When we cluster and classify we can do so in various ways across industry sector, customer size , revenue generated , length of tenure etc.
Which latent set of variables help point towards root cause of customer behavior ?
Different Entities may be named differently but might actually be the same Entity.
Identities are not unique. How can we tell if they are probably associated with the same Entity,
Each entity name has a unique identifier.
Market Basket Analysis using association rule learning is a candidate algorithm for performing account resolution analysis.
You have probabilistic indication of convergence on entity from possible multiple entities