837
views
0
recommends
+1 Recommend
1 collections
    0
    shares

      Teaching business & IT courses? Request a digital desk copy of any BCS Publishing title via the BCS Bookshop Inspection Copy Request form.

      scite_
       
      • Record: found
      • Abstract: found
      • Conference Proceedings: found
      Is Open Access

      Similarity Learning in Nearest Neighbor and Application to Information Retrieval

      Published
      proceedings-article
      ,
      Third BCS-IRSG Symposium on Future Directions in Information Access (FDIA 2009) (FDIA)
      Computers XXIII Celebrating People and Technology
      1 September 2009
      Similarity learning, kNN, Information Retrieval, Machine Learning
      Bookmark

            Abstract

            Many people have tried to learn Mahanalobis distance metric in kNN classification by considering the geometry of the space containing examples. However, similarity may have an edge specially while dealing with text e.g. Information Retrieval.We have proposed an online algorithm, SiLA (Similarity learning algorithm) where the aim is to learn a similarity metric (e.g. cosine measure, Dice and Jaccard coefficients) and its variation eSiLA where we project the matrix learnt onto the cone of positive, semidefinite matrices. Two incremental algorithms have been developed; one based on standard kNN rule while the other one is its symmetric version. SiLA can be used in Information Retrievalwhere the performance can be improved by using user feedback.

            Content

            Author and article information

            Contributors
            Conference
            September 2009
            September 2009
            : 131-133
            Affiliations
            [0001]LIG (Laboratoire d’Informatique de Grenoble)

            Université Joseph Fourier
            Article
            10.14236/ewic/FDIA2009.22
            82bec467-0a0e-42e3-8dd4-309165e73b99
            © Ali Mustafa Qamar. Published by BCS Learning and Development Ltd. Third BCS-IRSG Symposium on Future Directions in Information Access (FDIA 2009), Padua, Italy

            This work is licensed under a Creative Commons Attribution 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

            Third BCS-IRSG Symposium on Future Directions in Information Access (FDIA 2009)
            FDIA
            3
            Padua, Italy
            1 September 2009
            Electronic Workshops in Computing (eWiC)
            Computers XXIII Celebrating People and Technology
            History
            Product

            1477-9358 BCS Learning & Development

            Self URI (article page): https://www.scienceopen.com/hosted-document?doi=10.14236/ewic/FDIA2009.22
            Self URI (journal page): https://ewic.bcs.org/
            Categories
            Electronic Workshops in Computing

            Applied computer science,Computer science,Security & Cryptology,Graphics & Multimedia design,General computer science,Human-computer-interaction
            Similarity learning,kNN,Information Retrieval,Machine Learning

            Comments

            Comment on this article