Statistical Measures as Predictors of Compression Savings
William Culhane · The Knowledge Bank (The Ohio State University) · 2008
Different compression algorithms run on the same file will result in differently sized compressed files. Choosing the algorithm which will give the smallest resulting file has some advantages. It would be useful to be able to do this without having to take the resources to run every algorithm in question. In order to do this, some predictive measure has to be found.