Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
DEEP LEARNING-BASED METHOD AND APPARATUS FOR SKYLINE QUERY CARDINALITY ESTIMATION
Document Type and Number:
WIPO Patent Application WO/2024/021624
Kind Code:
A1
Abstract:
Disclosed in the present invention are a deep learning-based method and apparatus for Skyline query cardinality estimation. The method comprises: analyzing historical query log information of a database, acquiring a Skyline query on a given target data set and its corresponding cardinality, to construct a training set; according to distribution information of the target data set and the training set, constructing and training data distribution learning models, respectively; using a model parameter of a trained data distribution learning model as an initialization parameter of a cardinality estimation model, and training the cardinality estimation model according to the training set; and according to the trained cardinality estimation model, inputting a query point, to obtain a final cardinality estimation value. The present invention provides a solution for cardinality estimation of a Skyline query variant, and ensures a monotonic property present in the cardinality estimation of the Skyline query variant; an efficient and accurate cardinality estimation method is provided, and the method has the advantages of high accuracy, high efficiency, strong robustness, and strong expandibility. The method has a wide range of application scenarios in fields such as modern database management systems and query optimization.

Inventors:
MIAO XIAOYE (CN)
PENG JIAZHEN (CN)
WU YANGYANG (CN)
YIN JIANWEI (CN)
Application Number:
PCT/CN2023/080962
Publication Date:
February 01, 2024
Filing Date:
March 13, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV ZHEJIANG (CN)
International Classes:
G06F16/2457; G06N20/00; G06N3/04
Foreign References:
CN115392477A2022-11-25
CN114398395A2022-04-26
Other References:
YANG, ZONGHENG ET AL.: "Deep Unsupervised Cardinality Estimation", PROCEEDINGS OF THE VLDB ENDOWMENT, vol. 13, no. 3, 1 November 2019 (2019-11-01), pages 279 - 292, XP058445285, ISSN: 2150-8097, DOI: 10.14778/3368289.3368294
ZHANG, ZHENJIE ET AL.: "Kernel-Based Skyline Cardinality Estimation", PROCEEDINGS OF THE 2009 ACM SIGMOD INTERNATIONAL CONFERENCE ON MANAGEMENT OF DATA, 29 June 2009 (2009-06-29), pages 509 - 521
Attorney, Agent or Firm:
HANGZHOU JUNRUIDA INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: