The pearl algorithm as a method to extract information out of a database
JW Kruger · Unisa Institutional Repository (University of South Africa) · 1996
Big databases have been built up over time. The information in these databases can be converted by the Pearl algorithm [ Pearl 1988 ] from data to information. When a promotional drive is initiated, do we know who the potential customers are? The Pearl algorithm gives the belief that an individual will purchase. This means that the database can be used to select the individuals that are most likely to purchase. The break-even point, where the cost of the promotional contact and the expected return are the same, can be calculated. The Pearl algorithm uses Bayes' probabilities to propagate belief through a tree.