A Case for Hubness Removal in High–Dimensional Multimedia Retrieval View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2014

AUTHORS

Dominik Schnitzer , Arthur Flexer , Nenad Tomašev

ABSTRACT

This work investigates the negative effects of hubness on multimedia retrieval systems. Because of a problem of measuring distances in high-dimensional spaces, hub objects are close to an exceptionally large part of the data while anti-hubs are far away from all other data points. In the case of similarity based retrieval, hub objects are retrieved over and over again while anti-hubs are nonexistent in the retrieval lists. We investigate textual, image and music data and show how re-scaling methods can avoid the problem and decisively improve the overall retrieval quality. The observations of this work suggest to make hubness analysis an integral part when building a retrieval system. More... »

PAGES

687-692

References to SciGraph publications

  • 2004-11. Distinctive Image Features from Scale-Invariant Keypoints in INTERNATIONAL JOURNAL OF COMPUTER VISION
  • 2010. Can Shared-Neighbor Distances Defeat the Curse of Dimensionality? in SCIENTIFIC AND STATISTICAL DATABASE MANAGEMENT
  • 2011. The Role of Hubness in Clustering High-Dimensional Data in ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING
  • Book

    TITLE

    Advances in Information Retrieval

    ISBN

    978-3-319-06027-9
    978-3-319-06028-6

    From Grant

    Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/978-3-319-06028-6_77

    DOI

    http://dx.doi.org/10.1007/978-3-319-06028-6_77

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1036963898


    Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
    Incoming Citations Browse incoming citations for this publication using opencitations.net

    JSON-LD is the canonical representation for SciGraph data.

    TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

    [
      {
        "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
        "about": [
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Artificial Intelligence and Image Processing", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Information and Computing Sciences", 
            "type": "DefinedTerm"
          }
        ], 
        "author": [
          {
            "affiliation": {
              "alternateName": "Austrian Research Institute for Artificial Intelligence", 
              "id": "https://www.grid.ac/institutes/grid.432019.d", 
              "name": [
                "Austrian Research Institute for Artificial Intelligence (OFAI), Vienna, Austria"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Schnitzer", 
            "givenName": "Dominik", 
            "id": "sg:person.014564424411.74", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014564424411.74"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Austrian Research Institute for Artificial Intelligence", 
              "id": "https://www.grid.ac/institutes/grid.432019.d", 
              "name": [
                "Austrian Research Institute for Artificial Intelligence (OFAI), Vienna, Austria"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Flexer", 
            "givenName": "Arthur", 
            "id": "sg:person.0757053000.20", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0757053000.20"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Jo\u017eef Stefan Institute", 
              "id": "https://www.grid.ac/institutes/grid.11375.31", 
              "name": [
                "Artificial Intelligence Laboratory, Jo\u017eef Stefan Institute, Ljubljana, Slovenia"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Toma\u0161ev", 
            "givenName": "Nenad", 
            "id": "sg:person.014252560763.36", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014252560763.36"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "https://doi.org/10.1016/j.cviu.2005.09.012", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1004784969"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1145/1639714.1639771", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1006240814"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-642-13818-8_34", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1024575487", 
              "https://doi.org/10.1007/978-3-642-13818-8_34"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-642-13818-8_34", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1024575487", 
              "https://doi.org/10.1007/978-3-642-13818-8_34"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-642-20841-6_16", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1033916131", 
              "https://doi.org/10.1007/978-3-642-20841-6_16"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1002/sam.11161", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1041615658"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/b:visi.0000029664.99615.94", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1052687286", 
              "https://doi.org/10.1023/b:visi.0000029664.99615.94"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/tsa.2002.800560", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1061786085"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/iccp.2011.6047899", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094037868"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/iccv.2005.77", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1094132829"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.5244/c.23.2", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1099325633"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2014", 
        "datePublishedReg": "2014-01-01", 
        "description": "This work investigates the negative effects of hubness on multimedia retrieval systems. Because of a problem of measuring distances in high-dimensional spaces, hub objects are close to an exceptionally large part of the data while anti-hubs are far away from all other data points. In the case of similarity based retrieval, hub objects are retrieved over and over again while anti-hubs are nonexistent in the retrieval lists. We investigate textual, image and music data and show how re-scaling methods can avoid the problem and decisively improve the overall retrieval quality. The observations of this work suggest to make hubness analysis an integral part when building a retrieval system.", 
        "editor": [
          {
            "familyName": "de Rijke", 
            "givenName": "Maarten", 
            "type": "Person"
          }, 
          {
            "familyName": "Kenter", 
            "givenName": "Tom", 
            "type": "Person"
          }, 
          {
            "familyName": "de Vries", 
            "givenName": "Arjen P.", 
            "type": "Person"
          }, 
          {
            "familyName": "Zhai", 
            "givenName": "ChengXiang", 
            "type": "Person"
          }, 
          {
            "familyName": "de Jong", 
            "givenName": "Franciska", 
            "type": "Person"
          }, 
          {
            "familyName": "Radinsky", 
            "givenName": "Kira", 
            "type": "Person"
          }, 
          {
            "familyName": "Hofmann", 
            "givenName": "Katja", 
            "type": "Person"
          }
        ], 
        "genre": "chapter", 
        "id": "sg:pub.10.1007/978-3-319-06028-6_77", 
        "inLanguage": [
          "en"
        ], 
        "isAccessibleForFree": false, 
        "isFundedItemOf": [
          {
            "id": "sg:grant.6194214", 
            "type": "MonetaryGrant"
          }
        ], 
        "isPartOf": {
          "isbn": [
            "978-3-319-06027-9", 
            "978-3-319-06028-6"
          ], 
          "name": "Advances in Information Retrieval", 
          "type": "Book"
        }, 
        "name": "A Case for Hubness Removal in High\u2013Dimensional Multimedia Retrieval", 
        "pagination": "687-692", 
        "productId": [
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/978-3-319-06028-6_77"
            ]
          }, 
          {
            "name": "readcube_id", 
            "type": "PropertyValue", 
            "value": [
              "b01501480830ffe0d67914ca8eef34992da50fe550930a07cce01b8084c0728c"
            ]
          }, 
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1036963898"
            ]
          }
        ], 
        "publisher": {
          "location": "Cham", 
          "name": "Springer International Publishing", 
          "type": "Organisation"
        }, 
        "sameAs": [
          "https://doi.org/10.1007/978-3-319-06028-6_77", 
          "https://app.dimensions.ai/details/publication/pub.1036963898"
        ], 
        "sdDataset": "chapters", 
        "sdDatePublished": "2019-04-15T18:12", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000001_0000000264/records_8681_00000266.jsonl", 
        "type": "Chapter", 
        "url": "http://link.springer.com/10.1007/978-3-319-06028-6_77"
      }
    ]
     

    Download the RDF metadata as:  json-ld nt turtle xml License info

    HOW TO GET THIS DATA PROGRAMMATICALLY:

    JSON-LD is a popular format for linked data which is fully compatible with JSON.

    curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-06028-6_77'

    N-Triples is a line-based linked data format ideal for batch operations.

    curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-06028-6_77'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-06028-6_77'

    RDF/XML is a standard XML format for linked data.

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-06028-6_77'


     

    This table displays all metadata directly associated to this object as RDF triples.

    147 TRIPLES      23 PREDICATES      37 URIs      20 LITERALS      8 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/978-3-319-06028-6_77 schema:about anzsrc-for:08
    2 anzsrc-for:0801
    3 schema:author Nc941a3dc343644bc9a410bb472ab0947
    4 schema:citation sg:pub.10.1007/978-3-642-13818-8_34
    5 sg:pub.10.1007/978-3-642-20841-6_16
    6 sg:pub.10.1023/b:visi.0000029664.99615.94
    7 https://doi.org/10.1002/sam.11161
    8 https://doi.org/10.1016/j.cviu.2005.09.012
    9 https://doi.org/10.1109/iccp.2011.6047899
    10 https://doi.org/10.1109/iccv.2005.77
    11 https://doi.org/10.1109/tsa.2002.800560
    12 https://doi.org/10.1145/1639714.1639771
    13 https://doi.org/10.5244/c.23.2
    14 schema:datePublished 2014
    15 schema:datePublishedReg 2014-01-01
    16 schema:description This work investigates the negative effects of hubness on multimedia retrieval systems. Because of a problem of measuring distances in high-dimensional spaces, hub objects are close to an exceptionally large part of the data while anti-hubs are far away from all other data points. In the case of similarity based retrieval, hub objects are retrieved over and over again while anti-hubs are nonexistent in the retrieval lists. We investigate textual, image and music data and show how re-scaling methods can avoid the problem and decisively improve the overall retrieval quality. The observations of this work suggest to make hubness analysis an integral part when building a retrieval system.
    17 schema:editor N5f981b91c20a40f3819ca3caa8dcfb2e
    18 schema:genre chapter
    19 schema:inLanguage en
    20 schema:isAccessibleForFree false
    21 schema:isPartOf N9ecb5c100a394226ab68e6ef0c727905
    22 schema:name A Case for Hubness Removal in High–Dimensional Multimedia Retrieval
    23 schema:pagination 687-692
    24 schema:productId N1f8eea1c95d44b04af31bcad6c809dab
    25 N677c4c62a3134df2ba5a5951e36d223c
    26 Necc06cf71aea4e69a595fc24fdb06d1c
    27 schema:publisher N057fe965b6f54f9eae550f135f4aef0d
    28 schema:sameAs https://app.dimensions.ai/details/publication/pub.1036963898
    29 https://doi.org/10.1007/978-3-319-06028-6_77
    30 schema:sdDatePublished 2019-04-15T18:12
    31 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    32 schema:sdPublisher Nd9d093d3ca8f416e841d7bf7a5d71e31
    33 schema:url http://link.springer.com/10.1007/978-3-319-06028-6_77
    34 sgo:license sg:explorer/license/
    35 sgo:sdDataset chapters
    36 rdf:type schema:Chapter
    37 N050ef429278c44f1b0319e463cb1293e rdf:first Na14b7dcb5be441e4a0e6b60e2b8de9eb
    38 rdf:rest N198e6db9931d419c8ae5880bd7aee81d
    39 N057fe965b6f54f9eae550f135f4aef0d schema:location Cham
    40 schema:name Springer International Publishing
    41 rdf:type schema:Organisation
    42 N16a401493bb7472a8afa709d77ae5dee schema:familyName de Rijke
    43 schema:givenName Maarten
    44 rdf:type schema:Person
    45 N198e6db9931d419c8ae5880bd7aee81d rdf:first N421c735348ce4a4cb6d5d0f503eee241
    46 rdf:rest N75184d403ed141faa47f71b86cfcc858
    47 N1f8eea1c95d44b04af31bcad6c809dab schema:name dimensions_id
    48 schema:value pub.1036963898
    49 rdf:type schema:PropertyValue
    50 N28eca40cced14db28f16f5974535caa4 schema:familyName Kenter
    51 schema:givenName Tom
    52 rdf:type schema:Person
    53 N421c735348ce4a4cb6d5d0f503eee241 schema:familyName Zhai
    54 schema:givenName ChengXiang
    55 rdf:type schema:Person
    56 N5f981b91c20a40f3819ca3caa8dcfb2e rdf:first N16a401493bb7472a8afa709d77ae5dee
    57 rdf:rest Nf131a4aaa78e4b8f9ca4aeba77c4805f
    58 N677c4c62a3134df2ba5a5951e36d223c schema:name doi
    59 schema:value 10.1007/978-3-319-06028-6_77
    60 rdf:type schema:PropertyValue
    61 N75184d403ed141faa47f71b86cfcc858 rdf:first Nc6759b2483914274b6e032c1fd9f7aea
    62 rdf:rest Na7bdcc2360a34f9399f845e134cb411c
    63 N9ecb5c100a394226ab68e6ef0c727905 schema:isbn 978-3-319-06027-9
    64 978-3-319-06028-6
    65 schema:name Advances in Information Retrieval
    66 rdf:type schema:Book
    67 Na13e29e470f74f799e023f507fe0ff78 rdf:first sg:person.014252560763.36
    68 rdf:rest rdf:nil
    69 Na14b7dcb5be441e4a0e6b60e2b8de9eb schema:familyName de Vries
    70 schema:givenName Arjen P.
    71 rdf:type schema:Person
    72 Na7bdcc2360a34f9399f845e134cb411c rdf:first Nc699c3bdd4e0489db82545d653540721
    73 rdf:rest Na81de4261b564117b477270364e05ee6
    74 Na81de4261b564117b477270364e05ee6 rdf:first Nd5960056b9ac4d5097b98c642ef9187d
    75 rdf:rest rdf:nil
    76 Nc6759b2483914274b6e032c1fd9f7aea schema:familyName de Jong
    77 schema:givenName Franciska
    78 rdf:type schema:Person
    79 Nc699c3bdd4e0489db82545d653540721 schema:familyName Radinsky
    80 schema:givenName Kira
    81 rdf:type schema:Person
    82 Nc941a3dc343644bc9a410bb472ab0947 rdf:first sg:person.014564424411.74
    83 rdf:rest Ncb2b43864c6f49d29fb35865bf3e03a1
    84 Ncb2b43864c6f49d29fb35865bf3e03a1 rdf:first sg:person.0757053000.20
    85 rdf:rest Na13e29e470f74f799e023f507fe0ff78
    86 Nd5960056b9ac4d5097b98c642ef9187d schema:familyName Hofmann
    87 schema:givenName Katja
    88 rdf:type schema:Person
    89 Nd9d093d3ca8f416e841d7bf7a5d71e31 schema:name Springer Nature - SN SciGraph project
    90 rdf:type schema:Organization
    91 Necc06cf71aea4e69a595fc24fdb06d1c schema:name readcube_id
    92 schema:value b01501480830ffe0d67914ca8eef34992da50fe550930a07cce01b8084c0728c
    93 rdf:type schema:PropertyValue
    94 Nf131a4aaa78e4b8f9ca4aeba77c4805f rdf:first N28eca40cced14db28f16f5974535caa4
    95 rdf:rest N050ef429278c44f1b0319e463cb1293e
    96 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    97 schema:name Information and Computing Sciences
    98 rdf:type schema:DefinedTerm
    99 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
    100 schema:name Artificial Intelligence and Image Processing
    101 rdf:type schema:DefinedTerm
    102 sg:grant.6194214 http://pending.schema.org/fundedItem sg:pub.10.1007/978-3-319-06028-6_77
    103 rdf:type schema:MonetaryGrant
    104 sg:person.014252560763.36 schema:affiliation https://www.grid.ac/institutes/grid.11375.31
    105 schema:familyName Tomašev
    106 schema:givenName Nenad
    107 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014252560763.36
    108 rdf:type schema:Person
    109 sg:person.014564424411.74 schema:affiliation https://www.grid.ac/institutes/grid.432019.d
    110 schema:familyName Schnitzer
    111 schema:givenName Dominik
    112 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014564424411.74
    113 rdf:type schema:Person
    114 sg:person.0757053000.20 schema:affiliation https://www.grid.ac/institutes/grid.432019.d
    115 schema:familyName Flexer
    116 schema:givenName Arthur
    117 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0757053000.20
    118 rdf:type schema:Person
    119 sg:pub.10.1007/978-3-642-13818-8_34 schema:sameAs https://app.dimensions.ai/details/publication/pub.1024575487
    120 https://doi.org/10.1007/978-3-642-13818-8_34
    121 rdf:type schema:CreativeWork
    122 sg:pub.10.1007/978-3-642-20841-6_16 schema:sameAs https://app.dimensions.ai/details/publication/pub.1033916131
    123 https://doi.org/10.1007/978-3-642-20841-6_16
    124 rdf:type schema:CreativeWork
    125 sg:pub.10.1023/b:visi.0000029664.99615.94 schema:sameAs https://app.dimensions.ai/details/publication/pub.1052687286
    126 https://doi.org/10.1023/b:visi.0000029664.99615.94
    127 rdf:type schema:CreativeWork
    128 https://doi.org/10.1002/sam.11161 schema:sameAs https://app.dimensions.ai/details/publication/pub.1041615658
    129 rdf:type schema:CreativeWork
    130 https://doi.org/10.1016/j.cviu.2005.09.012 schema:sameAs https://app.dimensions.ai/details/publication/pub.1004784969
    131 rdf:type schema:CreativeWork
    132 https://doi.org/10.1109/iccp.2011.6047899 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094037868
    133 rdf:type schema:CreativeWork
    134 https://doi.org/10.1109/iccv.2005.77 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094132829
    135 rdf:type schema:CreativeWork
    136 https://doi.org/10.1109/tsa.2002.800560 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061786085
    137 rdf:type schema:CreativeWork
    138 https://doi.org/10.1145/1639714.1639771 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006240814
    139 rdf:type schema:CreativeWork
    140 https://doi.org/10.5244/c.23.2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1099325633
    141 rdf:type schema:CreativeWork
    142 https://www.grid.ac/institutes/grid.11375.31 schema:alternateName Jožef Stefan Institute
    143 schema:name Artificial Intelligence Laboratory, Jožef Stefan Institute, Ljubljana, Slovenia
    144 rdf:type schema:Organization
    145 https://www.grid.ac/institutes/grid.432019.d schema:alternateName Austrian Research Institute for Artificial Intelligence
    146 schema:name Austrian Research Institute for Artificial Intelligence (OFAI), Vienna, Austria
    147 rdf:type schema:Organization
     




    Preview window. Press ESC to close (or click here)


    ...