Concept Learning $[20]$

The data in Table $1$ show the content rating $($S$,$N$,$V$,$L$)$ of $6$ different movies, labeled

m$1$ to m$6,$ and whether Jane liked the movie, or not.

movie S N L V likes

m$1$ T F F T T

m$2$ T F F F T

m$3$ F T F F F

m$4$ F F T F F

m$5$ T T F F T

m$6$ T T T F F

Table $1$: Jane$$s movie preferences

We want to use the data in Table $1$ to learn which kind of movies, based on their

content rating, Jane likes to watch. A hypothesis for this data set can be written as

a conjunction of literals, for example:

h$1=$ S $$N $$V

$($a$)$ Use the Candidate$-$Elimination algorithm to find the G$-$ and S$-$boundaries of

the data in the table. Show all your steps. $(15)$

$($b$)$ What do you observe about the resulting version space that you computed

above? $(2)$

$($c$)$ Order the following hypotheses from least specific to most specific: $(3)$

i$.$ h$1=$ V

ii$.$ h$2=$ N $$V

iii. h$3=$N