網際網路發展至今，不僅成為應用程式開發的主要平台，也是人們獲取資訊最主要的管道。大量的網路爬蟲 (Web Crawler) 被建構來抓取網路上的資訊，藉以整合提供加值的資訊服務。根據網路安全公司 Imperva 及 Barracuda 統計，網際網路上有半數的流量來自網路機器人。為了防範惡意機器人的攻擊，網頁設計的架構日益複雜，透過 JavaScript 開發技術的使用，改變網頁嵌入和呈現數據的方式。這對於建構加值型網路應用服務來說，無疑是相當大的挑戰。例如在網址不變的情況下動態更新網頁內容。如何克服這類型的網站的網頁抓取是本文研究的主題。

為了取得動態網頁的資料，本研究在 Chrome extension 上開發一套模擬使用者點擊流程的系統，透過 Chrome 擴充套件來記錄使用者的點擊與輸入，達到重現使用者在網頁瀏覽時的操作並抓取網頁資料。幫助使用者在不用寫程式碼的前提下，成功抓取網頁資料並提供定期自動抓取的功能。改善 WebETL System，對高互動性及一頁式網站的動態網頁下載問題，達到資料擷取及重覆使用的目的 (Data extraction And Reuse)。針對自動分頁偵測 失敗與政府網址連結與Alex統計的熱門網站共75個動態網頁中，成功的抓取70個，有93.33%的成功率。

Since the development of the Internet, it has not only become the main platform for application development, but also the most important channel for people to obtain information. A large number of web crawlers are constructed to crawl information on the Internet, in order to integrate and provide value-added information services. According to statistics from Internet security companies Imperva and Barracuda, half of the Internet traffic comes from cyberbots.In order to prevent attacks from malicious robots, the architecture of web page design is becoming more and more complex. Through the use of JavaScript development technology, change the way web pages embed and present data. This is undoubtedly a considerable challenge for the construction of value-added network application services. For example, the content of the webpage is dynamically updated when the URL is unchanged. How to overcome web crawling of this type of website is the subject of this article.

In order to obtain the information on dynamic web pages, this research developed a system that simulates the user's click process on the Chrome extension. Use Chrome extensions to record user clicks and input, so as to reproduce the user's operations during web browsing and grab web data. Help users successfully crawl web page data without writing code and provide regular automatic crawling functions. For the dynamic webpage download problem of highly interactive and one-page websites, the purpose of data extraction and reuse is achieved. For automatic page detection failures, government URL links, and Alex’s statistics of 75 dynamic web pages, 70 were successfully crawled, with a success rate of 93.33%

[1] Thoma Bravo. Imperva. https://www.imperva.com/blog/bad-b ot-report-2021-the-pandeniic-of-the-internet/,2002.
[2] Berislav Kucan. helpnetsecurity. https://www.helpnetsecurity.com/2021/09/07/bad-bots-internet-traffic/, 1998.
[3] Google. Chrome extension. https://chrome.google.com/webstore/category/extensions, 2009.
[4] Cheng-Ju Wu. Large-scale web data api creation via automatic paginationrecognition -a case study on announcement monitoring. Master's thesis, National Central University, Taoyuan, Taiwan, 2021.
[5] Yu-An Chou. Web data etl system with unsupervised extractiori. Master's thesis, National Central University, Taoyuan, Taiwan, 2018.
[6] S. Chaudhari, R. Aparna, V. G. Tekkur, G. L. Pavan, and S. R. Karki. Ingredient/recipe algorithm using web mining and web scraping for smart chef. In 2020 IEEE International Conference on Electronics, Computing and Communication Technologies(CONECCT)),pages 1-3, Bangalore, India, 2020. IEEE.
[7] K. Sundaramoorthy, R. Durga, and S. Nagadarshini. Newsone — an aggregation system for news using web scraping method. In 2017 International Conference on Technical Advancements in Computers and Communications (ICTACC),pages 1-4, Melmaurvathur, India, 2017. IEEE.
[8] L. R. Julian and F. Natalia. The use of web scraping in computer parts and assembly price comparison. In 2015 3rd International Conference on New Media (CONMEDIA), pages 2-4, Tangerang, Indonesia, 2015. IEEE.
[9] Oviliani Y. Yuliana and Chia-Hui Chang. Dcade: divide and conquer alignment with dynamic encoding for full page data extraction. Applied Intelligence, pages 1-25, July 2019.
[10] wikipedia. Ajax. https://en.wikipedia.org/wiki/Ajax_(programming), 1999.
[11] wikipedia. Xpath. https://en.wikipedia.org/wiki/XPath, 1998.
[12] wikipedia. Css. https://en.wikipedia.org/wiki/CSS, 1996.
[13] wikipedia. Http. https://en.wikipedia.org/wiki/HTTP, 1996.
[14] Shore Group Associates. shoregrpleaderboard. https://www.shoregrp.com/blog/top-free-no-code-web-scraping-tools, 2006.
[15] Hevo. Hevo・ https://hevodata.com/leam/8-best-web-scraping-tools/, 2017.
[16] webhose.io. webhose.io. https://webhose.io/, 2015.
[17] Gil Elbaz. commonerawl. https://commoncrawl.org/, 2011.
[18] Shore Group Associates. shoregrp. https://www.shoregrp.com/, 2006.
[19] Proxy Crawl, scraperapi. https://www.scraperapi.com/, 2017 ・
[20] Zyte (formerly Scrapinghub). Scrapy. https://scrapy.org/, 2008.
[21] Web Scraper. Web scraper, https://webscraper.io/, 2017.
[22] Octoparse. Octoparse, https://www.octoparse.com/, 2016 ・
[23] Simplescraper. Simplescraper. https://simplescraper.io/, 2017.
[24] ParseHub. Parsehub. https://www.parsehub.com/, 2013.
[25] Dexi.io. Dexi.io. https://www.dexi.io/, 2015.

[26] Mozenda. Mozenda. https://www.mozenda.com/, 2007.
[27] Content Grabber. Content grabber. https://contentgrabber.com/Manual/understanding_the_concept.htm, 2020.
[28] Import.io. ImpoTt.io. https://www.import.io/, 2013.
[29] Enlyft. enlyftselenium. https://enlyft.com/tech/products/selenium, 2021.
[30] pcloudy. pcloudy. https: //www.pcloudy.com/blogs/best-selenium-python-fi'ameworks-for-test-automation-in-2021/, 2021.
[31] Holger Krekel. pytest. https://docs.pytest.org/en/6.2.x/, 2004.
[32] Pekka Klarck and Janne Harkonen. Robot framework. https://robotframework.org/, 2008.
[33] Benno Rice. behave. https://behave.readthedocs.io/en/stable/, 2012.
[34] St eve Purcell. pyunit. http://pyunit.Sourceforge.net/, 2001.
[35] Jason Pellerin. nose2. https://docs.nose2.io/en/latest/, 2010.
[36] J hong li Ding. Page-level information extraction system. Master's thesis, National Central University, Taoyuan, Taiwan, 2015.
[37] Bing Liu, Robert Grossman, and Yanhong Zhai. Mining data records in web pages. In Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining) pages 601-606, New York, 2003. ACM.
[38] Chia-Hui Chang, Tian-Sheng Chen, Ming-Chuan Chen, and Jhung-Li Ding. Efficient page-level data extraction via schema induction and verification. In Pacific-Asia Conference on Knowledge Discovery and Data Mining, pages 478-490, Switzerland, 2016. Springer.
[39] Elsevier. sciencedirect. https://www.sciencedirect.com/, 1997.
[40] Alexa Internet. Alexa. https://www.alexa.com/topsites, 1996.