Targeting diversity in photographic retrieval task with commonsense knowledge

Supheakmungkol Sarin, Wataru Kameyama

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    1 Citation (Scopus)

    Abstract

    Image search engines have a very limited usefulness since it is still difficult to provide different users with what they are searching for. This is because most research efforts to date have only been concentrating on relevancy rather than diversity which is also a quite important factor, given that the search engine knows nothing about the user's context. In this paper, we describe our approach for ImageCLEF 2008 photographic retrieval task. The novelty of our technique is the use of AnalogySpace [3], the reasoning technique over commonsense knowledge for document and query expansion, which aims to increase the diversity of the results. Our proposed technique combines AnalogySpace mapping with other two mappings namely, location and full-text. We then re-rank the resulting images from the mapping by trying to eliminate duplicate and near duplicate results in the top 20. We present our preliminary experiments and the results conducted using the IAPR TC-12 photographic collection with 20,000 natural still photographs. The results show that our integrated method with AnalogySpace yields slightly better performance in terms of cluster recall and the number of relevant photographs retrieved. We finally identify the weakness in our approach and ways on how the system could be optimized and improved.

    Original languageEnglish
    Title of host publicationCEUR Workshop Proceedings
    PublisherCEUR-WS
    Volume1174
    Publication statusPublished - 2008
    Event2008 Working Notes for CLEF Workshop, CLEF 2008 - Co-located with the 12th European Conference on Digital Libraries, ECDL 2008 - Aarhus, Denmark
    Duration: 2008 Sep 172008 Sep 19

    Other

    Other2008 Working Notes for CLEF Workshop, CLEF 2008 - Co-located with the 12th European Conference on Digital Libraries, ECDL 2008
    CountryDenmark
    CityAarhus
    Period08/9/1708/9/19

    Fingerprint

    Search engines
    Experiments

    Keywords

    • AnalogySpace
    • Commonsense knowledge
    • Diversity
    • Image retrieval
    • Query/document expansion
    • Re-rank

    ASJC Scopus subject areas

    • Computer Science(all)

    Cite this

    Sarin, S., & Kameyama, W. (2008). Targeting diversity in photographic retrieval task with commonsense knowledge. In CEUR Workshop Proceedings (Vol. 1174). CEUR-WS.

    Targeting diversity in photographic retrieval task with commonsense knowledge. / Sarin, Supheakmungkol; Kameyama, Wataru.

    CEUR Workshop Proceedings. Vol. 1174 CEUR-WS, 2008.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Sarin, S & Kameyama, W 2008, Targeting diversity in photographic retrieval task with commonsense knowledge. in CEUR Workshop Proceedings. vol. 1174, CEUR-WS, 2008 Working Notes for CLEF Workshop, CLEF 2008 - Co-located with the 12th European Conference on Digital Libraries, ECDL 2008, Aarhus, Denmark, 08/9/17.
    Sarin S, Kameyama W. Targeting diversity in photographic retrieval task with commonsense knowledge. In CEUR Workshop Proceedings. Vol. 1174. CEUR-WS. 2008
    Sarin, Supheakmungkol ; Kameyama, Wataru. / Targeting diversity in photographic retrieval task with commonsense knowledge. CEUR Workshop Proceedings. Vol. 1174 CEUR-WS, 2008.
    @inproceedings{032cba6b6fe6439194959f245e8079fa,
    title = "Targeting diversity in photographic retrieval task with commonsense knowledge",
    abstract = "Image search engines have a very limited usefulness since it is still difficult to provide different users with what they are searching for. This is because most research efforts to date have only been concentrating on relevancy rather than diversity which is also a quite important factor, given that the search engine knows nothing about the user's context. In this paper, we describe our approach for ImageCLEF 2008 photographic retrieval task. The novelty of our technique is the use of AnalogySpace [3], the reasoning technique over commonsense knowledge for document and query expansion, which aims to increase the diversity of the results. Our proposed technique combines AnalogySpace mapping with other two mappings namely, location and full-text. We then re-rank the resulting images from the mapping by trying to eliminate duplicate and near duplicate results in the top 20. We present our preliminary experiments and the results conducted using the IAPR TC-12 photographic collection with 20,000 natural still photographs. The results show that our integrated method with AnalogySpace yields slightly better performance in terms of cluster recall and the number of relevant photographs retrieved. We finally identify the weakness in our approach and ways on how the system could be optimized and improved.",
    keywords = "AnalogySpace, Commonsense knowledge, Diversity, Image retrieval, Query/document expansion, Re-rank",
    author = "Supheakmungkol Sarin and Wataru Kameyama",
    year = "2008",
    language = "English",
    volume = "1174",
    booktitle = "CEUR Workshop Proceedings",
    publisher = "CEUR-WS",

    }

    TY - GEN

    T1 - Targeting diversity in photographic retrieval task with commonsense knowledge

    AU - Sarin, Supheakmungkol

    AU - Kameyama, Wataru

    PY - 2008

    Y1 - 2008

    N2 - Image search engines have a very limited usefulness since it is still difficult to provide different users with what they are searching for. This is because most research efforts to date have only been concentrating on relevancy rather than diversity which is also a quite important factor, given that the search engine knows nothing about the user's context. In this paper, we describe our approach for ImageCLEF 2008 photographic retrieval task. The novelty of our technique is the use of AnalogySpace [3], the reasoning technique over commonsense knowledge for document and query expansion, which aims to increase the diversity of the results. Our proposed technique combines AnalogySpace mapping with other two mappings namely, location and full-text. We then re-rank the resulting images from the mapping by trying to eliminate duplicate and near duplicate results in the top 20. We present our preliminary experiments and the results conducted using the IAPR TC-12 photographic collection with 20,000 natural still photographs. The results show that our integrated method with AnalogySpace yields slightly better performance in terms of cluster recall and the number of relevant photographs retrieved. We finally identify the weakness in our approach and ways on how the system could be optimized and improved.

    AB - Image search engines have a very limited usefulness since it is still difficult to provide different users with what they are searching for. This is because most research efforts to date have only been concentrating on relevancy rather than diversity which is also a quite important factor, given that the search engine knows nothing about the user's context. In this paper, we describe our approach for ImageCLEF 2008 photographic retrieval task. The novelty of our technique is the use of AnalogySpace [3], the reasoning technique over commonsense knowledge for document and query expansion, which aims to increase the diversity of the results. Our proposed technique combines AnalogySpace mapping with other two mappings namely, location and full-text. We then re-rank the resulting images from the mapping by trying to eliminate duplicate and near duplicate results in the top 20. We present our preliminary experiments and the results conducted using the IAPR TC-12 photographic collection with 20,000 natural still photographs. The results show that our integrated method with AnalogySpace yields slightly better performance in terms of cluster recall and the number of relevant photographs retrieved. We finally identify the weakness in our approach and ways on how the system could be optimized and improved.

    KW - AnalogySpace

    KW - Commonsense knowledge

    KW - Diversity

    KW - Image retrieval

    KW - Query/document expansion

    KW - Re-rank

    UR - http://www.scopus.com/inward/record.url?scp=84921985616&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84921985616&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:84921985616

    VL - 1174

    BT - CEUR Workshop Proceedings

    PB - CEUR-WS

    ER -