Current location - Education and Training Encyclopedia - Graduation thesis - Why should the relative logarithm index variable take logarithm in the paper?
Why should the relative logarithm index variable take logarithm in the paper?
Why the relative logarithm index variable should be logarithmic in this paper: In some data processing, the original data is often logarithmic, and then further processing. This is done because the logarithmic function is monotonically increasing in its definition domain, and the relative relationship of data will not be changed after taking the logarithm.

Reduce the absolute value of data for calculation. For example, the value of each data item is very large, and the calculation of many such values may exceed the range of common data types. At this time, taking logarithm will reduce the value. For example, when calculating TF-IDF, the frequency of many words in large-scale corpus is very large.

adaptability

In the case of product form, quotient form, radical form, power form, exponential form or power exponential function form, logarithmic derivative method is more suitable for derivative, because logarithm can turn multiplication or division into addition or subtraction, and logarithm can turn radical form, power function, exponential function and power exponential function into multiplication and division.