【摘要】： 隨著移動(dòng)通信的發(fā)展,人們迫切希望能方便地利用手持移動(dòng)設(shè)備訪問(wèn)Web網(wǎng)站。由于移動(dòng)設(shè)備的小屏幕和低帶寬的缺點(diǎn),使得這一難題一直沒(méi)有得到很好的解決。Web網(wǎng)頁(yè)是為桌面計(jì)算機(jī)而設(shè)計(jì)的,桌面計(jì)算機(jī)的屏幕大,翻頁(yè)和滾屏十分方便。因而Web頁(yè)面的內(nèi)容很豐富。但移動(dòng)設(shè)備,例如手機(jī),具有內(nèi)存小、上網(wǎng)帶寬低和屏幕小的先天缺點(diǎn),不可能把一個(gè)完整的Web頁(yè)面直接映射到移動(dòng)設(shè)備屏幕上,因而如何在小屏幕設(shè)備上方便地瀏覽Web網(wǎng)頁(yè)一直是研究的熱點(diǎn)。 分析在Web信息分塊領(lǐng)域中已有的研究工作,其中包括純HTML文檔的轉(zhuǎn)換以及自然語(yǔ)言的總結(jié)等等。大部分算法都是單純的利用HTML標(biāo)簽之間的關(guān)系來(lái)對(duì)Web文檔進(jìn)行解析,這些分塊算法對(duì)固定格式以及規(guī)范標(biāo)簽用法的頁(yè)面來(lái)說(shuō),分塊效率以及正確率會(huì)比較高,但是,現(xiàn)在的大部分Web頁(yè)面都是復(fù)雜,不規(guī)則以及含有太多噪聲的,例如導(dǎo)航鏈接以及廣告鏈接等等,這時(shí)這樣的分塊算法的效果就不是很理想,在此基礎(chǔ)上設(shè)計(jì)出兩種適合于移動(dòng)設(shè)備小屏幕的Web頁(yè)面分塊算法,一種是基于DOM TREE的頁(yè)面分塊算法,將DOM TREE的葉子結(jié)點(diǎn)看做是不可分割的最小語(yǔ)義塊,然后再根據(jù)結(jié)點(diǎn)之間的語(yǔ)義關(guān)系進(jìn)行迭代,通過(guò)語(yǔ)義塊的提取,分割條的檢測(cè)以及權(quán)重的設(shè)置,找到各個(gè)頁(yè)面分塊,再根據(jù)內(nèi)容結(jié)構(gòu)的重構(gòu)形成適合于手持終端顯示的頁(yè)面。另一種算法利用Web網(wǎng)頁(yè)上對(duì)象的位置信息對(duì)信息塊進(jìn)行逐層聚類,生成一棵網(wǎng)頁(yè)分塊樹(shù),再根據(jù)移動(dòng)設(shè)備屏幕的特點(diǎn)把網(wǎng)頁(yè)分塊樹(shù)轉(zhuǎn)換成適合小屏幕瀏覽的頁(yè)面。 設(shè)計(jì)并實(shí)現(xiàn)了面向移動(dòng)設(shè)備的Web頁(yè)面分塊算法的小型實(shí)驗(yàn)系統(tǒng),系統(tǒng)能有效地將Web頁(yè)面轉(zhuǎn)換成適合小屏幕瀏覽的頁(yè)面。
[Abstract]:With the development of mobile communication, people are eager to use handheld mobile devices to visit Web website. Because of the shortcomings of small screen and low bandwidth of mobile devices, this problem has not been solved well. Web web pages are designed for desktop computers, which have large screens and are convenient to flip and scroll. As a result, Web pages are rich in content. But mobile devices, such as mobile phones, have the inherent disadvantages of small memory, low internet bandwidth, and small screens, making it impossible to map a complete Web page directly to a mobile device screen. Therefore, how to easily browse Web pages on small screen devices has been a hot topic. This paper analyzes the existing research work in the field of Web information partitioning, including the transformation of pure HTML documents and the summary of natural languages, and so on. Most algorithms simply use the relationship between HTML tags to parse Web documents. These block algorithms are more efficient and accurate for pages with fixed format and standard tag usage, but, Most of today's Web pages are complex, irregular and noisy, such as navigation links and ad links, and so on. On this basis, two Web page partitioning algorithms suitable for small screen of mobile devices are designed. One is page partitioning algorithm based on DOM TREE, which regards the leaf node of DOM TREE as the indivisible minimum semantic block. Then iterated according to the semantic relationship between nodes, through the extraction of semantic blocks, the detection of segmentation bar and the setting of weights, each page is found into blocks, and then according to the reconstruction of content structure, a page suitable for display by handheld terminals is formed. Another algorithm uses the location information of the objects on the Web web page to cluster the information block layer by layer to generate a web page block tree, and then according to the characteristics of mobile device screen, the web page block tree is converted into a page suitable for small screen browsing. A small experimental system of Web page partitioning algorithm for mobile devices is designed and implemented. The system can effectively transform Web pages into pages suitable for small screen browsing.
【學(xué)位授予單位】：華中科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2007
【分類號(hào)】：TP393.092

【引證文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前1條

1 于長(zhǎng)富;基于頁(yè)面分塊的Web檔案構(gòu)建技術(shù)研究[D];東北大學(xué);2009年

，

本文編號(hào)：2273883