Before the RME can cluster any data, it must first identify the key words in each document. It is important that the same number of key words are extracted from each documents, but identifying the optimum number of key words is a complicated process. The number of keywords chosen is also known as the dimensionality of the data. If the dimensionality is too low then some of the most important keywords will be missed, but if it is too high, then some unimportant words might be incorrectly used.
The RME uses complex statistical modeling techniques to determine the optimum dimensionality, but for large datasets this process can take some time. By using the precision slider you can alter how long the RME will take to find the optimum dimensionality. By moving the precison slider all the way to the right you will be guaranteed to find the optimum dimensionality, but the process could take a while. If you move it to the left the process will go very quickly, but the result might not be as accurate.