|
The purpose of this thesis is to make a comprehensive survey about text compression and find out an optimal algorithm for it. There have been some famous algorithms in this realm. How these algorithms work and why they compress data well are two topics we most concern about, so we survey these algorithms and analyze the superiority and limitation of them from the theoretical viewpoint at first. Then, the performance of algorithms are evaluated by experiments. Finally, the context modeling is suggested to improve the compression ratio further. By that, almost all the redundancy in source messages is exhausted under our understanding. A new algorithm using the context modeling is proposed and evaluated.
|