| Title: |
A Collective, Probabilistic Approach to Schema Mapping Using Diverse Noisy Evidence |
| Authors: |
Kimmig, Angelika; Memory, Alex; Miller, Renee J.; Getoor, Lise |
| Source: |
ISSN:1041-4347 ; ISSN:1558-2191 ; IEEE Transactions on Knowledge and Data Engineering, vol. 31 (8), Art.No. 8, (1426-1439. |
| Publisher Information: |
Institute of Electrical and Electronics Engineers |
| Publication Year: |
2019 |
| Subject Terms: |
Science & Technology; Technology; Computer Science; Artificial Intelligence; Information Systems; Engineering; Electrical & Electronic; Schema mapping; data integration; probabilistic logic; optimization; REFINEMENT; DESIGN; 08 Information and Computing Sciences; 46 Information and computing sciences |
| Description: |
IEEE We propose a probabilistic approach to the problem of schema mapping. Our approach is declarative, scalable, and extensible. It builds upon recent results in both schema mapping and probabilistic reasoning and contributes novel techniques in both fields. We introduce the problem of schema mapping selection, that is, choosing the best mapping from a space of potential mappings, given both metadata constraints and a data example. As selection has to reason holistically about the inputs and the dependencies between the chosen mappings, we define a new schema mapping optimization problem which captures interactions between mappings as well as inconsistencies and incompleteness in the input. We then introduce Collective Mapping Discovery (CMD), our solution to this problem using state-of-the-art probabilistic reasoning techniques. Our evaluation on a wide range of integration scenarios, including several real-world domains, demonstrates that CMD effectively combines data and metadata information to infer highly accurate mappings even with significant levels of noise. ; sponsorship: AK has been supported by Research Foundation Flanders (FWO). RJM is supported by NSERC. LG was supported by the National Science Foundation under Grant Numbers CCF-1740850 and IIS-1703331, AFRL and DARPA. The authors thank Boris Glavic, Patricia Arocena, GianniMecca, Donatello Santoro, and Radu Ciucanu for their valuable help with iBench and ++ Spicy; and Craig Knoblock for the Neuroscience problem set. (Research Foundation Flanders (FWO), NSERC, National Science Foundation|CCF-1740850, National Science Foundation|IIS-1703331, AFRL, DARPA) ; status: Published |
| Document Type: |
article in journal/newspaper |
| File Description: |
application/pdf |
| Language: |
English |
| Relation: |
https://lirias.kuleuven.be/handle/123456789/662369; https://doi.org/10.1109/TKDE.2018.2865785 |
| DOI: |
10.1109/TKDE.2018.2865785 |
| Availability: |
https://lirias.kuleuven.be/handle/123456789/662369; https://lirias.kuleuven.be/retrieve/0c393905-2081-4112-ac7d-cdff56ce1b03; https://doi.org/10.1109/TKDE.2018.2865785 |
| Rights: |
info:eu-repo/semantics/openAccess ; public |
| Accession Number: |
edsbas.6D62383C |
| Database: |
BASE |