$($d$)$ Two techniques of implementing recommendation systems are collaborative filtering and content$-$based filtering. $[7]$

$($i$)$ A user$-$item matrix is shown below. The values are the user ratings on the items, ranging from $1$ for the lowest to $5$ for the highest. $$NA$$ indicates an item that has not been consumed by a user.

Item $0$ Item $1$ Item $2$ Item $3$ Item $4$ Item $5$ Item $6$ Item $7$

User $01455$ NA NA NA $3$

User $151$ NA $151$ NA NA

User $24$ NA $22415$ NA

User $3254$ NA $241$ NA

The cosine similarity values of the users have been calculated and given below.

User $0$ User $1$ User $2$ User $3$

User $01\mathrm{.}000\mathrm{.}720\mathrm{.}820\mathrm{.}93$

User $10\mathrm{.}721\mathrm{.}000\mathrm{.}920\mathrm{.}67$

User $20\mathrm{.}820\mathrm{.}921\mathrm{.}000\mathrm{.}75$

User $30\mathrm{.}930\mathrm{.}670\mathrm{.}751\mathrm{.}00$

Applying collaborative filtering user similarity and using these data, recommend an item to user $0.$ Describe the steps and how the data are used in the process.

$($ii$)$ Give two advantages of collaborative filtering compared to content$-$based filtering.