题目到底问的是哪一个动作

问题如下：

Omar Khan is investigating the potential benefits of incorporating non-financial data, specifically weather and social media posts, to improve their stock selection process in the retail industry.

Dataset 1: A database from a large national weather provider that contains detailed weather data (temperature, humidity, rainfall, atmospheric pressure, etc.) at a very localized geographic level or zone recorded by GPS coordinates for the past 36 months.

In reviewing Dataset 1, Khan notices that there are many data fields included that would likely be highly irrelevant to their analysis and begins the process of selecting a subset of data fields that he believes are applicable.

Khan identifies a data field called that contains such results as drizzle, rain, freezing rain, snow, and hail. Khan believes that this could be much more valuable as a prediction variable if it is converted into binary form (1 = yes, 0 = no).

Khan’s selection of a subset of data from the weather dataset is best described as:

选项：

trimming

feature selection

feature engineering

解释：

B is correct. The process of identifying and removing unneeded, irrelevant, or redundant features in a dataset is known as feature selection.

A is incorrect. Trimming is a process for handling outliers in a dataset by simply removing the extreme values and is also known as truncation.

C is incorrect. Feature engineering is the process of combining, consolidating, or creating new features that do not exist in the current weather dataset.