Data Compression – Weissman Score


producers of HBO’s television series
Silicon Valley approached Tsachy Weissman, a professor at Stanford and Vinith Mishra a graduate student for developing an efficiency metric for
lossless compression professor Tsachy, developed an amazing equation that
compared both time required to compress a file and its compression ratio he
termed an equation as Weissman score after his own name the Weissman score W
is formulated as alpha times the ratio of compression ratio of a particular
file and the compression ratio of a universal standard compression ratio for
the file times the ratio of log of time taken by the standard universal
compression algorithm to compress a particular file and that of our
compression algorithm where alpha is a scaling constant by normalizing by the
performance of a standard compressor we take away variation in compressive
performance between types of data which means that Weissman score remains the
same whether your algorithm is compressing a text or an image or an
audio file according to Wikipedia Weissman score is only for fictional use it
cannot be used in real-world compression algorithms as the units used to measure
the times changes the score meaning if one person measures time in
seconds and other one in minutes let’s say 120 seconds and two minutes they
both refers to same amount of time but Weissman score used log function of time
in that sense log of 120 is not equal to log of two also the time taken by the
algorithm should not be equal to 1 as log of 1 is 0 which will yield the Weissman score
0 if you have any question ask it in the comments I’ll try to answer if it is
really important also leave a like and you found your answer and don’t forget
to subscribe I am Tanseer Saji and you are watching
data compression 101

, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

Post navigation

Leave a Reply

Your email address will not be published. Required fields are marked *