Approximating Certain and Possible Answers Beyond Sets
summary
Generalizes certain and possible answers from set semantics to semiring-annotated relations, proving that lattice-ordered semirings let query evaluation produce PTIME under-approximations of certain answers and over-approximations of possible answers; extends the model with attribute-level intervals into UA-DBs that remain closed under full relational algebra including aggregation and sort-based ops; provides two implementations (including GProM) and demonstrates feasible execution on TPCH 1GB with measurable but reducible overhead.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 1:11Incomplete databases
- 2:35Desired properties
- 4:01Semirings for databases
- 6:33K-relations semantics
- 7:15Defining certain and possible annotations
- 8:59Certain vs possible answers
- 12:10Mapping uncertainty models
- 12:52UA-DBs overview
- 15:00Bound-preserving semantics
- 17:07Handling difference operator
- 18:10Datalog for UA-DBs
- 19:14Existing works trade-offs
- 20:22Aggregations challenges
- 22:30Interval domains
- 25:01Attribute uncertainty bounds
- 26:06Semiring relations algebra
- 27:31Aggregations with intervals
- 28:59Group by sorting uncertain
- 29:42Conclusion future work
- 31:06Practical demo
- 32:11Q&A
related talks
01h 06mAlgebraic Structures in Logic and Query Evaluation 1
01h 05mAlgebraic Structures in Logic and Query Evaluation 2
01h 02mAlgebraic Structures in Logic and Query Evaluation 3
01h 08mAlgebraic Structures in Logic and Query Evaluation 4
43mA Dichotomy for Queries with Negation in Probabilistic Databases
01h 01mBrief Tutorial on Probabilistic Databases
01h 09mAccessing Answers to Unions of Conjunctive Queries with Ideal Time Guarantees: I
38mThe Intensional-Extensional Problem in Probabilistic Databases
