藉由「早期發現，早期治療」的方式，我們可以降低癌症的死亡率。因此找出與癌症病變有關的生物標記以期及早發現與治療是一項重要的工作。本研究分析了包含正常人以及攝護腺癌症病人實際的蛋白質質譜資料，而這些蛋白質質譜資料是來自於表面強化雷射解吸電離飛行質譜技術（SELDI-TOF MS）的蛋白質晶片實驗。表面增強雷射脫附遊離飛行時間質譜技術可有效地留存生物樣本的蛋白質特徵。如果沒有經過適當的事前處理步驟以消除實驗雜訊，ㄧ 個質譜中可能包含多於數百或數千的特徵變數。為了加速對於可能的蛋白質生物標記的搜尋，我們只考慮可以區分癌症病人與正常人的特徵變數。 基因演算法是一種類似生物基因演化的總體最佳化搜尋機制，它可以有效地在高維度空間中去尋找可能的最佳解。本研究中，我們利用仿基因演算法(GAL)進行蛋白質的特徵選取以區分癌症病人與正常人。另外，我們提出兩種兩階段仿基因演算法(TSGAL)，以嘗試改善仿基因演算法的缺點。 / Early detection and diagnosis can effectively reduce the mortality of cancer. The discovery of biomarkers for the early detection and diagnosis of cancer is thus an important task. In this study, a real proteomic spectra data set of prostate cancer patients and normal patients was analyzed. The data were collected from a Surface-Enhanced Laser Desorption/Ionization Time-Of-Flight Mass Spectrometry (SELDI-TOF MS) experiment. The SELDI-TOF MS technology captures protein features in a biological sample. Without suitable pre-processing steps to remove experimental noise, a mass spectrum could consists of more than hundreds or thousands of peaks. To narrow down the search for possible protein biomarkers, only those features that can distinguish between cancer and normal patients are selected. Genetic Algorithm (GA) is a global optimization procedure that uses an analogy of the genetic evolution of biological organisms. It’s shown that GA is effective in searching complex high-dimensional space. In this study, we consider GA-Like algorithm (GAL) for feature selection on proteomic spectra data in classifying prostate cancer patients from normal patients. In addition, we propose two types of Two-Stage GAL algorithm (TSGAL) to improve the GAL.

本研究係發想自政治大學「未來馬戲團」的展演活動表演方式，嘗試改進表演方式中的程式技術，以程式化方式整合展演藝術中實體與虛擬的互動平台，我們希望提供導演撰寫較為口語或展演描述方式的腳本敘述如『when ... other-wise ...』，如此一來就可以任意組合實體演員的肢體動作與指示虛擬環境的特效，因此我們採用了一套介接實體與虛擬環境應用程式的領域專屬語言- Digital In-teractive Performance Sketch (DIPS)，用以開發客製化的展演程式庫，並佈署於本團隊自行開發的執行引擎 Wearable Item Service runtimE (WISE)，提供導演在這個引擎上透過這個DIPS編寫前述口語的程式腳本，讓程式自行互動，達成展演效果的自動變化。 我們的系統會接收來自展演人員穿配的連網感應器上的訊號，並且根據導演寫好的腳本規則，自動根據接收到的裝置訊號判斷出該指示虛擬環境做出什麼樣的效果，以達到展演效果自動變化，完成虛擬與實體展互動的程式支援。 為了減少腳本程式撰寫前須具備的程式邏輯訓練，本研究開發一款所見即所得(WYSIWYG)的視覺化腳本編輯器 DIPS Creator，提供腳本編寫者可以直覺的方式組合編輯器中的展演詞彙方塊，完成腳本設計。 本研究展示了如何以較為口語或展演語意的方式敘述展演規則，以實現虛實互動的程式化，並且提供了具有彈性的客製化展演函式庫及圖形化展演規則編輯器的製作方式，未來可增加多演員層次的抽象支援以展現本研究系統的更多程式化能力，並加入表演階段設計、雙向溝通與規則互斥等能力，擴充系統功能。 / This research was inspired by “The Future Circus”, a cyber-physical interactive performance art developed in National Chengchi University. In this thesis, we pro-pose some mechanisms to support such performance art programmatically in a more effective manner. Specially, we provide a high-level scripting tool for directors to de-scribe the performance rules abstractly in the form of “when ... otherwise ...”, so that directors can compose arbitrary actions and effects easily. Underlying such abstract rules are a domain specific language – Digital Interactive Performance Sketch (DIPS), and a middleware. Wearable Item Service runtime (WISE), developed by our research team. Given a script with those abstract rules, our system will receive signals sent from a sensor on wearable devices of actors, and then it will command cyber environment perform effects, the performance effects or actions according to rules written by the director. Through our integration efforts, the performance effects in the cyber environment will change automatically in a programmatic way. Besides, for users without prior scripting experience, we developed a WYSIWYG GUI editor, DIPS Creator, that allows users to write a script intuitively by dragging and dropping pre-built rule blocks. We conduct a few experiments with real sensor device to demonstrate the programming support of our tool. The preliminary results are satisfactory in terms of prototype support. To further extend our tool for practical performance, we describe in detail a few directions such as support for multiple actor performance stage model-ing, and integrity check of related rules that will make our system more powerful.

因網際網路帶來的便利性與即時性，網路新聞成為社會大眾吸收與傳遞新聞資訊的重要管道之一，而累積的巨量新聞亦可反映出社會輿論對某特定新聞議題之即時反應、熱門程度以及情緒走向等。 因此，本研究期望借由意見探勘與情緒分析技術，從特定領域新聞中挖掘出有價值的關聯，並結合傳統機器學習建立一個房地產市場的預測模式，提供購屋決策的參考依據。 本研究搜集99年1月1日至103年6月30日共1,1150筆房地產新聞，以及8,165件捷運週邊250公尺內房屋買賣交易資料，運用意見探勘萃取意見詞彙進行情緒分析，並建立房市情緒與成交價量時間序列，透過半年移動平均、二次移動平均及成長斜率，瞭解社會輿情對房市行情抱持樂觀或悲觀，分析社會情緒與實際房地產成交間關聯性，以期能找出房地產買賣時機點，並進一步結合情緒及房地產的環境影響因素，藉由支援向量機建立站點房市的預測模型。 實證結果中，本研究發現房市情緒與成交價量之波動有一定的週期與相關性，且新捷運開通前一年將連帶影響整體捷運房市波動，當成交線穿越情緒線且斜率同時向上時，可做為適當的房市進場時機點。而本研究針對站點情緒與環境變數所建立之預測模型，其預測新捷運線站點之平均準確率為69.2％，而預測新捷運線熱門站點之準確率為78％，顯示模型於預測熱門站點上具有不錯的預測能力。 / Nowadays, E-News have become an important way for people to get daily information. These enormous amounts of news could reflect public opinions on a particular attention or sentiment trends in news topics. Therefore, how to use opinion mining and sentiment analysis technology to dig out valuable information from particular news becomes the latest issue. In this study, we collected 1,1150 house news and 8,165 house transaction records around the MRT stations within 250 meters over the last five years. We extracted the emotion words from the news by manipulating opinion mining. Furthermore, we built moving average lines and the slope of the moving average in order to explore the relationship and entry point between public opinion and housing market. In conclusion, we indicated that there is a high correlation between the news sentiment and housing market. We also uses SVM algorithm to construct a model to predict housing hotspots. The results demonstrate that the SVM model reaches average accuracy at 69.2% and the model accuracy increases up to 78% for predicting housing hotspots. Besides, we also provide investors with a basis of entry point into the housing market by utilizing the moving average cross overs and slopes analysis and a better way of predicting housing hotspots.

全球智慧型手機的出貨量持續提升，且熱門市集的App下載次數紛紛突破500億次。而在iOS和Android手機App市集中，App的評價和評論對App在市集的排序有很大的影響；對於App開發者而言，透過評論確實可掌握使用者的需求，並在產生抱怨前能快速反應避免危機。然而，每日多達上百篇的評論，透過人力逐篇查看，不止耗費時間，更無法整合性的瞭解使用者的需求與問題。 文字情感分析通常會使用監督式或非監督式的方法分析文字評論，其中監督式方法被證實透過簡單的文件量化方法就可達到很高的正確率。但監督式方法有無法預期未知趨勢的限制，且需要進行耗費人力的文章類別標注工作。 本研究透過情感傾向和熱門關注議題兩個面向來分析App評論，提出一個混合非監督式與監督式的中文情感分析方法。我們先透過非監督式方法標注評論類別，並作視覺化整理呈現，最後再用監督式方法建立分類模型，並驗證其效果。 在實驗結果中，利用中文詞彙網路所建立的情感詞集，確實可用來判斷評論的正反情緒，唯判斷負面評論效果不佳需作改善。在議題擷取方面，嘗試使用兩種不同分群方法，其中使用NPMI衡量字詞間關係強度，再配合社群網路分析的Concor方法結果有不錯的成效。最後在使用監督式學習的分類結果中，情感傾向的分類正確率達到87%，關注議題的分類正確率達到96%，皆有不錯表現。 本研究利用中文詞彙網路與社會網路分析，來發展一個非監督式的中文類別判斷方法，並建立一個中文情感分析的範例。另外透過建立全面性的視覺化報告來瞭解使用者的正反回饋意見，並可透過分類模型來掌握新評論的內容，以提供App開發者在市場上之競爭智慧。 / While the number of smartphone shipment is continuesly growing, the number of App downloads from the popular app markets has been already over 50 billion. By Apple App Store and Google Play, ratings and reviews play a more important role in influencing app difusion. While app developers can realize users’ needs by app reviews, more than thousands of reviews produced by user everday become difficult to be read and collated. Sentiment Analysis researchs encompass supervised and unsupervised methods for analyzing review text. The supervised learning is proven as a useful method and can reach high accuracy, but there are limits where future trend can not be recognized and the labels of individual classes must be made manually. We concentrate on two issues, viz Sentiment Orientation and Popular Topic, to propose a Chinese Sentiment Analysis method which combines supervised and unsupervised learning. At First, we use unsupervised learning to label every review articles and produce visualized reports. Secondly, we employee supervised learning to build classification model and verify the result. In the experiment, the Chinese WordNet is used to build sentiment lexicon to determin review’s sentiment orientation, but the result shows it is weak to find out negative review opinions. In the Topic Extraction phase, we apply two clustering methods to extract Popular Topic classes and its result is excellent by using of NPMI Model with Social Network Analysis Method i.e. Concor. In the supervised learning phase, the accuracy of Sentiment Orientation class is 87% and the accuracy of Popular Topic class is 96%. In this research, we conduct an exemplification of the unsupervised method by means of Chinese WorkNet and Social Network Analysis to determin the review classes. Also, we build a comprehensive visualized report to realize users’ feedbacks and utilize classification to explore new comments. Last but not least, with Chinese Sentiment Analysis of this research, and the competitive intelligence in App market can be provided to the App develops.

本研究的主要目的在探究以想法為中心的教學設計對大學生學習歷程與知識信念的影響。本研究透過知識論壇 (Knowledge Forum，KF) —一個以知識翻新（knowledge-building）教育理念為核心所建立的數位學習平台—來記錄學生的學習歷程。研究設計採個案研究法，以一個班級為研究單位，藉由多元資料的蒐集，以深入瞭解知識翻新教學對大學生在學習歷程上的影響。研究對象為某大學修習「生活科技概論」課程之22位學生。 資料來源包括：(1) 知識論壇上的學習筆記(note)—即學生整學期在知識論壇平台上的貼文與對話記錄；(2) 期末科技成品與影音記錄—即課程中學生所設計科技產品的期末發表; (3) 期末反思—即學生的期末學習心得; (4) 知識信念問卷(Chan & Elliott, 2004)的前後測資料。 本研究依據資料分析結果提出以下三點主要結論: (1)協作式的知識共構活動可以幫助學生跳脫傳統以個人為中心的學習方式，並進而以集體想法交流的形式來共構知識；(2) 從學生在科技產品上的創意表現可知，學生已逐漸能體認到知識是可以被創造的，人人都有能力創造知識；(3) 學生經過一學期的知識建構過程後，在知識信念上也有部分轉變，特別是在「天賦能力」的面向上，傾向學習能力並非天賦且不可改變的觀點。 根據上述結論，本研究在教育實務上提出以下三點建議：(1) 教師應多運用想法中心的教學設計，以培養有自主想法的學生，並藉此激勵學生積極參與學習社群中的知識共構活動；(2) 教師應將知識創造的能力視為學生學習的關鍵能力，以培養學生面對未來職場所需的競爭力；(3) 師資培育機構應注重師資培育生在知識信念上的發展，並協助其養成更具建構取向與學生中心的教學信念。 關鍵詞：想法中心、電腦支援協作學習、知識翻新、知識信念、主動學習 / The purpose of this research was to investigate the effects of idea-centered education on college students’ learning progress and change in epistemological beliefs. An online collaborative knowledge-building environment—Knowledge Forum (KF)—was employed in this study to document the process of students’ idea generation and development, as well as their belief change. This study employed a case-study design, using a college class as a unit for investigation, with multiple datasets collected to understand how knowledge building might affect students’ learning process and outcome. Participants were 22 students (16 females and 6 males) who took a course titled “Introduction to Living Technology” in a university, Taiwan. Major data sources include: (1) Students’ online discourse recorded in a KF database; (2) Students’ final presentations about the technology products they designed (the presentations were videotaped); (3) An open-ended survey employed to elicit students’ reflection on what they did and learned from this course; (4) A belief questionnaire, adopted from Chan & Elliott (2004), which was administered in the beginning and at the end of the course. There were three main findings: (1) An idea-centered approach was found helpful for students to move away from individual-based learning to taking more collective responsibility in their knowledge advancement. It was found that with support of KF, students were able to engage in continuous idea exchange and improvement; (2) As evidenced in students’ performance on the design of creative technology products, it was suggested that students have gradually realized that knowledge can be created and that all humans has potential to create knowledge; (3) After doing knowledge-building for a whole semester, the participating students also shifted their’ beliefs to become less inclined to believe in inborn ability as a fixed quality. Building on the findings, this study made the following suggestions: (1) To foster student capacity for more autonomous learning and collaborative knowledge building, teachers should try to encourage students to work with ideas and to engage them in sustained idea improvement; (2) Teachers should regard knowledge creation as a key ability that students need to acquire in order to face the competitive workplace in the future; (3) Teacher education programs should pay attention to the development of students’ epistemological beliefs, and to help student develop a more constructivist-oriented and student-centered teaching beliefs. Keywords: idea-centered, CSCL, knowledge building, epistemological beliefs, active learning

傳統的腫瘤指標篩檢方法，往往靈敏度、普及度及特異性有限，無法得到正確、即時的診斷結果。現今癌症的研究，則透過蛋白質體學經由光譜及影像觀察癌症不同時期的蛋白質表現變化，期望未來得以發展較佳之診斷工具。本研究中主要針對兩組攝護腺癌症病人之蛋白質質譜資料，此資料應用蛋白質晶片與表面強化雷射解吸電離飛行質譜技術（SELDI-TOF-MS）收集而來。我們的研究目的在於從大量的蛋白質特徵中篩選出一群有助於分類的蛋白質特徵變數。我們提出以最小分錯率特徵選取法與最小p值（ 檢定、Kruskal-Wallis檢定）特徵選取法進行初步特徵辨識度排序以及選取，並進一步發展出k-mean萃取法、最大相關係數萃取法與判定係數萃取法以改善變數間嚴重的共線性問題。我們利用支援向量機（Support Vector Machine）方法進行分類並評估分類效果，在不同的分類目的下萃取有助於辨識的蛋白質特徵，以決定最佳特徵集合。研究發現運用最小分錯率特徵選取法與最小p值分錯率特徵選取法，輔以判定係數萃取法，在各分類目的下皆有良好表現，為較佳的特徵選取方式。

本研究選擇工業電腦產業的領導廠商研華科技公司為個案研究對象，想藉由研究該公司的技術支援服務流程，來瞭解其中可能出現的問題，並進一步提出改善的建議。 在整個研究架構方面，採用Davenport and Short（1990）的企業流程再造的五階段，乃因為此架構既具有完整的流程再造觀念，也稍具執行細節的考量，因此決定以該兩位學者所提的企業流程再造五階段作為本研究架構。 根據本研究，發現其現行的技術支援服務流程有以下的問題： 1. 業務工程師花費過多時間於技術支援服務上，使得他們缺乏時間開發與創造新的商機。 2. 客戶找業務請求技術支援並不會留下記錄。 3. 業務工程師與應用工程師有遺忘客戶需求的可能。 4. 業務無法隨時掌握顧客需求被滿足的狀況。 5. 應用工程師被動等待客戶跟催。 因此本研究提出以下建議： 1. 使用「即時支援軟體」來讓客戶可以輕鬆解決自己的問題，並留下服務需求記錄。 2. 加強公司知識庫的完整性、豐富性及更新速度。 3. 使用新的技術支援服務流程以提升服務效率及反應速度。 4. 建立每日客戶需求未被滿足的清單之電腦自動回報機制。 5. 提供客戶即時查詢其需求處理狀況。 另外在進行技術支援服務流程再造的時候，需要特別注意的是，要有一個流程負責人來統管整個新流程的運作狀況，董事長與總經理應該給予最高的授權支持，而該負責人應請各相關部門派出經理級以上的人參與流程再造，藉此整合各方意見並凝聚共識，如此才可能增加推行的助力，並將可行性納入考量，不然有可能變成難以執行計畫。 除此之外，還要有搭配相對應的誘因機制，讓相關的人員可以配合與適應此一新的技術支援服務流程，因為人們面對改變總是偏向抗拒的，因此一個新流程的推行需要相對應的誘因制度設計，才能讓大家有動力去從事變革。 關鍵字：工業電腦、技術支援、企業流程再造、流程圖、服務管理

本研究旨在探究學習者在wiki共筆、數位閱讀標註及知識論壇三種不同知識建構導向之電腦支援合作學習環境下，其合作知識建構與問題討論歷程差異，再則這三種不同學習環境是否營造不同知識建構氛圍，以及對支援知識建構是否有所欠缺。最後，針對研究結果提出有效知識建構教學的實施策略與建議。 本研究採實驗研究法，以某國立大學數位碩士在職專班19名研究生為研究對象，分別依序體驗包括wiki共筆、合作式數位閱讀標註及知識論壇三種各具特色的知識建構導向合作學習環境，進行知識建構討論與分享，藉由搜集觀察上述三種學習環境之知識建構歷程討論及成果記錄，進行知識建構及問題解決討論概念編碼後，進行序列分析，依此觀察學員們知識建構演進之歷程，並評估問題討論的互動程度，最後輔以半結構式訪談，與序列分析結果進行交互驗證。 結果發現學員在wiki共筆、合作式數位閱讀標註及知識論壇三種知識建構環境上，共同合作建構產出的知識建構成果均呈現一定品質水準，確實有助於輔助學生之共同知識建構成長。此外，三種電腦支援合作知識建構學習環境中，知識論壇在適當教學設計與學習策略實施下，較能發揮知識建構水準；知識論壇相當適合於問題解決討論；三種知識建構導向之電腦支援合作學習環境，對於完整支援知識建構仍有不足需要強化之處，特別是為促使學員更深入討論，以達更深層的知識建構，需要更好的教學設計與學習策略。 最後本研究根據研究結果，歸納出幾點建議，作為教師在進行合作知識建構教學時，選擇電腦支援合作學習環境之參考，並對未來研究方向提出建議。 / The major purpose of the present study was investigate the learners would have the differences of the progress of knowledge construction and problem discussion, which were under the three different guided knowledge construction in the computer-supported collaborative learning environments, the three learning environments including wiki, knowledge-based annotation learning system, and Knowledge Forum. A secondary purpose of this study was to examine if these three different learning environments would build the variety atmospheres of knowledge construction, and then the deficiency in the computer-supported collaborative learning. Finally, the conclusion drawn above should be proposed the efficient policy and suggestion in relation to the effective teaching of knowledge construction. The method to carry out this study was using an experimental research. The participant in this research were 19 postgraduate students enrolled in executive master of digital systems in one national university. In this experiment, all participants experienced the three different guided knowledge construction in the computer-supported collaborative learning environments in order, each differs from one another, including, wiki, knowledge-based annotation learning system, and Knowledge Forum. Furthermore, all subjects focused on the discussion and shared the progress of the knowledge construction and the results of the accomplishment by searching and observing the above three varied environments. The data of knowledge construction and problem solving were to conceptual encoded and processed the sequential analysis in order to observe the evolution of the progress of knowledge construction of all subjects, and to estimate the level of interaction of problem discussion. Lastly, we used the auxiliary semi-constructed interview and the result of sequential analysis to work with the cross-validation. The findings suggest that the participants produced the conclusions with coordinated knowledge construction which appeared in a certain quality, and this result indeed helped learners to grow-up in the coordinated knowledge construction. Additionally, Knowledge Forum firstly developed well standard under the congruent design and tactic of teaching in these three computer-supported collaborative learning environments. And then Knowledge Forum would be even more proper to discuss the problem solving. However, there is an insufficient part which needs to be strength of the three guided computer-supported collaborative learning environments for the complete computer-supported learning environments, most particularly, to be able to stimulate the subjects would go deep into the discussion in order to achieve the more depth of the knowledge construction, at the same time require the preferable instructional design and learning strategy. On the basis of the findings we sum up the few suggestions: As teacher, this study would be a reference to choose computer-supported collaborative learning environment when teach the coordinated knowledge construction, and then to address the suggestion in relation to the future research.

本研究主要目的為探究知識翻新教學對學生知識共構歷程與想法概念理解的影響。以知識翻新原則（knowledge building principles）為教學理念（Scardamalia, 2002），輔以知識論壇(Knowledge Forum™)的教學設計，提供一個電腦支援協作學習的線上環境，幫助學生體驗與想法互動、想法共構與翻新想法的過程，並紀錄學生多元的想法。本研究採個案研究法，研究對象為某國立大學的學生（N=41），教學時程為一學期。資料來源與分析包含：(1)知識論壇平台活動情形--透過知識論壇分析工具與社會網絡分析法來瞭解學生在平台上的活動量與互動關係；(2)知識論壇中學生想法互動貼文--透過Dean, Hender, Rodgers與Santanenm（2006）建立的想法品質評估量表，以新穎性、可行性、關聯性與具體性四個面向進行學生想法的評分，另外也採用Garrison, Anderson與Archer（2001）提出的實用探索模式，將學生想法分為引發事件、探索、整合、決議四個探索層次進行分析；(3)想法概念開放式問卷前後測--由學生的回答進行開放式編碼，搭配Popper（1972）三個世界知識論（3-World epistemology）的理念框架進行分析。研究結果發現：(1)知識翻新與知識論壇輔助的教學設計，能幫助學生產生想法、進行想法互動，提升想法層次；(2)在知識翻新教學環境下，能提升學生對於想法概念的多元理解，並逐漸將想法視為可被交換與運用來解決真實問題的具體產物；(3)品質與數量並重的想法共構歷程，與學生對於想法概念的理解程度有相輔相成的可能性；(4)知識翻新學習環境，能促使學生提出多元化的想法，透過探索與整合進行知識共構。以上結果顯示，知識翻新環境提供了一個開放討論的空間，讓學生願意分享想法，使想法有改進的可能；此外，也讓學生在過程中，對於想法概念的理解更多元，也更認同想法是具體且共有的社會實體。 / The purpose of this research was to investigate the effects of knowledge building pedagogy on college students’ understanding of the concept of ‘ideas’. Knowledge building principles (Scardamalia, 2002) and Knowledge Forum (KF) technology were employed to support and record the process of idea generation, idea co-construction and idea elaboration. This study adopted a case study design. Participants were 41 undergraduates who engaged in a course titled “Introduction to Living Technology” for one semester. Data sources and analyses included: (1) students’ online activities in KF, which was assessed using Analytic Toolkit and social network analysis in order to understand the process of interaction among students; (2) students’ online discourse data, which were scored using idea evaluation scale (see Dean, Hender, Rodgers & Santanenm, 2006) including four dimensions--novelty, workability, relevance and specificity; moreover, students’ ideas were content-analyzed using practical inquiry model including four levels--triggering event, exploration, integration and resolution (Garrison, Anderson & Archer, 2001); (3) pre-post survey with five open-ended questions that assessed students’ understanding of the concept of ‘idea’ in several aspects (e.g., definition, purpose, and source); moreover, students’ answers were analyzed by open coding using Popper’s three-world epistemological framework. The findings were as follows: (1) knowledge building and technology was found helpful for students to generate ideas, facilitate idea interaction and improve ideas; (2) engaging students in knowledge building was likely to enhance students’ diverse perceptions about ideas, and students tended to see ideas as improvable objects of value for potential knowledge construction; (3) idea co-construction processes, emphasizing both the quality and quantity of ideas, were also likely to enhance students’ diversified perceptions about ‘ideas’; (4) knowledge building environment was beneficial for students to implement knowledge co-construction through exploration and integration. To conclude, knowledge building environment provided an open space for students to share ideas and made idea elaboration more easily; besides, via co-construction process, students’ understanding of the concept of ‘idea’ becomes more diverse, and tended to recognize ideas as tangible, social epistemic entities.

En France comme au Japon, le système d'éducation à destination des enfants handicapés a connu d'importantes transformations ces dernières années, avec le passage à une logique d'éducation inclusive, privilégiant la scolarisation dans les écoles ordinaires plutôt qu'en établissement spécialisé. Cependant, les établissements spécialisés n'ont pas disparu pour autant et le milieu spécialisé se reconfigure pour s'adapter aux changements. Cette thèse s'appuie sur l'idée que ces transformations correspondent à l'aboutissement d'un mouvement historique allant d'une approche catégorielle du handicap (dans laquelle les différentes catégories médicales de déficiences sont sources de spécialisation) à une approche que l'on peut qualifier, par opposition, de généraliste (dans laquelle les catégories de déficiences s'effacent pour laisser la place à une catégorie unique : celle de « besoins particuliers »). Comment les écoles spécialisées, fondées sur la logique de spécialisation catégorielle, se transforment-elles à l'heure du généralisme ? On étudie plus précisément le cas des écoles d'aveugles, emblèmes historiques de la logique de spécialisation. La démarche comparative franco-japonaise permet ainsi d'éclairer une zone d'ombre, peu étudiée, du phénomène international de passage à une éducation inclusive. En mobilisant le champ de la sociologie du travail, plus spécifiquement le concept de compétences, on explore les transformations de l'institution spécialisée vécues par ses représentants: les enseignants spécialisés. On montre ainsi, à travers la comparaison, que le travail des enseignants spécialisés dans ces écoles n'est plus fondé sur une expertise de la déficience visuelle mais sur des compétences relationnelles d'adaptation aux "besoins éducatifs particuliers" de chaque enfant, dans des conditions d'enseignement en profonde mutation, où la mise en œuvre de ces compétences se fait de plus en plus difficile. / In France as in Japan, the educational system for children with disabilities has undergone major changes for the last years, through the development of inclusive education, promoting enrollment in mainstream schools rather than special institutions. However, special schools have not disappeared; they are rather evolving so as to adapt to changes. This work is grounded on the idea that these transformations represent the completion of a historical trend shifting from a category-based approach of disability (in which different medical categories result in specialized treatments) to an approach that, as opposed to the latter, we can call generalist (in which categories of impairments fade into the unified category of “special educational needs”). How do special schools, deeply rooted in the category-based specialized approach, evolve in the days of generalism? We examine the specific case of Schools for the Blind, embodiments of the specialized approach. Through a French-Japanese comparison, we shed light on some gray areas of the international move towards inclusive education. Using the field of work sociology, especially the concept of “skills”, we explore the transformations of special institutions as experiences by their representatives: special teachers. Through comparison, we show that special teachers' work in these schools is no longer based on an expertise of visual disability but relies on the use of relational skills, especially involving adaptation to every child's “special educational needs”, in a context of radically changing teaching conditions, where it becomes growingly difficult to put those skills into practice. / 要旨 : 近年、フランスと日本における障害児教育制度は、インクルーシブ教育の理念に従って、特殊学校よりも通常学校での在学を推進することで、大きく変わってきた。しかし、特殊学校が全くなくなるわけではなく、特殊教育制度はその変化に応じるように形を変えてきた。本論文はその推移が「障害」の取り扱いの歴史的転換と符合するという公準を基本とする。「各障害の医療的な特性に応じる専門的な取り扱いが必要」という障害種別アプローチから「障害の諸カテゴリーが「特別なニーズ」という単一なカテゴリーに溶け込む」という逆の一般的なアプローチへの転換である。障害種別の専門性を踏まえていた特殊学校はその一般的なアプローチの時代にどう応じてきたか。本研究は障害種別アプローチの体現である盲学校を中心の課題とする。日仏比較により、インクルーシブ教育への歩みという国際現象のあまり検討されていない面を調査することを目的とする。そのために、職業社会学（特に、「能力」の理論）に基づき、盲学校の教員が見た特殊教育の変化を分析する。その結果、盲学校教員の仕事は今、視覚障害の専門性よりも「各生徒の個人的なニーズに応じる」対人関係能力を踏まえているということが示され、盲学校における教育条件の変化により、その対人関係能力の実現が難しくなりつつあることが指摘される。