Streamlining Context Models for Data Compression

Context modeling has emerged as the most promising new approach to com-pressing text. While context-modeling algorithms provide very good compression, they suer from the disadvantages of being slow and requiring large amounts of main memory in which to execute. We describe a context-model-based algorithm that runs signicantly faster, uses much less space, and provides compression ratios close to those of earlier context modeling algorithms. We achieve these improvements through the use of self-organizing lists