Orange Forum • View topic - Typos in Orange white paper

Typos in Orange white paper

Report bugs (or imagined bugs).
(Archived/read-only, please use our ticketing system for reporting bugs and their discussion.)
Forum rules
Archived/read-only, please use our ticketing system for reporting bugs and their discussion.

Typos in Orange white paper

Postby roschler » Thu Apr 28, 2005 16:03

Hello,


Just got done reading the Introductory Orange White Paper:

http://www.ailab.si/orange/wp/orange.pdf

I found some typographical errors which I felt you might be interested in.

===================== BEGIN EDITS
  • Page 2

    Is: "write short Python scripts for data analysis or enjoy in powerful while easy-to-use"

    Change To: "write short Python scripts for data analysis or enjoy a powerful while easy-to-use"

  • Page 4

    Is: "library, reads the data file and prints out what we were interesting in."

    Change To: "library, reads the data file and prints out what we were interested in."

    Is: "To find out what the right classifications where"

    Change To: "To find out what the right classifications were,"

  • Page 5

    Is: "and we may be interested for how much we have"

    Change To: "and we may be interested in how much we have"

    Is: "Just for some more illustrative examples"

    Change To: "For some more illustrative examples"

    Is: "that uses 10-fold cross validation to test naïve Bayesian classifier"

    Change To: "that uses 10-fold cross validation to test a naïve Bayesian classifier"

  • Page 7

    Is: "condition-based data splitting and component that implements"

    Change To: "condition-based data splitting and a component that implements"

    Is: "Induction procedure for classification trees uses"

    Change To: "The induction procedure for classification trees uses"

    Is: "To see if that makes the difference"

    Change To: "To see if that makes a difference,"

  • Page 8

    Is: "But just like Orange methods are modular and component based"

    Change To: "But just like Orange methods which are modular and component based"

    Is: "For a feeling, a sidebar example shows how"

    Change To: "To get a feeling for this, the sidebar example shows how"

    Is: "Once you get used to, programming"

    Change To: "Once you get used to it, programming"

  • Page 9

    Is: "is real fun."

    Change To: "is really fun."

  • Page NNN

    Is: "A number of Orange components were build as an implementation"

    Change To: "A number of Orange components were built as an implementation"
===================== END EDITS

Also, what is PMML and ROC? Got any links to web pages that explain them?

I really enjoyed the white paper. I am very eager to start working with Orange, especially since I'm a big fan of Visual Programming environments.

Thanks.

Postby Blaz » Thu Apr 28, 2005 17:32

Thanks for the list -- we'll encorporate your suggestions asap.

For the links on PMML check http://www.dmg.org/pmml-v3-0.html. For ROC, there are plenty of refrences that can go back to

- Beck, J.R. and Shultz, E.K. (1986) The use of relative operating characteristic (ROC) curves in test performance evaluation, Arch Pathol Lab Med, 110, 13-20.

whereas ML really started to use them a lot after the papers by Provost and Fawcett (e.g., F. Provost and T. Fawcett (2001). Robust classification for imprecise environments. Machine Learning, 42, 203-231.). See also Peter Flach's tutorial at http://www.cs.bris.ac.uk/%7Eflach/ICML04tutorial/index.html.

Orange has an ROC widget with implementations that directly follow the procedures as proposed in Provost and Fawcett (2001)).

Postby Blaz » Mon May 30, 2005 12:13

I finally found time to fix the typos. The fixed white paper is now online. Thanks again!

Postby Guest » Tue May 31, 2005 1:27

My pleasure. 8)


Return to Bugs