The Experts below are selected from a list of 12 Experts worldwide ranked by ideXlab platform

Carolyn Penstein Rose - One of the best experts on this subject based on the ideXlab platform.

  • whatâ s in a domain multi domain learning for multi Attribute data
    North American Chapter of the Association for Computational Linguistics, 2013
    Co-Authors: Mahesh Joshi, Mark Dredze, William W Cohen, Carolyn Penstein Rose
    Abstract:

    Multi-Domain learning assumes that a single Metadata Attribute is used in order to divide the data into so-called domains. However, real-world datasets often have multiple Metadata Attributes that can divide the data into domains. It is not always apparent which single Attribute will lead to the best domains, and more than one Attribute might impact classification. We propose extensions to two multi-domain learning techniques for our multi-Attribute setting, enabling them to simultaneously learn from several Metadata Attributes. Experimentally, they outperform the multi-domain learning baseline, even when it selects the single “best” Attribute.

Mahesh Joshi - One of the best experts on this subject based on the ideXlab platform.

  • whatâ s in a domain multi domain learning for multi Attribute data
    North American Chapter of the Association for Computational Linguistics, 2013
    Co-Authors: Mahesh Joshi, Mark Dredze, William W Cohen, Carolyn Penstein Rose
    Abstract:

    Multi-Domain learning assumes that a single Metadata Attribute is used in order to divide the data into so-called domains. However, real-world datasets often have multiple Metadata Attributes that can divide the data into domains. It is not always apparent which single Attribute will lead to the best domains, and more than one Attribute might impact classification. We propose extensions to two multi-domain learning techniques for our multi-Attribute setting, enabling them to simultaneously learn from several Metadata Attributes. Experimentally, they outperform the multi-domain learning baseline, even when it selects the single “best” Attribute.

Mark Dredze - One of the best experts on this subject based on the ideXlab platform.

  • whatâ s in a domain multi domain learning for multi Attribute data
    North American Chapter of the Association for Computational Linguistics, 2013
    Co-Authors: Mahesh Joshi, Mark Dredze, William W Cohen, Carolyn Penstein Rose
    Abstract:

    Multi-Domain learning assumes that a single Metadata Attribute is used in order to divide the data into so-called domains. However, real-world datasets often have multiple Metadata Attributes that can divide the data into domains. It is not always apparent which single Attribute will lead to the best domains, and more than one Attribute might impact classification. We propose extensions to two multi-domain learning techniques for our multi-Attribute setting, enabling them to simultaneously learn from several Metadata Attributes. Experimentally, they outperform the multi-domain learning baseline, even when it selects the single “best” Attribute.

William W Cohen - One of the best experts on this subject based on the ideXlab platform.

  • whatâ s in a domain multi domain learning for multi Attribute data
    North American Chapter of the Association for Computational Linguistics, 2013
    Co-Authors: Mahesh Joshi, Mark Dredze, William W Cohen, Carolyn Penstein Rose
    Abstract:

    Multi-Domain learning assumes that a single Metadata Attribute is used in order to divide the data into so-called domains. However, real-world datasets often have multiple Metadata Attributes that can divide the data into domains. It is not always apparent which single Attribute will lead to the best domains, and more than one Attribute might impact classification. We propose extensions to two multi-domain learning techniques for our multi-Attribute setting, enabling them to simultaneously learn from several Metadata Attributes. Experimentally, they outperform the multi-domain learning baseline, even when it selects the single “best” Attribute.

Irene Bosch - One of the best experts on this subject based on the ideXlab platform.

  • Metadata driven comparative analysis tool for sequences meta cats an automated process for identifying significant sequence variations that correlate with virus Attributes
    Virology, 2013
    Co-Authors: Brett E Pickett, Mengya Liu, Eva Sadat, R B Squires, Jyothi Noronha, W Jen, Sam Zaremba, L Zhou, Christopher N Larsen, Irene Bosch
    Abstract:

    The Virus Pathogen Resource (ViPR; www.viprbrc.org) and Influenza Research Database (IRD; www.fludb.org) have developed a Metadata-driven Comparative Analysis Tool for Sequences (meta-CATS), which performs statistical comparative analyses of nucleotide and amino acid sequence data to identify correlations between sequence variations and virus Attributes (Metadata). Meta-CATS guides users through: selecting a set of nucleotide or protein sequences; dividing them into multiple groups based on any associated Metadata Attribute (e.g. isolation location, host species); performing a statistical test at each aligned position; and identifying all residues that significantly differ between the groups. As proofs of concept, we have used meta-CATS to identify sequence biomarkers associated with dengue viruses isolated from different hemispheres, and to identify variations in the NS1 protein that are unique to each of the 4 dengue serotypes. Meta-CATS is made freely available to virology researchers to identify genotype-phenotype correlations for development of improved vaccines, diagnostics, and therapeutics.