Mining surprising patterns using temporal description length

Summary: Introduces a parameter-free MDL measure of temporal surprise for itemsets, rewarding compressible steady correlations and discounting patterns explained by known subsets. Mining thus rejects obvious frequent complements and surfaces itemsets with evolving correlations over time. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID: 8713
Venue: VLDB
Year: 1998
Pagerank: 6.9709253e-05
Overall Rank: 3,990 | 72.63%
DOI: -

Incoming Non-self Citations Over Time

Authors

1. Soumen Chakrabarti (IBM)
2. Sunita Sarawagi (IBM)
3. Byron Dom (IBM)

BibTeX Citation

@article{chakrabarti_vldb98,
        title = {{Mining surprising patterns using temporal description length}},
        author = {Chakrabarti, Soumen and Sarawagi, Sunita and Dom, Byron},
        journal = {PVLDB},
        series = {{VLDB} '98},
        pages = {606--617},
        year = {1998}
}

Incoming Citations (Sorted by Pagerank)

Showing 4 of 4 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
1,029	Anomaly Detection in Time Series: A Comprehensive Evaluation	2022	VLDB	0.00012557065
3,785	Detecting Change in Data Streams	2004	VLDB	7.1265312e-05
6,305	A Regression-Based Temporal Pattern Mining Scheme for Data Streams	2003	VLDB	5.9204295e-05
7,210	A Framework for Measuring Changes in Data Characteristics	1999	PODS	5.6720634e-05

Outgoing Citations (Sorted by Pagerank)

Showing 2 of 2 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
711	Beyond Market Baskets: Generalizing Association Rules to Correlations	1997	SIGMOD	0.00014713746
1,318	Querying Shapes of Histories	1995	VLDB	0.00011178588

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	10,069	Mining and Forecasting of Big Time-series Data	2015	SIGMOD
2	9,212	Feasible Itemset Distributions in Data Mining: Theory and Application	2003	PODS
3	5,179	An Efficient Rigorous Approach for Identifying Statistically Significant Frequent Itemsets	2009	PODS
4	7,691	Finding Relevant Patterns in Bursty Sequences	2008	VLDB
5	5,043	Efficient Temporal Pattern Mining in Big Time Series Using Mutual Information	2022	VLDB
6	12,150	Beyond Itemsets: Mining Frequent Featuresets over Structured Items	2015	VLDB
7	11,249	Efficient Discovery of Significant Patterns with Few-Shot Resampling	2024	VLDB
8	704	Dynamic Itemset Counting and Implication Rules for Market Basket Data	1997	SIGMOD
9	711	Beyond Market Baskets: Generalizing Association Rules to Correlations	1997	SIGMOD
10	12,349	Mining Flipping Correlations from Large Datasets with Taxonomies	2012	VLDB