Association Rules
See Frequent Itemset Mining. Once a set of frequent itemsets has been found, association rules can be generated. Association rules are of the form A→B, and could be read as “A implies B”. Each association rule has support (how common the precondition is in the dataset), confidence (how often the precondition leads to the consequence in the dataset) and lift (how much more common the consequence is in instances covered by the rule compared to the whole dataset). ref
Applications
Identify promising secondary phenotype candidates. Predictions rely on a large gene–phenotype annotation set that is used to find occurrence patterns of phenotypes. ref.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.