1、<p><b>　　外文翻譯</b></p><p>　　Semantic Advertising for Web 3.0</p><p>　　Material Source：Department of Computer Science Author: Edward Thomas</p><p>　　University

2、 of Aberdeen Aberdeen, Scotland</p><p>　　Abstract: Advertising on the World Wide Web is based around automatically matching web pages with appropriate advertisements, in the form of banner ads, interactive a

3、dverts, or text links. Traditionally this has been done by manual classification of pages, or more recently using information retrieval techniques to find the most important keywords from the page, and match these to key

4、words being used by adverts. In this paper, we propose a new model for online advertising, based around light weig</p><p>　　1 Introduction</p><p>　　Advertising is the main economic force which d

5、rives the development of the World Wide Web. According to a report by Price Water House Coopers advertising revenues totaled $6.1 billion for the fourth quarter of 2008, an increase over the previous year even during an

6、economic recession. Of this, banner advertising accounted for the second largest piece of this revenue, following search revenue with 21 percent of the total market. Search revenue typically uses the keywords entered by

7、the user to ma</p><p>　　This paper outlines a third alternative. By using lightweight semantics on a web page, and RDF descriptions of adverts (and more importantly, what web sites should a particular advert

8、 appear on), combined with some existing semantic web technologies, we can produce an open market for online advertising which offers automatic targeting, with more accurate targeting, combined with the zero cost of entr

9、y which keyword based advertising currently operates.</p><p>　　In this paper, we will first discuss the technical motivations of our approach, before proposing a new model for online advertising, based on li

10、ghtweight embedded</p><p>　　semantics. Furthermore, we propose a system architecture for the new model, based on our scalable semantic reasoning infrastructure TrOWL. Finally we will present two case studies

11、 on Semantic Advertising, and conclude the paper with a discussion of areas of future work.</p><p>　　2 Approaches</p><p>　　Traditional approaches such as strict keyword matching are quite limite

12、d in a sense that it can not disambiguate the keywords in different context. Also, the synonyms or hyponyms have to be manually specified by the advert providers but not automatically derived. Other approaches such as th

13、e one from Double Click require large amount of work for both the advert provider and advertising agency to classify the website’s content and fit it into a pre-defined taxonomy. This is inconvenient for own o</p>

14、<p>　　Our approach attempts to provide a more accurate and easy-to-use matching between web pages and adverts by making use of semantics embedded in both. This can, on the one hand, enable the web developers and ad

15、vert providers describing their documents (web pages and adverts) and requirements in a intuitive and flexible manner, and on the other hand, make use of existing semantic web resources such as ontologies, thesaurus and

16、reasoners to discover the relationsin between. The advert providers no l</p><p>　　This approach includes two major aspects: (1) the automatic reasoning in matching and (2) the manual or automatic annotation

17、of the documents. Like any other web-based application, a crucial technical feature of this service is efficiency. Neither the web publisher, nor the advertising provider would like to an advert matching delays the rende

18、ring of the web page. In the semantic web context, the efficiency of a reasoning-related service is strongly restricted by the language used to describe th</p><p>　　As for the annotation aspect, one can crea

19、te an RDF document and render it as in HTML through transformation techniques such as XSLT. For web developers, it will be more convenient to embed RDF data into normal web pages and further validate them w.r.t. its sche

20、ma. RDFa, an application of RDF bridges the gap between web page composing language such as XHTML and RDF. It can express structured data such as RDF in any markup language by specifying attributes of web page elements.

21、In this paper, we </p><p>　　3 How Does System Work</p><p>　　In this section, we propose the system architecture for semantic advertising and show you how does it works. The semantics embedded on

22、 the page will be converted into RDF graphs, and the constraints given by the advertisers will be rewritten as SPARQL queries. By running each query against the repository of graphs extracted from content, we can produce

23、 a map of the best advertising for each webpage. We propose that the advertising system performs the matching process at the point when</p><p>　　new content or new adverts are added to the system. This can t

24、hen be stored in a cache to improve performance on repeated matching. </p><p>　　When a user requests an advert for a particular page, the system can consult the map of appropriate adverts and select the most

25、 lucrative. Additional techniques could, for example, ensure that a user does not see the same advert on the same site too many times, but this is outside the scope of this paper.</p><p>　　4 Conclusions and

26、Future Work</p><p>　　In this paper we have outlined a vision for publishing advertising specifications and matching these to semantically enabled web pages. We see this as a general approach that can work ac

27、ross a number of different domains without changing the underlying method. There are some issues still to resolve before this can be realized on a large scale.</p><p>　　The first and most difficult problem i

28、s that embedded semantics are currently not widely used on commercial web sites. RDFa is a new format which is not greatly understood, and also there is no compelling application for these semantics which would encourage

29、 large publishers to add them to their web sites. Our hope is that by giving a financial incentive for web sites to deploy RDFa, by improving the matching of advertisements to web pages, we may help to bootstrap these ne

30、w technologies into th</p><p>　　The second issue occurs on highly dynamic web pages, where the content is different for every user. The cost of performing the extraction of RDFa, RDFS reasoning, and matching

31、 this to the most suitable advert would make this method prohibitive for these web sites. There is some research being made into methods for approximate matching and querying.</p><p><b>　　譯文</b>&

32、lt;/p><p>　　Web3.0的語(yǔ)義廣告 </p><p>　　資料來(lái)源：蘇格蘭亞伯丁計(jì)算機(jī)科技大學(xué) 作者：愛(ài)德華?托馬斯</p><p>　　摘要：萬(wàn)維網(wǎng)的廣告是將廣告與適當(dāng)?shù)木W(wǎng)頁(yè)自動(dòng)匹配，主要形式有旗幟廣告、互動(dòng)廣告或文本鏈接。傳統(tǒng)的形式是手工分類(lèi)來(lái)匹配,或是使用最近信息檢索技術(shù),從頁(yè)面上尋找最重要的關(guān)鍵字

33、,并將這些關(guān)鍵字與廣告中的關(guān)鍵字進(jìn)行比對(duì)。在本文中,我們提出一個(gè)新的模型基于語(yǔ)義的在線嵌入式廣告。這將提高萬(wàn)維網(wǎng)上廣告的相關(guān)性，有助于將語(yǔ)義作為一種機(jī)制并把它歸屬為網(wǎng)絡(luò)的一種屬性。此外，我們還提出了一種新的模型——web3.0的語(yǔ)義廣告。</p><p><b>　　1簡(jiǎn)介</b></p><p>　　廣告的是推動(dòng)萬(wàn)維網(wǎng)發(fā)展的主要經(jīng)濟(jì)力量。Price Water Ho

34、use Coopers調(diào)查報(bào)告顯示，2008年第四季度廣告收入總計(jì)為61億美元。即使是在經(jīng)濟(jì)衰退時(shí)期相對(duì)于以前也是增長(zhǎng)的。因此,旗幟廣告占第二大的一塊收入，調(diào)查結(jié)果顯示占整個(gè)市場(chǎng)的21%。搜索引擎的收入一般通過(guò)廣告購(gòu)買(mǎi)客戶購(gòu)買(mǎi)關(guān)鍵詞得到的。這是一個(gè)激烈的廣告商比賽，如果你想關(guān)鍵字能包括同類(lèi)的或者相關(guān)下位的產(chǎn)品，你就得購(gòu)買(mǎi)長(zhǎng)尾關(guān)鍵字。因?yàn)閺V告商只支付每次瀏覽,或是每點(diǎn)擊,沒(méi)有涵蓋大范圍關(guān)鍵詞，用戶購(gòu)買(mǎi)的是簡(jiǎn)單匹配的關(guān)鍵字。由用戶輸入的關(guān)鍵

35、字，以及廣告客戶購(gòu)買(mǎi)的關(guān)鍵字簡(jiǎn)單的匹配可以很容易地理解，因此成為了一個(gè)受廣告商歡迎的路線。匹配到網(wǎng)頁(yè)上的橫幅廣告是一個(gè)更難的問(wèn)題。在這種情況下，我們應(yīng)該考慮到網(wǎng)站的內(nèi)容以及它所處的環(huán)境?，F(xiàn)在有這樣一個(gè)隨收隨付系統(tǒng)，并且在這個(gè)網(wǎng)絡(luò)上發(fā)布廣告的成本非常低。比如被Double Click使用的，與出版商密切合作的一些系統(tǒng)，以稅收的內(nèi)容、嵌入到網(wǎng)也本身的自定義標(biāo)簽或關(guān)鍵字來(lái)對(duì)網(wǎng)站進(jìn)行分類(lèi)來(lái)提高匹配度。以廣告或者出版的方式會(huì)花費(fèi)很大的成本，Dou

36、ble Click和類(lèi)似的網(wǎng)絡(luò)只需要在網(wǎng)站上具有一定數(shù)</p><p>　　本文概述了第三個(gè)選擇。使用輕量級(jí)語(yǔ)義網(wǎng)頁(yè)和RDF描述廣告(更重要的是, 廣告出現(xiàn)在那些網(wǎng)站某一特定地點(diǎn)),結(jié)合現(xiàn)有的語(yǔ)義及其他的網(wǎng)絡(luò)技術(shù), 現(xiàn)有的零成本關(guān)鍵字的基礎(chǔ)廣告，我們可以生產(chǎn)出自動(dòng)瞄準(zhǔn),更準(zhǔn)確的在線廣告。</p><p>　　在本文中,在提出一種基于語(yǔ)義網(wǎng)的在線廣告全新的模型之前，我們將首先討論技術(shù)動(dòng)機(jī)。此

37、外,根據(jù)我們的TrOWL可擴(kuò)展的語(yǔ)義推理得出一種系統(tǒng)新的體系結(jié)構(gòu)模型，并且總結(jié)說(shuō)明本文未來(lái)的討論領(lǐng)域。</p><p><b>　　2 研究</b></p><p>　　傳統(tǒng)的方法,如嚴(yán)格的關(guān)鍵字匹配受關(guān)鍵字在在不同的上下文之間存在歧義的限制。同時(shí),同義詞或下位詞廣告供應(yīng)商不會(huì)自動(dòng)推導(dǎo)，必須由人工所指定的方法才能解決。其他的方法,如一個(gè)來(lái)Double Click的點(diǎn)擊

38、需要廣告機(jī)構(gòu)和廣告代理公司大量的工作，將網(wǎng)站的內(nèi)容放到一個(gè)預(yù)先設(shè)定的分類(lèi)中。這對(duì)我們自己開(kāi)設(shè)的網(wǎng)站來(lái)說(shuō)非常的不方便。此外,當(dāng)您的網(wǎng)頁(yè)實(shí)時(shí)自動(dòng)產(chǎn)生的就很難應(yīng)用這種方法。我們?cè)噲D提供了更加準(zhǔn)確、容易地使用嵌入的方式來(lái)匹配網(wǎng)頁(yè)和廣告之間的語(yǔ)義的方法。一方面，這樣可以直觀和靈活讓網(wǎng)絡(luò)開(kāi)發(fā)者和廣告提供商描述他們的文件（網(wǎng)頁(yè)和廣告），另一方面利用語(yǔ)義網(wǎng)發(fā)現(xiàn)本體與同義詞之間的關(guān)系。因本體分級(jí)系統(tǒng)會(huì)自動(dòng)的幫他們匹配，廣告宣傳供應(yīng)商不用再擔(dān)心同義詞的問(wèn)題

39、。同時(shí)，站長(zhǎng)們也不用一個(gè)個(gè)地對(duì)網(wǎng)頁(yè)進(jìn)行分類(lèi)，這些都可以通過(guò)語(yǔ)義網(wǎng)來(lái)實(shí)現(xiàn)。</p><p>　　這個(gè)研究包括兩大方面:(1)自動(dòng)推理匹配(2)手動(dòng)或自動(dòng)注釋的文件。像提供關(guān)鍵性技術(shù)服務(wù)的其他網(wǎng)絡(luò)應(yīng)用程序一樣。網(wǎng)站發(fā)布商和廣告供應(yīng)商都不希望因?yàn)橐粋€(gè)廣告匹配延遲網(wǎng)頁(yè)網(wǎng)頁(yè)的發(fā)布。在語(yǔ)義網(wǎng)頁(yè)中，服務(wù)的效率嚴(yán)格受用來(lái)描述語(yǔ)義的語(yǔ)言限制。目前，現(xiàn)實(shí)意義上的語(yǔ)義就是W3C開(kāi)發(fā)的RFD格式的網(wǎng)絡(luò)本體語(yǔ)言，許多應(yīng)用成熟的軟件都支持網(wǎng)

40、絡(luò)本體語(yǔ)言的語(yǔ)言描述，但是，很多網(wǎng)絡(luò)本體語(yǔ)言的推廣代價(jià)卻是巨大的，如OWL2 DL和 OWL2DL。另一方面，RFD格式的網(wǎng)絡(luò)本體語(yǔ)言廣泛應(yīng)用在數(shù)據(jù)交換和集成領(lǐng)域中，但是它所應(yīng)用的領(lǐng)域是有限制的。唯一的解決方式就是采用TrOWL,提供可擴(kuò)展的推理和查詢(xún)回答服務(wù),不僅RDF-DL,OWL2 –QL各項(xiàng)目功能顯而易見(jiàn)(以及其他OWL2 profiles2,包括OWL2-EL和OWL2-RL),而且也表達(dá)本體語(yǔ)言O(shè)WLDL和OWL2-DL(語(yǔ)

41、義的推理質(zhì)量也有了良好的保證)。在注釋方面，一個(gè)人能夠創(chuàng)造RFD文件，并且可以把它像XSLT一樣在網(wǎng)頁(yè)中進(jìn)行轉(zhuǎn)化。對(duì)于網(wǎng)開(kāi)發(fā)者,它將更方便RDF數(shù)據(jù)正常嵌入網(wǎng)頁(yè)并進(jìn)一步驗(yàn)證它們w.r.t.。RDFa是XHTML和RDF溝通的橋梁，它無(wú)法用</p><p><b>　　3 實(shí)現(xiàn)的方法</b></p><p>　　在這一節(jié)中,我們提出了該系統(tǒng)的體系結(jié)構(gòu)為語(yǔ)義廣告并介紹其是

42、如何工作。語(yǔ)義嵌入到頁(yè)面中將會(huì)轉(zhuǎn)化為RDF圖表,廣告商的要求將被改寫(xiě)成查詢(xún)的語(yǔ)句。每個(gè)詢(xún)問(wèn)跑向數(shù)據(jù)庫(kù)中提取圖的內(nèi)容,就可以為每個(gè)網(wǎng)頁(yè)制作出一幅最佳的廣告。我們建議對(duì)廣告系統(tǒng)進(jìn)行匹配過(guò)程的內(nèi)容或新的廣告點(diǎn)被添加到系統(tǒng)。這可以被存儲(chǔ)在高速緩沖區(qū)來(lái)提高性能,重復(fù)的匹配。當(dāng)一個(gè)用戶要求做某一特定的廣告頁(yè),系統(tǒng)可以參考一下適當(dāng)?shù)膹V告，選擇最有利的發(fā)布方式。額外的技術(shù),例如,確保用戶不看到同樣的廣告在同一地點(diǎn)太多次了,但這已經(jīng)超出了本文的范圍。&l

43、t;/p><p><b>　　4 總結(jié)和展望</b></p><p>　　本文闡述了如何讓刊登的廣告在網(wǎng)頁(yè)中與語(yǔ)義匹配，我們把他看做是在不同領(lǐng)域中應(yīng)用的一般方法，而不改變?cè)械幕A(chǔ)方式。在實(shí)現(xiàn)這種方法是還需要解決一些問(wèn)題。第一步,也是最困難的問(wèn)題是, RDFa是一種新型的格式，語(yǔ)義現(xiàn)在還沒(méi)辦法讓大多數(shù)人信服，嵌入式語(yǔ)義是目前也沒(méi)有鼓勵(lì)大出版商將它們添加到他們的網(wǎng)站，在商業(yè)