外文資料翻譯--基于本體的應用框架的網(wǎng)絡教育資源庫

1、畢業(yè)設計外文資料翻譯學 院： 信息科學與工程學院 專 業(yè)： 計算機科學與技術(shù) 姓 名： xxx 學 號： xxx (用外文寫)外文出處：_fengyang,youquan chen.ontology-based application framework for network eduction resources library. j.tsinghua science and technology, 2007,(4):3545. 附 件： 1.外文資料翻譯譯文；2.外文原文。 指導教師評語： 簽名： 年 月 日附件1：外文資料翻譯譯文基于本體的應用框架的網(wǎng)絡教育資源庫馮陽1，2，陳友全31（

2、計算機科學與技術(shù)，吉林大學，長春，中國學院）2（信息學院，吉林教師的工程技術(shù)，長春中國研究所）3（機械工程系，長春工程技術(shù)研究所，長春中國）摘要在過去的二十年，隨著互聯(lián)網(wǎng)應用的迅速增長，網(wǎng)絡教育已經(jīng)發(fā)展成為現(xiàn)行教育的主要組成部分。同時，基于語義的網(wǎng)絡教育資源成為教育技術(shù)在計算機和人工智能的重要研究領域。到目前為止，然而，在其申請的過程中了出現(xiàn)了一系列嚴重的瓶頸，例如：對網(wǎng)絡資源的標準是不兼容的，定義不嚴格和沒有有效性的語義，缺乏良好的結(jié)構(gòu)。解決這一問題，本文提出的工作顯示了網(wǎng)絡教育資源庫的新的應用框架設計、實施。通過使用基于本體的語義技術(shù)，它創(chuàng)造了一個新的教育資源的開放式平臺，實現(xiàn)異構(gòu)互操作資

3、源，實現(xiàn)資源的集中管理和分布存儲的內(nèi)容。關鍵詞：網(wǎng)絡教育資源、語義web、本體論一、引言隨著網(wǎng)絡技術(shù)的進一步發(fā)展和應用需求的不斷出現(xiàn)，語義web成為一個在www研究領域的熱點和智能網(wǎng)絡服務的關鍵技術(shù)1。語義網(wǎng)在機器理解的基礎上可以提供關更多的功能，執(zhí)行語義層次的計算。它不僅能提供更清晰和更全面的語義的網(wǎng)絡信息和知識，而且還實現(xiàn)了計算機可以了解網(wǎng)絡信息和知識，以及網(wǎng)絡數(shù)據(jù)處理和網(wǎng)絡服務的智能化。語義網(wǎng)絡在網(wǎng)絡服務，基于代理的分布式計算，基于語義的搜索引擎，基于語義的數(shù)字圖書館等領域成為一個更重要的應用。在同時，隨著應用研究的持續(xù)的深入，作為語義網(wǎng)絡的關鍵技術(shù)本體論技術(shù)趨于成熟，語義元數(shù)據(jù)描述方法

4、的形成和語義網(wǎng)絡的發(fā)展開始進入實際應用階段，例如基于語義的信息表示和語義為基礎的數(shù)字圖書館、數(shù)字博物館有了初步的探索和應用和其他應用程序搜索應用2。網(wǎng)絡教育資源是計算機教育和人工智能的重要研究領域。建構(gòu)主義學習理論認為學習環(huán)境的四個要素是：上下文、協(xié)作、會話和意義建立。隨著計算機網(wǎng)絡和多媒體技術(shù)的發(fā)展，由于建構(gòu)主義理論的影響，人們學習的方式改變了很多，從以傳統(tǒng)的教學資源為基礎的教學到以資源為基礎的學習3。隨著建構(gòu)主義學習理論作為指導，它是人類社會在21世紀的數(shù)字識字發(fā)展的重要方向，建立堅固的網(wǎng)絡教育資源，它可以選擇教學內(nèi)容和教學模式，根據(jù)學生的實際情況、實施個性化指導學生。網(wǎng)絡教育資源起源于電

5、腦輔助教學系統(tǒng)的發(fā)展，其主要特征是，根據(jù)用戶的具體情況提供適當?shù)慕滩?，換句話說，它有個別教學的功能，此功能的實現(xiàn)是通過人工智能的手段4。但是，當前網(wǎng)絡教育資源的也有一些問題，如下：（1）檢索效率低教授的知識過多而且廣泛，傳統(tǒng)的基于關鍵字的搜索技術(shù)無法滿足需要。使用適當?shù)谋倔w來描述教育資源庫和建立以本體論為基礎的知識架構(gòu)來解決這些問題是非常必要的。（2）弱語義高的語義對于應用不同的教育平臺之間的交流是必要的。使交流和知識共享成為困難的原因是描述教學知識的基本標準并不是任何時間都是一樣的，知識以不同的方式表示5。（3）缺乏情報當前的很多教學系統(tǒng)不能學生做有針對性的學習，使教師無法準備合適的學習內(nèi)容

6、跟隨學生的認知模式，通過了解他們的認知特點和學習的變化來引導學生。為應對當前的教育資源建設和應用需求的問題，本文介紹了語義的關鍵技術(shù)是建設教育資源，提出了基于語義網(wǎng)絡技術(shù)為基礎的教育資源的主要框架，開始于教育資源建立的有意義的方面，如基于語義的元數(shù)據(jù)描述、教育、資源庫本體模型、資源池之間的模式匹配、討論了框架和關鍵技術(shù)的核心理念。二、本體在本節(jié)中，提出了最容易被人接受的本體理解。由于本體是用于描述在語義和知識層次的信息系統(tǒng)中的概念模型，它已被用于許多領域。由于本體是不是一個靜態(tài)的模型，它必須有能力獲取意義和關系的變化。在人工智能領域的研究，有許多關于本體的定義。其中，最經(jīng)常采用的是格魯伯的關于

7、本體的定義，即本體是一個明確的規(guī)范概念化6。一個本體o是一個五元組o=(c, i, r, f, a)，其中c，i是分別的概念和實例。r是一個關系集，f是c上面的函數(shù)，a是一系列（本體論）公理- 指定了在一些論述領域的詞匯的解釋。以格魯伯為基礎的定義，還有一些本體匹配類型：概念與概念、屬性的概念、屬性與性質(zhì)，場景和約束等6。本體論在教育資源庫的構(gòu)造中起著重要的作用，它可以實現(xiàn)三方共享5：(1)資源內(nèi)容共享。這意味著，資源庫像每個人提供資源，與此同時，從每個人那里獲取自己本身沒有的資源。(2)技術(shù)分享。這意味著，當不同的組織或個人建立他們自己的資源庫，如果他們按照相同的標準，同樣的資源表示方式和相

8、同的屬性標記方法，那么他們可以共享同一接口。(3)概念共享。這意味著，在實現(xiàn)語義理解，確保不同的對象建立，維護和使用過程中這方面的知識分享同一個概念。三、框架概述 我們?yōu)榫W(wǎng)絡教育資源庫設計了一個以本體論為基礎的應用程序框架。它可以實現(xiàn)教育資源的訪問和重用，實現(xiàn)教育資源和優(yōu)化資源之間的生命周期管理機制的分布式資料庫的互操作性。最重要的是，它提供基于語義的搜索，因為這種制度框架提供一個綜合的、面向全球的本體表示的概念和機制，以及基于語義的資源規(guī)劃。該框架結(jié)構(gòu)由三部分組成。如圖1所示。這一框架的更多的技術(shù)細節(jié)將在第四部分討論。a、表示層它是用戶和系統(tǒng)之間的互動界面。用戶通過瀏覽器訪問系統(tǒng)。用戶界面負

9、責接收查詢和顯示搜索結(jié)果給用戶。在這層，雖然他們是被采用的，但是本體保護用戶的網(wǎng)頁。因此，網(wǎng)頁語義信息對用戶是透明的。為了實現(xiàn)顯示與邏輯分離這一目標，jsp和javabean技術(shù)已被用于在本層的開發(fā)。b、邏輯層作為主要的應用程序邏輯層，它進行系統(tǒng)的知識檢索。它由五個部分組成，即本體管理組件、語義分析組件、推理引擎、查詢組件、web信息訪問組件。在用戶的查詢請求被搜索引擎提交后，語義分析組件獲得概念和語義關系，用戶通過語義分析需求，然后jena owl的推理引擎負責對這個概念和語義關系的基礎上的推理，最終查詢beans類訪問此資源的本體庫和查詢結(jié)果返回給用戶。c、數(shù)據(jù)層數(shù)據(jù)層是由三個部分組成：本

10、體庫、教育資源描述數(shù)據(jù)庫和mysql/db2/oracle數(shù)據(jù)庫。作為知識資源本體的存儲介質(zhì)，這一層是負責建立和從新結(jié)構(gòu)性的描述owl為基礎的知識本體，這是知識檢索的直接來源。此外，為了實現(xiàn)知識共享和重用，它會進行語義標注相關資源的網(wǎng)頁并存儲在數(shù)據(jù)采集元件本體庫中。圖1、本體論為基礎的框架的網(wǎng)絡教育資源庫四、框架的關鍵技術(shù)本節(jié)中對這一框架的關鍵技術(shù)進行介紹。它包括網(wǎng)絡服務、用戶的要求處理、語義分析和推理引擎。a、網(wǎng)絡服務表示層也可以作為網(wǎng)絡服務層稱，其主要任務是實現(xiàn)與用戶的交互。在很多資源建設方式的情況下，選擇網(wǎng)絡服務技術(shù)，以集成分布式有效的教育資源庫和共享它們之間的信息。在各種平臺的異構(gòu)基礎

11、上，網(wǎng)絡服務可用來建立一個共同的與平臺無關，語言無關的技術(shù)，不同平臺的應用程序依賴于去實現(xiàn)連接以及它們之間的融合。在整個資源庫，作為服務供應商，每個地方的資源制定自己的服務和功能補充全球系統(tǒng)，它的uddi（通用描述，發(fā)現(xiàn)和集成）。uddi是負責注冊和發(fā)送工作。在此，所有的信息通過soap（簡單對象訪問協(xié)議）被發(fā)送到整個系統(tǒng)。b、用戶請求處理在我們的框架，用戶通過普通的連接方法進行資源搜索，這些領域被限制或自然聲明。學習對象屬于不同的本地資源，它們在地理位置上分開。不同的資源庫通常根據(jù)不同的計算機平臺、數(shù)據(jù)格式和編程語言建立，它提供了他們自己的機制來查詢和訪問資源。在這種情況下，中央服務器需要一

12、定的機制來確定當?shù)氐目商峁┓掌髻Y源庫，然后用戶的請求將被發(fā)送到本地服務器。在具體系統(tǒng)設計我們指的是一系列數(shù)學的方法8。圖2說明了用戶要求如何由這些模塊處理。c、語義分析 在我們的框架中，語義分析是負責通過搜索引擎，提交用戶的請求。該機器的用戶表示需要語義本體庫查詢，分析用戶意圖和使用jena owl推理機制來完成。由于用戶查詢在一定程度上的不確定性，根據(jù)用戶的需要擴大查詢的概念集合是可行的。語義擴展規(guī)則的定義為：如果o是literlas類型，則s是擴張中心，查詢的概念可根據(jù)學的概念和屬性關系獲得。如果s不存在，這些概念是不能采取。 如果o是一種資源，那么o是擴張中心，查詢的概念可根據(jù)o的概念

13、和屬性關系獲得。查詢處理是語義檢索最重要的部分。它可以被描述為以下國際扶輪模型：本體模型和基于本體的網(wǎng)絡資源、識別模型。在這些中，網(wǎng)絡資源r和查詢q是基于本體o構(gòu)建的，他們是有關的，當且僅當r滿足q：它描述了r和o意味著邏輯上查詢q：or-q9。圖2、用戶要求處理有兩種行之有效的方法，可以把查詢關鍵字轉(zhuǎn)換成本體描述。文獻10提出了一種基于語義的檢索方法：本體匹配圖，也就是語義實在本體、關系和本體圖之間的相關性提出來的。根據(jù)這些語義的相關性，本體圖匹配的框架，它可以計算這些相互關系設計。在文獻11中給出了另外一種語義檢索想法。它提出了一個搜索的體系結(jié)構(gòu)，它結(jié)合激活擴散應用于一個特定領域的語義模型

14、古典搜索技術(shù)。給定的本體，權(quán)衡被分配給連接基于特定本體的屬性，因此他們測量的關系的強度。激活擴散技術(shù)用于尋找在本體相關概念給出的概念和相應的初始值，初始設置激活。這些初始值是從應用到與本體的概念相關的數(shù)據(jù)古典搜索的結(jié)果。兩個測試例實施，具有非常積極的成果。有人還指出，提出的混合激活擴散，結(jié)合象征和子象征性的辦法，當和每個單獨的方法相比取得了較好的結(jié)果11。d、推理引擎這是整個框架的關鍵控制模塊。通過推理和搜索，將得到知識用戶的需要，與此同時，智能教學方法和策略也可以得到。推理引擎負責解析和推理owl文件負責。這樣做的目的從一般的文件中讀取本體，用特定的模式存儲是為了方便處理，然后在按照一定的規(guī)

15、則進行基于本體的語義推理，這是語義檢索的關鍵一步。此過程使用jena開發(fā)套件來執(zhí)行。如圖3所示，推理機的工作原理是：推理引擎注冊機制，推理機的基礎上創(chuàng)建基本的rdf三元組和本體。經(jīng)過這一過程，模型對象中包含推理機制，inefreneegnarh和ingfraph可以生成。在jena中，garph也被稱為模型及其表現(xiàn)形式是modelinterafee。然后，它可以對運行和處理這一模式利用該模型api和onotlogy空氣污染指數(shù)，因此語義層次上的信息檢索可以實現(xiàn)。圖3、推理機的工作原理五、結(jié)論與未來在這個文件中，我們提出了一個新的網(wǎng)絡教育資源庫的應用框架設計、實施。通過使用基于本體的語義技術(shù)，我

16、們創(chuàng)建了一個新的教育資源的開放平臺，實現(xiàn)異構(gòu)資源之間的互操作，實現(xiàn)資源的集中管理和分布存儲的內(nèi)容。在此基礎上，資源庫可以成為共享、分布、開放和異構(gòu)資源之間進行互操作得到支持。隨著教育的快速發(fā)展，課程也在發(fā)生變化。生產(chǎn)和教育資源的更新是越來越頻繁。通過利用本體論表達和工作流程的標準化，建立教育資源庫的基礎上的共同標準，廣泛的資源共享和再利用可以達到的理念和形成共識。這將是建設和發(fā)展教育資源庫的重要方向。在今后的工作中，我們將沿著三個方向延伸我們當前的工作。第一個是該體系結(jié)構(gòu)的改善。第二個方向是實現(xiàn)當設計到我們這個框架的數(shù)據(jù)集可用時，設計出更復雜的本體。最后是與其他網(wǎng)絡教育資源庫進行比較。鳴謝這項

17、工作是受國家自然科學基金委員會中國委員會（nsfc），410073no.60873044和專項科研基金批準號：20060183044下的中國高等教育的教育博士課程。附件2：外文原文（復印件）ontology-based application framework for network education resources libraryfeng yang1,2, youquan chen31(college of computer science and technology, jilin university, changchun, china)2(college of informat

18、ion, jilin teachers institute of engineering and technology, changchun china)3(department of mechanical engineering, changchun institute of engineering technology, changchun china), qabstractin past twenty years, as the rapid growth of internet application, n

19、etwork education has grown a main part of current education. at the same time, semantic-based network educational resource became an important research field in computer education technology and ai. so far, however, there is a serious of bottlenecks in the progress of their applications, e.g. the cr

20、iterion of web resource is not compatible, the semantic definition is not strict and the validity and well-structure are absent. addressing this issue, the work presented in this paper shows the design, implementation of a new application framework for network education resources library. by using o

21、ntology-based semantic technology, it creates an open platform for new educational resources, achieves interoperation between heterogeneous resources and realizes the centralized management of resources and contents distributed stored. keywords- network education resources; semantic web; ontology i.

22、 introduction with the further development of network technology and the unceasing emergence of application requirements, semantic web becomes a hotspot in www research field and a key technology of intelligent network services 1.the semantic web can provide more extensive application functions on t

23、he basis of the machine understanding and perform computation at the semantic level. it does not only give clearer and more completely semantic to the network information and knowledge, but also achieves that computers can understand the network information and knowledge, and the intelligentization

24、of the network data processing and network services. semantic web becomes a more important application in the fields as network services, agent-based distributed computing, semantic-based search engine, semantic-based digital library and so on. at the same time, with the continuous deepening of the

25、applied research, the ontology technique as a key research of semantic web tends to be sophisticated, and there is a formation of semantic metadata describing methods, and development of semantic web begins to enter the practical application stage, such as that there is an initial exploration and ap

26、plication of semantic-based information representation and semantic-based searching in digital libraries and digital museums and other applications2. network educational resource is an important research field on computer education technology and ai. constructivist learning theory holds that the fou

27、r elements of the learning environment are context, collaboration, conversation, and meaning built. with the development of computer network and multimedia technology, the ways people learn changed a lot under the influence of the constructivist theory, from traditional teaching to the resource-base

28、d teaching, and then to resource-based learning 3. with constructivist learning theory as a guide, it is an important direction of the development of human society in the 21st century digital literacy to build the stablish network educational resources which can select teaching content and teaching

29、pattern according to the actual situation of students, implement individualized instruction for students. network educational resources originated in the development of computer-aided education systems, its main feature is that according to the specific circumstances of the user to provide appropria

30、te teaching material, in other words, it has the function of individual teaching, this feature is achieved by means of ai 4. but, there are some problems of current network educational resources as follows: (1) low retrieval efficiency for teaching knowledge is excessive and extensive, the tradition

31、al keyword-based search technology can not meet the needs. it is necessary to use appropriate ontology to describe educational resources library and build ontology-based knowledge architecture for solving these problems. (2) weak semantic high semantic is necessary for communication between differen

32、t educational application platforms. the reason why making the exchange and sharing of knowledge difficult is that standard of describing the teaching knowledge base is not the same all the time and the knowledge represents in different ways5. (3) lack of intelligence a lot of current teaching syste

33、ms can not provide student information automatically for students to do targeted studying, and make teachers unable to prepare the suitable learning content following the cognitive model of students, and guide the student to learn by their cognitive characteristics and the changes in learning. in re

34、sponse to problems of the current education resources construction and application needs, this paper introduced key technology of semantic to the building of education resources, presented the primary frame of education resources based on semantic web technologies, started with the significant aspec

35、ts of the building of education resources, such as semantic-based metadata description, education and resource library ontology model, pattern matching between the resource pool, and discussed the core concept of the framework and key technology. ii. ontology in this section the most generally accep

36、ted understanding of ontology is presented. since ontology is a conceptual model which is used for describing information systems on the level of semantics and knowledge, it has been used in many fields. since ontology is not a static model, it has to have the ability to capture the changes of meani

37、ngs and relations. in research field of ai, there are many definitions of ontology. among them, the most frequently adopted is gruber, that is, an ontology is an explicit specification of a conceptualization 6. an ontology o is a 5-tuple o=(c, i, r, f, a), where c, i are concepts and instances respe

38、ctively. r is a relation set, f is a function above c, and a is a set of (ontological) axioms-specifying the intended interpretation of the vocabulary in some domain of discourse. base on grubers definition, there are some types of ontology matching: concept to concept, property to concept, property

39、 to property, scenario and constraint and so on 6.ontology plays an important role in the construction of the educational resources library, which can achieve tripartite sharing 5: sharing of resources contents. it means that resource libraries provide resources to each other, at the same time get r

40、esources they do not have from each other. sharing of technology. it means that when different organizations or individual build their own resources library, if they can follow the same standard , the same resource representation mode and the same attribute marking method, then they can share one in

41、terface. sharing of concepts. it means that for achieving semantic understanding, make sure that the different objects being established, maintened and used of knowledge during this area share the same concept. iii. framework overview we designed an ontology-based application framework for network e

42、ducation resources library. it can realize the access and reuse of education resources, achieve the interoperability between the distributed repositories of educational resources and optimize resource life-cycle management mechanism. most importantly, it provides semantic-based search, because such

43、a system framework provides an integrated, global-oriented concepts and ontology representation mechanisms, and the semantic-based resource formulation. the structure of the framework consists of three parts. it is shown in figure 1. more technology details of this framework will be discussed in par

44、t iv. a. representation layer it is the interactional interface between user and system. users access the system through a browser. the user interface is responsible for receiving queries and displaying the search results to the user. in this layer, the ontologies are shielded to users web page, tho

45、ugh they are adopted. so the semantic information of web page are transparent to the user. in order to achieve this separation of display and logic, jsp and java bean technology have been used in the development of this layer. b. logic layer as the main application logic layer, it carries on the ret

46、rieval of system knowledge. it consists of five parts, namely, ontology management component, semantic analysis component, inference engine, query component, and web information access component. after users query requests are submitted by search engine, semantic analysis component obtains the conce

47、pts and semantic relations that user needs through the semantic analysis, and then jena owl inference engine takes on reasoning on the basis of this concepts and semantic relations, ultimately the query beans accesses this resource ontology library and returns the query results to the user. c. data

48、layer data layer is composed of three parts: ontology library, education resource description database and mysql/db2/oracle database. as the storage medium of knowledge resource ontology, this layer is responsible for creating and refining a structural description of the owl-based knowledge ontology

49、, which is the direct source of knowledge retrieval. in addition, in order to achieve knowledge sharing and reuse, it will carry out semantic annotation to the relevant resources in the web page and store them in the ontology library by the data acquisition components.fig 1. ontology-based framework

50、 for network education resources libraryiv. key technologies of framework in this section the key technologies of this framework are presented. it includes web services, processing of user request, semantic parsing and inference engine. a. web services representation layer may also be known as the n

51、etwork service layer, its main task is to achieve interaction with the user. under the case of a variety of the resource building ways, the web service technologies are selected in order to integrate distributed resource library of education effectively and share information between them. on the bas

52、is of a variety of heterogeneous platforms, web services are available to build a common platform-independent, language-independent technical level, which applications of different platform rely on to implement the connection and integration between them. throughout the resource library, as a servic

53、e provider, each local resource make their own services and features promulgate to the global system, which uddi (universal description discovery and integration) on it. and then the uddi is responsible for the registration and sending. in this, all the information is send to the entire system throu

54、gh the soap (simple object access protocol). b. processing of user request in our framework, user takes on resource search by an ordinary combination method that fields are limited or natural statement. learning objects belong to different local resources, they are geographically separated. differen

55、t resource libraries are usually built on different computer platforms, data formats and programming languages, which provide their own mechanisms to query and access resources. in this case, the central server needs a certain mechanism to orient local resource libraries that can provide server, and

56、 then user request will be sent to local servers. in the specific system design, we refer to a number of mathematical methods 8. figure 2 illustrates how a user request to be processed by these modules. c. semantic parsing in our framework, semantic parsing is responsible for the transition from the

57、 user requests submitted by search engine to concepts of ontology. the conversional process of user semantic to machine presentation need query ontology library, analyze user intents and use jena owl inference mechanism to complete. as user query are uncertainty in a certain degree, it is feasible t

58、hat expand the concept set of queries according to the needs of users. semantic expansion rule is defined as: if o is literlas type, then s is the center of expansion, the concepts for a query can be obtained according to the concepts and attribute relations of s. if s does not exist, the concepts are not to be taken. if o is a reso