒農業區的八家中小型業者為目標進行訪談。在資料蒐集和分析之後，本研究發現operand資源與operant資源皆會正向的影響到中小企業對科技化合作的參與，其中又以operant資源的影響為更大。此外，從研究中也驗證了科技化合作能促成中小企業績效的提升。根據研究結果，本研究將operant資源視為影響中小企業導入科技化合作模式進而達成績效提升的主要關鍵資源。 / Small and medium sized enterprises (SMEs) are the main force to economic growth, but they usually restricted by resource shortage when they want to upgrade goods or services. Past literature has indicated collaboration with partners and customers through information technology (IT) may overcome their weakness and serve better. However, there still are blurs, like what kind of SMEs are more likely to engage IT-enabled collaboration? Will IT-enabled collaboration really enhance SMEs performance? Our research framework based on resource-based view and service-dominant logic categorizing SME resources into (1) operand resources, (2) operant resources, (3) resource complementarity and (4) resource similarity. We attempt to find out what key resources in SMEs are related to the usage of IT-enabled collaboration and the contributions to SME performance. We conduct case study with 8 SMEs in Mt. Pillow Leisure Agriculture Area. After data collection and analysis, we figure out that both operand and operant resources have positive impact on IT-enabled collaboration; furthermore, operant resources contribute more for SMEs to communicate and coordinate with each other and customers. In addition, from our research, we demonstrate that IT-enabled collaboration helps SMEs to perform better. Therefore, we suggest operant resources as important resources for SMEs engaging in IT-enabled collaboration and consequently enhancing their performance.

背景: 生醫文獻語意角色標註（Semantic Role Labeling, SRL）是一種自然語言處理的技術，其可用來將描述生物過程的語句以predicate-argument structures ( PASs ) 表示。SRL 經常受限於arguments的unbalance problem而且需要花費許多的時間和記憶體空間在學習 arguments 之間的相依性。 方法: 我們提出一Markov Logic Network ( MLN ) -based SRL之系統，且此系統使用自動化生成之SRL 模板同時辨識constituents與候選之語意角色。 結果及結論: 我們的方法在BioProp語料上來評估。實驗結果顯示我們的方法勝過目前最先進的系統。此外，使用SRL模板後，在時間及記憶體之花費上亦大幅的減少，而且我們自動化生成之模板亦能幫助建立這些模板。我們認為本論文提出之方法可以透過增加新的SRL模板例如：由生物學家手動寫的模板，而得到進一步的提升，而且本方法也為於需要處理大量SRL 語料時，提供一種可能的解法。 / Background: Biomedical semantic role labeling ( SRL ) is a natural language processing technique that expresses the sentences that describe biological processes as predicate-argument structures ( PASs ) . SRL usually suffers from the unbalanced problem of arguments and consuming time and memory on learning the dependencies between the arguments. Method: We constructed a Markov Logic Network ( MLN ) -based SRL system, and the system uses SRL patterns, which utilizes automatically generated approaches, to simultaneously recognize the constituents and candidates of semantic roles. Results and conclusions: Our method is evaluated on the BioProp corpus. The experimental result shows that our method outperforms the state-of-the-art system. Furthermore, after applying SRL patterns, the costs of the time and memory are greatly reduced, and our automatically generated patterns are helpful in the development of these patterns. We consider that our method can be further improved by adding new SRL patterns such as biological experts manually written patterns and it also provide a possible solution to process large SRL corpus.

逆物流回收的複雜度遠比正向物流高，企業為專注核心價值多半將逆物流活動委外專業物流服務供應商。對第三方逆物流業者而言，選擇適當的回收處理中心位址為其重要核心能力之一，而現今研究對於選址模式中之回收不確定性，大多以歷史資料作為參數，無根據區域特性不同而有所分別。故本研究希望探討在零售商提供資訊的情境下，結合消費者問卷建構廢棄產品的使用年限機率、並以二元迴歸邏輯分析建構回收機率以此預測區域回收數量，透過資訊分享以建立更好的回收處理中心選址設置模式，使第三方逆物流業者可按照此模式選擇最適當的回收點位置與回收處理量安排用以求得利潤最大化。 / Since reverse logistics is much more complex than forward logistics, third-party logistics providers are often the prior choice for firms to obtain their core value when a reverse logistic activity is needed. For third-party logistics providers, the location is one of their crucial core values; while most of them can only rely on historical data to assume the best location, due to the uncertainty of recycling in present studies.Therefore, this paper tries to construct the probability of products’ used-years by combining the retailers’ information with consumer-oriented questionnaires. Binary logistic regression is the methodology used to analyze and predict recycling probability. By information-sharing the third-party logistics providers will be able to construct a better selecting model for the best facility location, which will reach the most suitable recycling quantity to maximize their profits.

近年線上教學平台有著很大的發展，不管是國內的均一教學平台，或國外知名的可汗教育平台，都提供各種學科便利學生自主學習。而在高中化學計算的領域中，這些平台上均提供各種教學課程。美中不足的是在線上的練習系統中，往往題目數量少、題目變化少、無詳細解題步驟，這樣將不足以透過題目衡量一個學生在各個主題的學習上有無明顯的進步。 本論文的目的是改善上述問題。我們設計並實做一系統，只要使用者輸入簡單需求，即可自動產生高中化學問題以及伴隨詳細解答，可方便出題者快速產生各式不同主題的高中化學應用題目。我們的系統提供一個Web前端供使用者輸入所需要生成的題目之資訊。系統由此收齊相關參數之後，接著即可依據參數產生符合題目限制條件的化學問題生成模型。此問題模型為一hypergraph，節點代表已知或未知相關化學量，超連結(hyperedge)則代表數個化學量間的相依關係。有了此一以ASP(Answer Set Programming)表達的問題模型之後，系統即可利用ASP求解器(Solver)進行單一或多個題目生成，後續工作則是驗證每一生成題目之可行性並產生解題步驟，最後經由Django整合呈現於Web上。 / In recent years there has been great progress in the development of online learning. Well-known platforms such as international Khan Academic or local Junyi Academy in Taiwan provide courses in various subjects allowing interested students to study in a very convenient and autonomous way. As expected, courses on common subjects such as high school chemistry are offered with rich content by these platforms. However, there are shortcomings in these courses about the problems they provide for the students to practice or test. In addition to rich content, an ideal course should provide abundant problems of all possible topics, with each given detailed solution, so that students can evaluate their achievement of study by practicing or testing themselves with these problems. Unfortunately, no courses on these platforms meet the above requirements. The purpose of this thesis is to improve the above shortcoming by providing a system which can generate automatically word problems on various topics of high school chemistry, together with detailed accompanied solutions. Our system is a web-based application implemented using Django. It provides a front-end enabling the users to enter related information for the word problems they want the system to generate. According to the parameters collected from the front-end, our system will generate a corresponding chemical problem model. The model is a hypergraph with nodes representing known or unknown chemical quantities related to the problem and hyperedges representing relations or dependencies among these quantities. After the model is generated as a logic program of ASP(Answer-set Programming), the system will use an ASP solver to generate one or more candidate problems. Subsequent works are then used to verify the feasibility of each problem and produce a solution for the feasible one. Finally the generated problems as well as solutions are wrapped in the server side and then sent to and presented friendly in the client's browser.

本文主要在探討自變數存在有測量誤差時，邏輯式迴歸模型的估計問題，並設計實驗使得測量誤差能滿足遞減假設，進一步應用序貫分析方法，在給定水準下，建立一個信賴範圍。 當自變數存在有測量誤差時，通常會得到有偏誤的估計量，進而在做決策時會得到與無測量誤差所做出的決策不同。在本文中提出了一個遞減的測量誤差，使得滿足這樣的假設，可以證明估計量的強收斂，並證明與無測量誤差所得到的估計量相同的近似分配。相較於先前的假設，特別是證明大樣本的性質，新增加的樣本會有更小的測量誤差是更加合理的假設。我們同時設計了一個實驗來滿足所提出遞減誤差的條件，並利用序貫設計得到一個更省時也節省成本的處理方法。 一般的case-control實驗，自變數也會出現測量誤差，我們也證明了斜率估計量的強收斂與近似分配的性質，並提出一個二階段抽樣方法，計算出所需的樣本數及建立信賴區間。 / In this thesis, we focus on the estimate of unknown parameters, experimental designs and sequential methods in both prospective and retrospective logistic regression models when there are covariates measured with errors. The imprecise measurement of exposure happens very often in practice, for example, in retrospective epidemiology studies, that may due to either the difficulty or the cost of measuring. It is known that the imprecisely measured variables can result in biased coefficients estimation in a regression model and therefore, it may lead to an incorrect inference. Thus, it is an important issue if the effects of the variables are of primary interest. When considering a prospective logistic regression model, we derive asymptotic results for the estimators of the regression parameters when there are mismeasured covariates. If the measurement error satisfies certain assumptions, we show that the estimators follow the normal distribution with zero mean, asymptotically unbiased and asymptotically normally distributed. Contrary to the traditional assumption on measurement error, which is mainly used for proving large sample properties, we assume that the measurement error decays gradually at a certain rate as there is a new observation added to the model. This kind of assumption can be fulfilled when the usual replicate observation method is used to dilute the magnitude of measurement errors, and therefore, is also more useful in practical viewpoint. Moreover, the independence of measurement error and covariate is not required in our theorems. An experimental design with measurement error satisfying the required degenerating rate is introduced. In addition, this assumption allows us to employ sequential sampling, which is popular in clinical trials, to such a measurement error logistic regression model. It is clear that the sequential method cannot be applied based on the assumption that the measurement errors decay uniformly as sample size increasing as in the most of the literature. Therefore, a sequential estimation procedure based on MLEs and such moment conditions is proposed and can be shown to be asymptotical consistent and efficient. Case-control studies are broadly used in clinical trials and epidemiological studies. It can be showed that the odds ratio can be consistently estimated with some exposure variables based on logistic models (see Prentice and Pyke (1979)). The two-stage case-control sampling scheme is employed for a confidence region of slope coefficient beta. A necessary sample size is calculated by a given pre-determined level. Furthermore, we consider the measurement error in the covariates of a case-control retrospective logistic regression model. We also derive some asymptotic results of the maximum likelihood estimators (MLEs) of the regression coefficients under some moment conditions on measurement errors. Under such kinds of moment conditions of measurement errors, the MLEs can be shown to be strongly consistent, asymptotically unbiased and asymptotically normally distributed. Some simulation results of the proposed two-stage procedures are obtained. We also give some numerical studies and real data to verify the theoretical results in different measurement error scenarios.

在資料分析上，調查者通常需要決定，不同的樣本是否可被視為來自相同的母體。一般最常使用的統計量為Pearson’s 統計量。然而，傳統的統計方法皆是利用二元邏輯觀念來呈現。如果我們想要用模糊邏輯的概念來做樣本調查，此時，使用傳統 檢定來分析這些模糊樣本資料是否仍然適當？透過這樣的觀念，我們使用傳統統計方法，找出一個能處理這些模糊樣本資料的公式，稱之為模糊 。結果顯示，此公式可用來檢定，模糊樣本資料在不同母體下機率的一致性。 / In the analysis of research data, the investigator often needs to decide whether several independent samples may be regarded as having come from the same population. The most commonly used statistic is Pearson’s statistic. However, traditional statistics reflect the result from a two-valued logic concept. If we want to survey sampling with fuzzy logic concept, is it still appropriate to use the traditional -test for analysing those fuzzy sample data? Through this concept, we try to use a traditional statistic method to find out a formula, called fuzzy , that enables us to deal with those fuzzy sample data. The result shows that we can use the formula to test hypotheses about probabilities of various outcomes in fuzzy sample data.

本篇論文以《祖堂集》裡出現的所有否定詞為研究對象。首先，用演繹法來建立假設，用歸納法（完全歸納法）來驗證其假設。我們所處的現實世界都是由時間和空間構成的，這種世界與時間結合而造成「事」，與空間結合而產生「物」，我們通常合稱為「事物」。基於這種基本觀點，可說語言結構是一種現實世界的表達，因而它如實地或刻意地反映這種現實世界，而且這些看法也可適用於漢語否定詞的體系。由此，筆者主要用演繹法來大膽地假設說漢語的m系否定詞表達「空間性概念」，而p系否定詞表達「時間性概念」。第參、肆、伍和陸章，利用這種假設，剖析《祖堂集》裡出現的所有否定詞而驗證此假設與《祖堂集》的否定詞用法是否吻合。第二，在語言的邏輯結構層次分析上，在本篇論文的第參章裡，用謂詞邏輯研究法分析《祖堂集》裡出現的否定句以解釋早期近代漢語否定詞的邏輯結構與語言結構的同異。第三，用具有普遍性的語言學方法論，如「區別性特徵（distinctive feature） 」、「標記成分（marked member）」、「直接成分分析法（immediate constituent analysis,ICA）」、「互補分布（complementary distribution）」、「有定（definite reference）和無定（indefinite reference）」、「焦點（focus）、新信息（new information）以及舊信息（old information） 」、「義素分析法（seme analysis或componential analysis）」等，在第肆、伍和陸章裡分析探討《祖堂集》否定詞在邏輯語言結構層次或純粹語言結構層次上的語法結構和語義特點。第四，用計量和統計方法來分析《祖堂集》否定詞的分布情況以解釋近代漢語否定詞的面貌和體系。最後，用漢語否定詞的歷時演變來解釋《祖堂集》否定詞的共時結構，並且透過這種解釋來檢討本篇論文的結論是否具有妥當性。就研究步驟來說，先把《祖堂集》全文輸入到電腦，按否定詞類別歸類而分析探討否定詞的邏輯、語法和語義。 本篇文論共分八章，579頁。第壹章敘論，主要論及研究動機、研究目的、研究範圍、研究方法、否定界說、否定範圍和否定焦點。第貳章邏輯結構、語言結構和否定詞，主要討論「邏輯結構與語言結構之間的同異」、「邏輯的同異問題與否定」以及「時空、語言實詞分類和《祖堂集》否定詞分類（m系否定詞和p系否定詞）之間的關係」。第參章《祖堂集》否定詞之邏輯意義分析，主要用謂詞邏輯學的方法分析《祖堂集》否定詞之邏輯結構和邏輯意義。第肆章《祖堂集》m系「否定詞」之語義分析(一)－「無」，主要用語法學和語義學的研究方法探討m系否定詞當中「無」字在邏輯語言層次和純粹語言層次上之邏輯結構、語法結構和語義特點。第伍章《祖堂集》m系「否定詞」之語義分析(二)，主要用語法學和語義學的研究方法探討其他m系否寫詞在純粹語言層次上之語法結構和語義特徵，如「未」、「莫」、「勿」、「沒」、「亡」、「罔」、「靡」和「微」。第陸章《祖堂集》p系「否定詞」之語義分析，主要用語法學和語義學的研究方法探討《祖堂集》的p系否定詞在邏輯語言層次和純粹語言層次上之邏輯意義、語法結構和語義特點，如「不」、「非」、「匪」、「否」和「弗」。本章後面附帶討論「休」字，上古漢語時期它的否定用法不存在，到後來產生於方言之中，因此它不屬於所謂m/p系否定詞。第柒章《祖堂集》中「否定詞」之邏輯和語義特點，主要探討《祖堂集》否定詞在邏輯層次和語言層次上之特點，如「矛盾關係（contradictory relation）、反對關係（contrary relation）與否定詞」、「時空關係與否定詞」、「主客關係與否定詞」、「常變關係、動靜關係與否定詞」、「時間的空間化（量化）」、「空間的時間化」、「客觀現象的意志化」、「肯定與否定的選擇（疑問）」、「測度否定」、「名詞的述語化」、「雙重否定」、「否定詞之互用」等等。第捌章結論，用三張圖表，把《祖堂集》裡出現的所有否定詞系統化，扼要地解釋而究明第壹章敘論中的六條研究目標。最後附有引用及主要參考書目和《祖堂集》否定詞索引。 總之，筆者利用《祖堂集》裡出現的否定詞探討否定詞在符號隱性系統和符號顯性系統上的特徵。由此，筆者發現到幾點新理論，第一、語言研究要區分符號隱性系統和符號顯性系統，才能得到完善的結果。第二、邏輯學的「矛盾（contradiction）」和「反對（contrary）」顯現為語言的否定詞時，在《祖堂集》裡，區別「矛盾」和「反對」的關鍵在於否定詞是否替代其相應的肯定形式。第三、《祖堂集》的漢語否定詞不僅具有一個完整的系統（paradigm），而且從古到今一直維持m系和p系兩組系統，因為m系和p系否定詞分別表示邏輯意義上的空間概念和時間概念。其他純粹語言結構層次上的否定詞用法都是在這種邏輯意義的基礎上衍生出來的，如「時間的量化否定」、「疑問否定」、「測度否定」、「『無』義否定」等等。第四、邏輯學的矛盾概念和反對概念顯現為語言符號時，產生一些問題，因此研究符號顯性系統的否定詞，反對概念還要分成「確定反對（definite contrary）」和「不定反對（indefinite contrary）」。

當母體中有部份對象因被治癒或免疫而不會失敗時，需考慮這群對象所佔的比率，即存活分率。本文主要在探討如何以貝氏方法對含存活分率之治癒率模式進行分析，並特別針對兩種含存活分率的迴歸模式，分別是Weibull迴歸模式以及對數邏輯斯迴歸模式，導出概似函數與各參數之完全條件後驗分配及其性質。由於聯合後驗分配相當複雜，各參數之邊際後驗分配之解析形式很難表達出。所以，我們採用了馬可夫鏈蒙地卡羅方法（MCMC）中的Gibbs抽樣法及Metropolis法，模擬產生參數值，以進行貝氏分析。實證部份，我們分析了黑色素皮膚癌的資料，這是由美國Eastern Cooperative Oncology Group所進行的第三階段臨床試驗研究。有關模式選取的部份，我們先分別求出各對象在每個模式之下的條件預測指標（CPO），再據以算出各模式的對數擬邊際概似函數值（LPML），以比較各模式之適合性。 / When we face the problem that part of subjects have been cured or are immune so they never fail, we need to consider the fraction of this group among the whole population, which is the so called survival fraction. This article discuss that how to analyze cure rate models containing survival fraction based on Bayesian method. Two cure rate models containing survival fraction are focused; one is based on the Weibull regression model and the other is based on the log-logistic regression model. Then, we derive likelihood functions and full conditional posterior distributions under these two models. Since joint posterior distributions are both complicated, and marginal posterior distributions don’t have closed form, we take Gibbs sampling and Metropolis sampling of Markov Monte Carlo chain method to simulate parameter values. We illustrate how to conduct Bayesian analysis by using the data from a melanoma clinical trial in the third stage conducted by Eastern Cooperative Oncology Group. To do model selection, we compute the conditional predictive ordinate (CPO) for every subject under each model, then the goodness is determined by the comparing the value of log of pseudomarginal likelihood (LPML) of each model.