When using Data Refinery, what is the process of customizing data by filtering, sorting, or removing columns?

The process of customizing data by filtering, sorting, or removing columns is referred to as "Shaping" the data. This encompasses the various tasks performed to organize the data appropriately for analysis, ensuring that it is in a suitable form for the specific insights or outcomes desired. When shaping data, users can modify the dataset to enhance its quality, reduce noise, and create a more focused dataset by selecting only the relevant columns. This manipulation is crucial in preparing the data for subsequent analytical processes.

The other terms do not capture this specific function. Training pertains to the process of teaching a model to recognize patterns based on input data. Cleansing involves correcting or removing inaccuracies and inconsistencies in the data, which might overlap with shaping but is more focused on ensuring data integrity rather than adjusting its structural layout for analysis. Prediction relates to the act of making forecasts based on trained models and is not involved in the preprocessing or structuring of raw data. Therefore, shaping is the correct term for the customization processes mentioned in the question.