http:^^www.cs.washington.edu^education^courses^473^project.html
来自「This data set contains WWW-pages collect」· HTML 代码 · 共 117 行
HTML
117 行
Date: Mon, 02 Dec 1996 15:59:27 GMTServer: NCSA/1.4.2Content-type: text/html <html><TITLE>CSE 473 PROJECT</TITLE><body><H1>CSE 473 Project<br></h1> <p>Topic descriptions due Friday, May 17.Preliminary demonstrations due Friday, May 24.Final demonstrations and reports due Friday, May 31.<p><H2>Rationale</H2>To do justice to more than just a few topics in artificial intelligencewould require more than a single 3-credit, one-quarter course.The project offers you an opportunity to choose for yourself one topic forin-depth examination. This can be one that we don't get to coverin the course, or it can be one we have already covered.You have the flexibility to work either alone or in a partnershipwith another student in the class. You may be able to apply some ofthe representation and inference techniques we have studiedso far. In any case, you can put to good use the fluency in Lispthat you have developed over the past few weeks.<H2>Topics</H2>You may select any of Chapters 5-14of the text as your topic area. Within the chapter, selecta specific topic, such as the use of augmented transitionnetworks in the Stone World program within Natural LanguageUnderstanding in Chapter 11.If you choose a topic that we have already covered in class,then your project should explore that topic in greater depth.If you choose a topic that we haven't gotten to, then you maywish to use one of the sample programs in the text as a startingpoint, and then implement additional features and perform someexperiments.<p>On Friday, May 17, turn in a project topic description that includesthe following:<ol><li> Name(s)<li> Project name (should be descriptive)<li> Book chapter selected<li> Name of Sample program, if any, you plan to work from.<li> New features or ideas you wish to explore.<li> How the work will be divided, if you are working in a partnership.<li> (Not required, but strongly encouraged:)A bibliographic reference (other than the text)for additional reading on yourchosen topic. Give a two or three sentence summary of what thisreference contains that is relevant to your project.</ol>Whether you select machine learning, natural-language understanding,vision, or neural networks, you may wish to design your projectto work with input data from the World-Wide Web.For example, you may wish to train a neural network todistinguish between different types of HTML files, or you may wishto do pattern recognition experiments on images downloaded fromvarious online galleries or home pages.Jeremy and I strongly suggest that your project be executed ina way that can help others learn something about AI. To makeyour project useful in this regard, give informal demonstrationsto classmates from time to time, and ask them for suggestionson how to make the program better or more understandable.<H2>Preliminary Demonstrations</H2>Be prepared to show a Lisp program on Friday, May 24, thatdoes something interesting. If you are using a sample programfrom the text, your demo should include some novel new feature(s).<p>Alternatively, instead of giving a demo at this time, youmay turn in a 1-page progress report in class.<p><H2>Final Demonstrations and Reports</H2>On May 31, turn in both a diskette with your Lisp program,and a hardcopy report on your project. Here are the guidelinesfor <a href="http://www.cs.washington.edu/education/courses/473/writeups.html">the form of the report.</a><p><p><p><HR><address>tanimoto@cs.washington.edu</address></html>⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?