Please Answer with detailed solutionConsider the following documents in standard frequency$-$based vector space. The dissimilarity

metric is the Manhattan distance; $m(D_k,D_j)={\displaystyle \underset{i}{\overset{?}{}}}|D_{k,i}-D_{j,i}|,$ where $D_k$ and $D_j$ represent

the document$-$vectors and $D_{k,i}$ gives the frequency of term $i$ in document $D_k.$

$D_1$: dil, chahta, hai, dil, chahta, hai, hai, hai, hai

$D_2$: dilwale, chahta, chahta, bade

$D_3$: dil, bade, bade, dil, dil

$D_4$: dilwale, dilwale, dilwale

$4$a$.$ Construct the term$-$document matrix and then document$-$document matrix under the as$-$

sumption that the terms are not stemmed.

$4$b$.$ On the basis of the document$-$document matrix, perform complete$-$link clustering, showing

the output as well as intermediate results.

Ac$.$ Describe each step of single$-$link clustering.

$4$d$.$ Will there be any change if you consider stemming? Justify.