First Image-recognition Software

　　(1) Dartmouth researchers and their colleagues have created an artificial intelligence software that uses photos to locate documents on the Intemet with far greater accuracy than ever before.

　　(2) The new system, which was tested on photos and is now being applied to videos, shows for the first time that a machine learning algorithm (运算法则)for image recognition and retrieval is accurate and efficient enough to improve large-scale document searches online.The system uses pixel (像素) data in images and potentially video- rather than just text- to locate documents.It learns to recognize the pixels associated with a search phrase by studying the results from text-based image search engines. The knowledge gleaned (收集) from those results can then be applied to other photos without tags or captions (图片说明), making for more accurate document search results.

　　(3) "Over the last 30 years," says Associate Professor Lorenzo Torresani, a co-author of the study, "the Web has evolved from a small collection of mostly text documents to a modem,massive, fast-growing multimedia data set, where nearly every page includes multiple pictures or videos. When a person looks at a Web page, he immediately gets the gist (主旨) of it by looking at the pictures in it. Yet, surprisingly, all existing popular search engines, such as Google or Bing, strip away the information contained in the photos and use exclusively the text of Web pages to perform the document retrieval. Our study is the first to show that modem machine vision systems are accurate and efficient enough to make effective use of the information contained in image pixels to improve document search."

　　(4) He researchers designed and tested a machine vision system-- a type of artificial intelligence that allows computers to learn without being explicitly programmed--that extracts semantic(语义的) information from the pixels of photos in Web pages. This information is used to enrich the description of the HTML page used by search engines for document retrieval. The researchers tested their approach using more than 600 search queries (查询) on a database of 50 million Web pages. They selected the text-retrieval search engine with the best performance and modified it to make use of the additional semantic information extracted by their method from the pictures of the Web pages. They found that this produced a 30 percent improvement in precision over the original search engine purely based on text.

　　23～26概括大意

　　23. Paragraph 1__________

　　24. Paragraph 2 __________

　　25. Paragraph 3 __________

　　26. Paragraph 4 __________

　　A. Function of the new system

　　B. Improvement in document retrieval

　　C. Publication of the new discovery

　　D. Problems of the existing search engines

　　E. Popularity of the new system

　　F. Artificial intelligence software created

　　27～30完成句子

　　27. The new system does document retrieval by __________.

　　28. The new system is expected to improve precision in __________.

　　29. When performing document retrieval the existing search engines ignore __________.

　　30. The new system was found more effective in document search than the __________.

　　A. using photos

　　B. description of the HTML page

　　C. current popular search engines

　　D. document search

　　E. information in images

　　F. machine vision systems

　　23.F。题干：第一段的主要内容是__________。第一段就一句话，大意是说Dartmouth的研究者及其同事发明了一种人工智能软件，该软件使用图片来定位网络文件，准确性较以前更高，故F(已经发明的人工智能软件)是正确答案。

　　24.A。题干：第二段的主要内容是__________。第二段首旬是主题句，新系统已在图片上进行过检测，现在正应用于视频，第一次表明一种学习图像辨识和检索运算法则的机器能精确和有效地改进大规模的网络文件搜索。该句讨论的是新系统的功用，故A(新系统的功能)是正确答案。

　　25.D。题干：第三段的主要内容是__________。第三段的主题句为中间的句子，“yet”之前讨论的是网页的发展，“yet”之后讨论的则是现存的搜索引擎，诸如谷歌和必应等，无法使用图像中的信息而只能用文字进行文件的检索。因而，D(现存搜索引擎的问题)是正确答案。

　　26.B。题干：第四段的主要内容是__________。第四段没有主题旬，第一句讲新系统的机制，第二句讲新系统在文件检索中的功能，第三句讲研究者对新系统的检测以及检测效果，检测发现准确性提升了30%。故B(文件检索的改进)是正确答案。

　　27.A。题干：新系统通过__________进行文件检索。利用出题顺序与段落顺序一致的原则，根据第一段可知，Dartmouth的研究者及其同事发明了一种人工智能软件，该软件使用图片(use photo)来定位网络文件，所以答案为A(使用图片)。

　　28.D。题干：新系统有望在__________方面提升精确性。利用题干关键词new system和improve precision可以定位到第二段第一句，可知新系统己在图片上进行过检测，现在正应用于视频，第一次表明一种学习图像辨识和检索运算法则的机器能精确和有效地改进大规模的网络文件搜索(document searches)，故选D(文件搜索)。

　　29.E。题干：在进行文件检索时，现存的搜索引擎忽视了__________。利用题干关键词the existing search engines可以定位到第三段的第三旬，也就是yet引导的句子，可知现在流行的搜索引擎都跳过了包含在图片中的信息而只能用网页的文字进行文件检索，由此可知这些搜索引擎忽视了利用图片信息进行文件检索，所答案为E(图片中的信息)。

　　30.C。题干：新系统比__________在文件检索方面更有效。利用题干关键词new system和more effective可以定位到第三段的最后一句，可知与那些目前流行的搜索引擎如谷歌和必应相比，新系统有效使用包含在图像像素中的信息来检索文件，更加准确和有效。因此，该题答案选C(当前流行的搜索引擎)。

（编辑：何莹莹）