Question $5$:

Books$-$plus is an NGO working towards providing quality education for children belonging to social and economically backward communities. Currently, they are in the midst of preparing a curriculum for imparting basic communication and critical thinking skills. Based on expert inputs, they have come up with a Basic module and an Advanced module. They are planning to launch these modules in the districts across North India. To maximize effective absorption of the modules, they are planning to launch one module $($either Advanced or Basic$)$ in each of the districts during phase $1$ depending on the socio$-$economic conditions and the current skill level of the district. To assess the same, they conduct a pilot in $14$ districts and obtain the following result.

$\backslash$table$[[$District ID$,$Income Level,Literacy,$\backslash$table$[[$Social$],[$Inclusion$]],$District Type,Module$],[1,$High $\frac{?}{{?}_{?}}.,$Low ${?}_v,$Poor $v,$Rural,Basic.$],[2,$High Low $-,$Good dots,Rural $\frac{?}{?},$Basic $.],[3,$High Low vlongrightarrow,Poor$-$r$,$Semi$-$Urban $-,$Advanced$],[4,$Medium,How $vv,$Poor Urban $=-,$Advanced$],[5,$Low,High,Poor,Urban,,Advanced$],[6,$Low $.,$High,Good,Urban $-,$Basic $.],[7,$Low $,,$High,Good,Semi$-$Urb$$n $-,$Advanced$],[8,$Medium $,$Low $\sqrt[\phantom{2}]{?},$Poor $\frac{?}{?},$Rural :$-,$Basic$],[9,-,$High,Poor?,Rural, $,,$Advanced$],[10,$Medium $/,$High,Poor $,$Urban.,Advanced$],[11,$Medium $/,$High,Good,Rural $/,$Advanced$],[12,$Medium Low Good,Semi$-$Urban $,,$Advanced$],[13,$High $,,$High,Poor.a$,$Semi$-$rban $,,$Advanced$],[14,$Medium $,$Low $J,$Food,Urban $-,$Basic :$]]$

Based on the data above, construct a decision tree using any method of your choice. $[9$ marks$]$

ii What module would you suggest for a Semi$-$Urban district with Poor Social Inclusion and High Income Level? $[2$ marks$]$

iii. Let's say you run the decision tree model that you have built for a testing dataset of $200$ data points and obtain the following result.

$\backslash$table$[[,$Observed Beginner,Observed Advanced$],[$Predicted Beginner,$45,35],[$Predicted Advanced,$75,45]]$

Calculate the misclassification rate? $[1$ mark$]$