Computing, IT & AI

Association rule mining (Apriori)

1993-1994 CERakesh Agrawal and Ramakrishnan Srikant (IBM Almaden)Fast algorithms for mining association rules (VLDB 1994)Diaspora (work abroad)

1993

An algorithm that finds items that often occur together in large databases, such as products bought together.

Dated: 1993–1994
As scholars put it: 1993-1994 CE
Originator: Rakesh Agrawal and Ramakrishnan Srikant (IBM Almaden)
Source text: Fast algorithms for mining association rules (VLDB 1994)
Region: Diaspora (work abroad)
Discipline: Computing, IT & AI
Evidence: Documented
Confidence: High
Verification: Verified-web

Background

Rakesh Agrawal (with Tomasz Imielinski and Arun Swami) defined association rule mining in 1993. In 1994 Agrawal and Ramakrishnan Srikant published Apriori, which prunes the search using the fact that every subset of a frequent itemset must also be frequent.

Significance

Among the founding papers of data mining, and among the most cited papers in computer science. Agrawal also pioneered privacy-preserving data mining.

Path to modern innovation

Market-basket analysis, recommendations and fraud detection.

Global impact

"Customers who bought this also bought" recommendations.

References

3Fast algorithms for mining association rules R. Agrawal and R. Srikant, 1994