公務人員的人事陞遷是一個複雜性極高，其中隱藏著許多不變的定律及過程，長官與部屬、各公務人員人之間的關係，更是如同蜘蛛網狀般的錯綜複雜，而各公務人員的陞遷狀況，更是隱藏著許多派系之間的鬥爭拉扯連動，或是提攜後進的過程，目前透過政府公開的總統府公報－總統令，可以清楚得知所有公務人員的任職相關資料，其中包含各職務之間的陞遷、任命、派免等相關資訊，而每筆資料亦包含機關、單位、職稱及職等資料，可以提供各種研究使用。 本篇係整理出一種陞遷序列的資料模型來進行研究，透過資料探勘的相關演算法－支撐向量機(Support Vector Machine，簡稱SVM)及決策樹(Decision Tree)的方式，並透過人事的領域知識加以找出較具影響力的屬性，來設計實驗的模型，並使用多組模型及多重資料進行實驗，透過整體平均預測結果及圖表方式來呈現各類別的預測狀況，再以不同的屬性資料來運算產生其相對結果，來分析其合理性，最後再依相關數據來評估此一方法的合理及可行性。 透過資料探勘設計的分類預測模型，其支撐向量機與決策樹都具有訓練量越大，展現之預測結果也愈佳之現象，這跟一般模型是相同的，而挖掘的主管職務屬性參數及關鍵屬性構想都跟人事陞遷的邏輯不謀而合，而預測結果雖各有所長，但整體來看則為支撐向量機略勝一籌，惟支撐向量機有一狀況，必須先行排除較不具影響力之屬性參數資料，否則其產生超平面的邏輯運算過程將產生拉扯作用，導致影響其預測結果；而決策樹則無是類狀況，且其應用較為廣泛，可以透過宣告各屬性值的類型，來進行不同屬性資料類型的分類實驗。 而透過支撐向量機與決策樹的產生的預測結果，其正確率為百分之77至82左右，如此顯示出國內中高階文官的陞遷制度是有脈絡可循的，其具有一定的制度規範及穩定性，而非隨意的任免陞遷；如此透過以上資料探勘的應用，藉著此特徵研究提供公務部門在進行人力資源管理、組織發展、陞遷發展以及組織部門精簡規劃上，作為調整設計參考的一些相關資訊；另透過一些相關屬性的輸入，可提供尚在服務的公務人員協助其預估陞遷發展的狀況，以提供其進行相關生涯規劃。 / The employee promotion is a highly complexity task in Government office, it include many invariable laws and the process, between the senior officer and the subordinate, various relationships with other government employees, It’s the similar complex with the spider lattice, and it hides many clique's struggles in Government official’s promotion, and help to process the promote for the junior generation, through the government public presidential palace - presidential order, it‘s able to get clearly information about all government employees’ correlation data, include various related information like promotion, recruitment , and each data also contains the instruction, like the job unit, job title and job rank for all research reference. It organizes a promoted material model to conduct the research, by the material exploration's related calculating method – Support Vector Machine (SVM) and the decision tree, and through by knowledge of human resource to discover the influence to design the experiment's model, and uses the multi-group models and materials to process, and by this way , it can get various categories result by overall average forecasting and the graph, then operates by different attribute material to get relative result and analyzes its rationality, finally it depends on the correlation data to re-evaluate its method reasonable and feasibility. To this classification forecast model design, the SVM and the decision tree got better performance together with the good training quality, it’s the same with the general model, and it’s the same view to find the details job description for senior management and employee promotion, however the forecasting result has their own strong points, but for the totally, the SVM is slightly better, only if any accidents occurred, it needs to elimination the attribute parameter material which is not have the big influence, otherwise it will have the planoid logic operation process to produce resist status, and will affect its forecasting result, but the decision tree does not have this problem, and its application is more widespread, it can through by different type to make the different experiment. The forecasting result through by SVM and decision tree, its correction percentage can be achieved around 77% - 82% , so it indicated the high position level promotion policy should be have its own rules to follow, it has certain system standard and the stability, but non-optional promoted, so trough by the above data mining, follow by this characteristic to provide Government office to do the Human resource management, organization development, employee promotion and simplify planning to the organization, takes the re-design information for reference, In addition through by some related attribute input, it may provide the government employee who is still on duty and assist them to evaluate promotion development for future career plan.

全球化潮流方興未艾，基於與國際接軌目標，我國金融業自2006年起實施新巴塞爾資本協定，期於現今日新月異金融環境中以全球一致性的銀行管理方法及制度落實其精神。實施新巴塞爾協定後，首當其衝者便是台灣產業發展主體—中小企業。以信用風險中資本計提為例，中小企業不若大型企業體質健全，且財務透明度亦為人詬病，相對提升金融機構授信風險，進而導致中小企業融資授信審查趨於嚴格與保守，中小企業融資難度與成本皆大幅增加。 有鑑於此，本研究以中小企業中電子業為主要研究對象，採資料採礦流程進行信用評等模型建置。為求配適最佳違約機率模型，分別以不同精細抽樣比例逐一配適羅吉斯迴歸、類神經網路及分類迴歸樹等統計模型，經評估後篩選出羅吉斯迴歸模型建置信用評等系統。再者，為確認模型與信用評等系統建置適當，係遵循新巴塞爾協定相關規範進行各項測試及驗證，結果顯示模型於樣本外資料測試表現良好，信用評等系統亦通過正確性分析、等級區隔同質性檢定及穩定度分析等驗證準則，冀能提供金融機構一套有效且精簡的信用管理機制，建立與中小企業間資訊對稱管道，於兩造雙方取得互利平衡，防範危機於未然。 / Globalization trend is still growing. Because of the objective of connecting to the world, the banking and finance industry in Taiwan has implemented the New Basel Capital Accord since 2006, hoping to make use of globally consistent banking management method and system to implement its spirit in this changing financial environment today. After the implementation of the New Basel Capital Accord, the principal development part in Taiwan industry, medium- and small-sized enterprises, is the first to be affected. For example, with regard to the capital requirements in credit risks, the constitution of medium- and small-sized enterprises is not as sound as large-sized enterprises’, and the financial transparency of medium- and small-sized enterprises is insufficient that the credit risk of financial institution would be lifted comparatively; and then, the finance and credit investigation of medium- and small-sized enterprises would become strict and conservative, thus the finance difficulty and cost of medium- and small-sized enterprises would be increased substantially. In view of this, this study regards the electronics industry from medium- and small-sized enterprises as the main study objects, and data mining procedures are used so as to establish the credit scoring system. To get the best probability model of default, different oversampling ratios are used one by one to match such statistical models and logistic regression, Neural Network Analysis, and C&R Tree; and logistic regression model is selected for the establishment of credit scoring system after assessment. Moreover, relevant the New Basel Capital Accord standards are followed to carry out every test and verification so as to confirm that the establishments of model and credit scoring system are appropriate. The result indicates that the model has good performance in out-sample test, while credit scoring system also passes such verification standards as accuracy analysis, level segment homogeneity test, and stability analysis. Hopefully, this study result can provide a set of effective and simple credit management system for the financial institution to establish information symmetrical channel with the medium- and small-sized enterprises, so that both parties can obtain mutual balance and the crisis can be alerted in advance.

近年來臺灣高齡人口比例有明顯之增長，兩性平均餘命自1906年至今不斷的往高齡延伸，伴隨著這兩種趨勢下，臺灣高齡人口結構快速的轉型和變動，使得瞭解高齡人口死亡率模式成為估算未來臺灣人口結構發展趨勢的重要依據。然而，過去許多人口研究所依賴的資料來源是以「戶籍人口統計資料」為基礎，其資料內容雖涵蓋時間範圍甚廣，但在高齡人口的死亡率資料記載則有稍嫌簡化的問題及死亡人數紀錄不準確的限制，因此本研究擬以搭配「死因資料檔」，擷取其對於死亡人口數及死亡時間詳細紀錄的優點，來結合運用以探討臺灣高齡人口死亡率模式。 本研究以「參數式模型」、「相關模型」、「外推法」及「APC模型」四種不同估算取徑的運用，並結合現有實際的臺灣高齡人口死亡率資料，說明臺灣65歲以上高齡人口死亡率的變遷模式及發展軌跡。研究結果顯示，自1975年臺灣65歲以上高齡人口的死亡率變遷趨勢，確實往更高齡方向發展，同時，其死亡率的變遷波動越大，本身除資料紀錄上可能有所偏誤外，也可能因為90歲以上人口資料逐漸增加的情況下，變異性逐漸的明顯。此外，以各死亡率模型估算配適下，大多在90歲以上高齡人口的計算，其估算不準確的情形則越形明顯，但在65歲至85間的估算死亡率模式則有相當不錯的配適。

稅捐分處是最接近民眾之第一線機關，直接影響民眾對稅捐機關之觀感，其亦為最小的單位，卻肩負稅收稽徵、防止逃漏、為民服務等業務，故衡量分處的稽徵績效刻不容緩。本研究以臺北市稅捐稽徵處所屬13個分處為研究對象，並以員額、稽徵成本與設備數量等作為投入項、以各稅實徵淨額、違章裁罰金額、欠稅清理金額與人民申請案件數作為產出項，運用三階段資料包絡分析法進行衡量，評估臺北市稅捐稽徵處所屬各分處之技術效率，探討外生因素對產出差額的影響，並與原稽徵績效考核成績進行比較。實證結果顯示，在排除外生因素之影響後，整體效率值是較前提升的，而造成無效率的原因，部分可歸責於資源的浪費、部分可歸咎於未達最適規模；外生變數對大部分的產出差額具顯著影響，轄區內營業家數愈多，稅收金額較高；位於臺北市南區，稅課收入亦較多；前年平均國民生產毛額對大部分技術效率有負向影響；實務面之稽徵績效考核因考慮面向較廣、涉及過多非相關考評項目而淡化了各分處稽徵效率，為使對分處考核制度更為嚴謹，建議使用一套客觀的效率準則來衡量稽徵效率並加以評估，較能反映分處實際之稽徵績效。

本研究運用台灣教育長期追蹤資料庫（Taiwan Education Panel survey）之2001年與2003年針對同一批國中生（N=16,530）蒐集的資料，檢證James Coleman代間封閉性網絡有助於學生學習成效之理論。對於Coleman的理論，過往實證的研究發現並不一致。台灣亦有研究顯示，在升學制度的壓力下，家長間網絡不見得出現效力。本研究旨在檢證兩種不同家長網絡形式在學校與個體層次對學生學習成效的影響，研究結果發現：（1）個人層次方面，僅校外家長連帶有益於學習成效，但社會經濟地位具有間接影響力；（2）學校層次方面，則僅代間封閉性網絡具影響力，封閉性越高越有益於學生學習成效，且其影響力與社會經濟地位無關。

本文研究目的旨在探討所得不均度對自殺率之影響，採用臺灣2004年至2010年23個縣市的追蹤資料（panel data），使用行政院主計處家庭收支調查報告之年度原始資料，計算自2004年至2010年台灣地區23縣市之吉尼係數作為所得不均度代理變數。 實證結果顯示吉尼係數對總自殺率、男性與25至44歲年齡組自殺率的效果是不顯著的正相關。本文認為，2004至2010年吉尼係數均維持在0.34左右，而且變化不大，以及樣本年度不足，可能是實證結果不顯著的原因之一。此外，由於用以計算吉尼係數之可支配所得並未納入資本利得，導致吉尼係數偏低亦可能是實證結果不顯著的另一個原因。 然而，吉尼係數對於男性自殺率及25至44歲自殺率估計係數的大小卻是值得注意，隱含這二個族群非常關注所得分配的公平性，建議政府在短期政策上可就獨厚富人之賦稅不公平現況加以改善，以達杜漸防微之效。 / This study investigates the impact of income inequality on suicide rate in Taiwan. Using panel data of 23 countries for the period 2004-2010, As a proxy for income inequality, Gini coefficients based on the Survey of Family Income and Expenditure, compiled by the Department of Government of Budget. Empirical results show Gini coefficient has a positive but statistically insignificant effect on total, male and aged 25-44 suicide rates. This paper argues that the Gini coefficient remain around 0.34 for the period of 2004-2010 without significant change, and only 7 years of data may be part of the reason why empirical result is not significant, another reason that might cause Gini coefficient lower is one of the element "disposable income" it did not included capital gain. However, the most noteworthy feature is the magnitude of the Gini coefficients for male and aged 45-64. In other words, these two groups are concerned on the fairness of income distribution. In order to prevent the suicide rate goes higher, government should change the unfair taxes that benefit the riches right away.

在數據分析的領域中，尤其在大數據的領域之中，因常含有相當高維度的預測變數，故特徵選擇是一個很重要的主題。這個主題在半導體的應用上，已經獲得相當豐碩的成果，但在IC基板的應用上，成果就相對顯得貧乏。所以，此次的研究(以IC基板中鑽孔製程為例)將透過以下的試驗方法(含：GR-SNBC (Gain Ratio with Naive Bayes Classifier)、SU-SNBC (Symmetrical Uncer-tainty with Naive Bayes Classifier)與SU-CART (Symmetrical Uncer-tainty with Classification and Regression Tree Classifier))，來建立可應用於IC基板製程時間預測上的一組屬性。最後，此一研究的成果不僅在於，使用資料挖礦的方法，來找出一組具有顯著性，而且可以用來預測的IC基板製程時間的產品特徵屬性；而且，發現若為了縮短製程時間，來自產品結構本身的因子，會比來自產品在生產管理上的因子更具顯著的效果。 / Feature selection is significate subject in domain of data analysis, especially in big-data with a lot of high dimension predictive variables. In semi-conductor field, this subject has already gotten a plenty of achievement, but not in IC-substrate; so in this research for example of drilling operation, through experiments, it builds a group of se-lective features for this field to predict process time, and the methods used are GR-SNBC (Gain Ratio with Naive Bayes Classifier), SU-SNBC (Symmetrical Uncertainty with Naive Bayes Classifier) and SU-CART (Symmetrical Uncertainty with Classification and Regression Tree Classifier). The contributions of this research are not only a selective product characteristics subset suggested to predict process-time in IC-substrate fab via the data-mining methods here, but also an observation that in order to shorten the process time, the factors of product construction weighs more than production management.

「民以食為天」，飲食與我們的生活息息相關。近年來由於食安風暴肆虐，自行烹煮的需求隨之高漲。然而在家自行烹煮時常會面臨不知道該烹煮什麼料理的問題，因此有便利的食譜搜尋系統對烹煮的人而言將是相當方便的。然而使用搜尋系統時，由於我們只知道想用某些特定食材進行烹煮，而不知道哪些食譜含有特定食材，因此在以少數食材進行查詢時不免會得到過多的食譜結果而難以快速找到喜好的食譜。我們建立了一個食譜搜尋的自動完成機制，並依照該機制實做出了食譜搜尋引擎。使用者使用系統進行搜尋時，我們將會依照使用者輸入的食材尋找適合搭配的食材推薦給使用者，幫助使用者在查詢時使用更完整的Query讓搜尋系統可以找到更少更精準的食譜，幫助使用者更快的找到喜歡的食譜。然而只推薦搭配性食材，可能會推薦出與Query中的食材是替代關係的食材，也就是通常不會一起出現的食材，因此我們也進行了替代性食材的研究。給定由兩個食材組成的食材配對，我們研究如何自動的判斷替代性食材。我們將問題轉化成分類問題來解決，並使用One-Class Classification的技術解決分類問題中的Imbalanced Problem。我們使用f1-score觀看One-Class Classification與傳統分類器的比較。經實驗測試，One Class Classification與傳統分類器相比，One Class Classification較能協助我們解決Imbalanced Problem。

長期性資料(longitudinal data)，是指對同一個觀察個體(subject)或實驗單位(experiment unit)，在不同時間點上重複觀察或測量一個或多個變數。雖然觀察個體之間互相獨立，但就同一個個體而言，不同時間的觀察或測量常常是有相關性的。且觀察的個體之間可能由於一些無法測量的環境因素造成個體之間有差異，因此在傳統橫斷面分析中，假設其有相同迴歸係數的邊際模型可能不合理。隨機效果模型可以解決長期資料分析的相關，並假設每個個體的迴歸係數不同；此模型不但可以說明橫斷面資料的cohort效果，也可直接解釋長期資料的age效果；更可以區分個體之間與個體之內的變異。 本研究以1995年至2000年台灣11個產業中的100家公司之每股盈餘與各財務比率，作為實證分析的資料；分別配適每股盈餘與時間、產業別、時間產業別交互作用及財務比率及排除每股盈餘有異常值後之邊際效果模型(一般迴歸分析)及隨機效果模型，並比較其參數估計之異同。實證結果顯示，一般迴歸分析與假設誤差不相關且等變異下的隨機效果模型參數估計相似，但後者能區分變異為個體之間(between-subjects)與個體之內(within-subject)的變異。而假設誤差不相關且不等變異與假設誤差服從AR(1)且不等變異下的隨機效果模型估計相近。實證結果並顯示，在排除異常值後的模型參數估計，一般迴歸分析不論是估計值及顯著性大多沒有很大差別；而隨機效果模型的估計在排除異常值前後較有差別。特別是現金流量比率(CFR)原本為不顯著變數，在排除異常值後的模型配適全部變顯著性變數。 / The defining characteristic of a longitudinal study is that individuals are measured repeatedly through time. Although it is independent between subjects, the set of observations on one subject tends to be inter-correlated. Because there is some natural heterogeneity due to unmeasured factors between subjects, it is not corrected to assume they have the same regression coefficients. A random effect model is a reasonable description about the different regression coefficients, and it can resolve the inter-correlation of the observations on one subject. The major advantages of the random effect model are its capacity to separate what in the context of population studies are called cohort and age effects, and it can distinguish the variations between subjects and within subjects. This study describes the marginal model and random effect model, and shows their difference by real data analysis. We apply these models to the earnings per share (EPS) and other financial ratios of one hundred companies in Taiwan, which are distributed in eleven industries. The results show that the parameter estimates of the marginal model and random effect model are similar when error structure is independent and of equal variance. Furthermore, the latter can distinguish the variations between subjects and within subjects. However, the residual analysis reveals that the error structure may not be constant. Therefore, we consider heteroscedasticity error in random effect model. We also assume that error follows an autoregressive process (e.g. AR(1) model), which leads to the optimum among our results in terms of residual analysis. There are some observations that appear to be outlying from the majority of data. The results show little difference in the marginal models no matter whether those outliers are included. However, we obtain different results in the random effect models. Especially, the variable of “cash flow ratio” becomes significant once those potential outliers have been excluded, while it is not significant when all cases are fitted in the model.

本研究以台灣8家生物科技公司為研究對象，探討公司於民國85年到88年之間的經營效率評估，比較孰優孰劣。本研究應用資料包絡分析法(data envelopment analysis, DEA)來計算相對效率值。投入要素包括：資本額、研發支出、員工人數，及員工素質共四項；產出項目則為公司營業額一項。研究中分別求解CCR效率和A&P效率，此外，亦將CCR效率進一步區分為純粹技術效率(BCC效率)與規模效率，除了效率值比較和衍生的相關討論外，還進行規模報酬分析、虛擬乘數分析、差額變數分析，與敏感度分析，最後則是獲利能力與經營效率之比較。 研究結果顯示，效率排名以杏輝表現最佳，其次依序為濟生、葡萄王、永日、永信、生達、中化、五鼎；依年度區分的平均效率值分析，可觀察到的共通現象，皆是從民國85年一路衰退到88年，顯示八家生技公司的營運效率在此期間總體表現不佳。整體來看，投入項目需縮減幅度最大者為員工素質，資本額次之。表示八家生技公司在此期間高素質人力的投入，並無產生相當的營收，原因可能是公司開發的產品未能符合市場的需求，因此，未來在開發新產品方面應加強結合行銷功能，才能充分滿足消費者或客戶的需求。此外，資本額投入過多，造成資源的浪費，也需要公司管理者加強成本的控管，以期達到資源有效分配。從獲利能力與經營效率之比較分析得知，經營效率高的公司大多獲利能力亦較高。 / Utilizing Data Envelopment Analysis (DEA), this paper examines the relative efficiency of 8 companies over a period of 4 years in Taiwan biotech industry. The study has indicated how to use DEA to identify individual companies that are less efficient than other comparable units of output factors relative to input factors. These DEA models basing on the data of 1996-1999 provide CCR efficiency, and A&P efficiency. Furthermore, CCR efficiency is divided into pure technical efficiency (BCC efficiency) and scale efficiency. Besides the comparison of these efficiencies and the discussion about related content, the present study also performs scale analysis, multiplier analysis, slack analysis, and sensitivity analysis. Finally, the comparison of profitability and operating efficiency is conducted. The research shows that Sinphar Pharm. Corp. is ranked first in efficiency, and then Chi Sheng Chemical Corp., Grape King Corp., Yung Zip Chemical Corp., Yung Shin Pharm. Corp., Standard Chemical & Pharm. Corp., China Chemical & Pharm. Corp., and Apex Biotech Corp. in order. Average efficiency of these eight companies declined from 1996 to 1999. As a whole, the personnel Ability and Capital should be the first two input item that needs to be reduced. The comparison of profitability and operating efficiency indicates that most efficient companies can have good profitability.