• Login
    View Item 
    •   Shocker Open Access Repository Home
    • Graduate Student Research
    • GRASP: Graduate Research and Scholarly Projects
    • Proceedings 2016: 12th Annual Symposium on Graduate Research and Scholarly Projects
    • View Item
    •   Shocker Open Access Repository Home
    • Graduate Student Research
    • GRASP: Graduate Research and Scholarly Projects
    • Proceedings 2016: 12th Annual Symposium on Graduate Research and Scholarly Projects
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Maximum inner product search using memory efficient randomized partition trees

    View/Open
    Abstract (511.5Kb)
    Date
    2016-04-29
    Author
    Keivani, Omid
    Advisor
    Sinha, Kaushik
    Metadata
    Show full item record
    Citation
    Keivani, Omid. 2016. Maximum inner product search using memory efficient randomized partition trees. --In Proceedings: 12th Annual Symposium on Graduate Research and Scholarly Projects. Wichita, KS: Wichita State University, p. 63
    Abstract
    Finding the maximum inner product is a well-known problem and it has been used in many applications such as recommender systems. It is also known that this problem can be converted to a nearest neighbor problem using various transformations. Local sensitivity hashing (LSH) is the most used algorithm for this purpose, but it has some disadvantages (e.g. many parameters should be fixed initially and there is no optimal way to fix them). In this paper we use an existing method called Random projection tree (RPT) and also propose two space efficient versions of it at nearly no additional cost. Moreover, we prove what the best transformation for RPT is. Finally, we test our method on many real world datasets such as Movielens and Netflix. The results connote that with the same amount of computations (even less), RPT have higher accuracy than LSH.
    Description
    Presented to the 12th Annual Symposium on Graduate Research and Scholarly Projects (GRASP) held at the Heskett Center, Wichita State University, April 29, 2016.

    Research completed at Department of Electricial Engineering and Computer Science, College of Engineering
    URI
    http://hdl.handle.net/10057/12210
    Collections
    • EECS Graduate Student Conference Papers
    • Proceedings 2016: 12th Annual Symposium on Graduate Research and Scholarly Projects

    Browse

    All of Shocker Open Access RepositoryCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsBy TypeThis CollectionBy Issue DateAuthorsTitlesSubjectsBy Type

    My Account

    LoginRegister

    Statistics

    Most Popular ItemsStatistics by CountryMost Popular Authors

    DSpace software copyright © 2002-2023  DuraSpace
    DSpace Express is a service operated by 
    Atmire NV