Fusing Data with Correlations
by Ravali Pochampally, Anish Das Sarma, Xin Luna Dong, Alexandra Meliou, Divesh Srivastava
Abstract:
Many applications rely on Web data and extraction systems to accomplish knowledge-driven tasks. Web information is not curated, so many sources provide inaccurate, or conflicting information. Moreover, extraction systems introduce additional noise to the data. We wish to automatically distinguish correct data and erroneous data for creating a cleaner set of integrated data. Previous work has shown that a naïve voting strategy that trusts data provided by the majority or at least a certain number of sources may not work well in the presence of copying between the sources. However, correlation between sources can be much broader than copying: sources may provide data from complementary domains (negative correlation), extractors may focus on different types of information (negative correlation), and extractors may apply common rules in extraction (positive correlation, without copying). In this paper we present novel techniques modeling correlations between sources and applying it in truth finding. We provide a comprehensive evaluation of our approach on three real-world datasets with different characteristics, as well as on synthetic data, showing that our algorithms outperform the existing state-of-the-art techniques.
Citation:
Ravali Pochampally, Anish Das Sarma, Xin Luna Dong, Alexandra Meliou, and Divesh Srivastava, Fusing Data with Correlations, in Proceedings of the ACM SIGMOD International Conference on Management of Data (SIGMOD), 2014, pp. 433–444.
Bibtex:
@inproceedings{DBLP:conf/sigmod/PochampallyDDMS14,
    Abstract = {Many applications rely on Web data and extraction systems to accomplish
    knowledge-driven tasks. Web information is not curated,
    so many sources provide inaccurate, or conflicting information.
    Moreover, extraction systems introduce additional noise to the data.
    We wish to automatically distinguish correct data and erroneous
    data for creating a cleaner set of integrated data. Previous work
    has shown that a na\"ive voting strategy that trusts data provided by
    the majority or at least a certain number of sources may not work
    well in the presence of copying between the sources. However,
    correlation between sources can be much broader than copying:
    sources may provide data from complementary domains (negative
    correlation), extractors may focus on different types of information
    (negative correlation), and extractors may apply common rules in
    extraction (positive correlation, without copying). In this paper we
    present novel techniques modeling correlations between sources
    and applying it in truth finding. We provide a comprehensive
    evaluation of our approach on three real-world datasets with different
    characteristics, as well as on synthetic data, showing that our
    algorithms outperform the existing state-of-the-art techniques.},
    Author = {Pochampally, Ravali and Das Sarma, Anish and Dong, Xin Luna and Meliou, Alexandra and Srivastava, Divesh},
    Booktitle = {Proceedings of the ACM SIGMOD International Conference on Management of Data (SIGMOD)},
    Title = {\href{http://people.cs.umass.edu/ameli/projects/dataIntegration/papers/corrFusion-SIGMOD2014.pdf}{Fusing Data with Correlations}},
    Venue = {SIGMOD},
    address = {Snowbird, Utah},
    month = {June},
    Year = {2014},
    pages = {433--444},
    doi = {10.1145/2588555.2593674},
}