Data that either produce low layer info or simply restate that which is already known is called as data redundancy.

For e.g. Using lots of data to convey little info.

Let $N_{1}$ and $N_{2}$ be no. of information carrying unit into data set that represents same info Then

Where,

R $\rightarrow$ negative data redundancy of first data set. CR $\rightarrow$ Compression ratio $N_{1} / N_{2}$

1. If $N_{1} = N_{2}$

   CR = 1

   It indicates that first representation of information i.e. $N_{1}$ contains no redundant data.

2. If $N_{1}$ >> $N_{2}$

   CR $\rightarrow$∞

   Implies significant compression and highly redundant data.

3. If $N_{1}$ << $N_{2}$

   CR $\rightarrow$0

   It indicates that second data set contains much more data than original representation.

   (Undesirable case of data expansion)