Thank you Dr. Manashty for making things simple and straight forward and yet deeply insightful. My question is how do you determine the thresholds for min_support and min_Confidence as a young Data Scientist? Thank you.
@Alberto: Thank you for watching the video. I think it depends on the size of the dataset and number of items you have. If you have a diverse set of items and no popular items, the chances are that your support numbers are quite low compared to your total number of transactions. The more variety you have, the lower your min_support and min_confidence should be. It is best to try a few different thresholds and see how many rules you get. If you get too many or too few, try increasing or decreasing your thresholds.
Thank you Dr. Manashty for making things simple and straight forward and yet deeply insightful. My question is how do you determine the thresholds for min_support and min_Confidence as a young Data Scientist? Thank you.
@Alberto: Thank you for watching the video. I think it depends on the size of the dataset and number of items you have. If you have a diverse set of items and no popular items, the chances are that your support numbers are quite low compared to your total number of transactions. The more variety you have, the lower your min_support and min_confidence should be. It is best to try a few different thresholds and see how many rules you get. If you get too many or too few, try increasing or decreasing your thresholds.