CS404_Exam_2_110402

CS404_Exam_2_110402 - CSc 401 Data Mining Exam #2 November...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
CSc 401 – Data Mining Name:______________________________ Exam #2 November 4, 2002 Score:_________________/100 Directions: Carefully answer each of the following questions. This is an open-book, open-note exam. You may use calculators. You are NOT to get help from others. Points will be assigned on answer quality as well as answer correctness. CLEARLY show all work. 1. Consider the following data: Calculate the gain(C) when attribute C is considered to have two sets of values; those less than or equal to (164 + 180)/2 and those > (164 + 180)/2. CLASS is the classification attribute. Clearly show work. [10 pts.] This was on previous exam. Skipping for now # A B C CLASS 1 b u 202 P 2 a u 43 N 3 a u 280 N 4 b u 100 P 5 b u 120 P 6 b u 360 P 7 b u 164 N 8 b y 180 N - = = | | ) , ( log * | | ) , ( ) ( info 2 1 T T C freq T T C freq T j j k j ) info( * | | | | ) ( info 1 X i i n i T T T T = =
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Name:______________________________ page 2/8 2. Using the data shown in problem #1, develop datasets which can be used to do 3-fold cross- validation. [Careful] Clearly explain. [10 pts] Split the data up into three non-overlapping data sets, which encompass the full set of data. Randomly selecting the 3 sets would be an ideal way. The model will be trained in 3 different passes (or folds). In each pass, all data sets except for one will be used to train. The one not used for training will be used for testing.
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 8

CS404_Exam_2_110402 - CSc 401 Data Mining Exam #2 November...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online