The 'Darmstadt Indexing Approach' (DIA) is a dictionary-based indexing approach for automatic indexing from document titles and abstracts, with descriptors from a prescribed indexing vocabulary. The approach uses an indexing dictionary which contains information about terms, descriptors and weighting functions r(t,s), where r(t,s) is a probabilistic weight for the term-descriptor pair (t,s) (see below). This information can be derived from manually indexed documents. The indexing task starts with the identification of terms in the text. In contrast to other approaches, which are based on a binary decision about the occurrence of a term (and then only count the number of occurrences within the document), several forms of occurrence which correspond to different levels of certainty of identification as well as to the significance of the term with respect to the document are distinguished.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Log-linear indexing functions and the Darmstadt indexing approach


    Additional title:

    Ein neues Wahrscheinlichkeitsmodell für die Darmstädter Indexierungsmethode


    Contributors:
    Fuhr, N. (author)


    Publication date :

    1989


    Size :

    5 Seiten, 8 Quellen


    Type of media :

    Report


    Type of material :

    Print


    Language :

    English




    INDEXING APPARATUS AND INDEXING METHOD

    JORDAN S ERICKSON / LOUIE MICHAEL K | European Patent Office | 2022

    Free access

    Indexing apparatus and indexing method

    ERICKSON JEFFREY S / LOUIE MICHAEL K | European Patent Office | 2022

    Free access

    INDEXING APPARATUS AND METHOD OF INDEXING

    ERICKSON JORDAN S / LOUIE MICHAEL K | European Patent Office | 2022

    Free access

    INDEXING APPARATUS AND METHOD OF INDEXING

    JORDAN S ERICKSON / LOUIE MICHAEL K | European Patent Office | 2022

    Free access

    Indexing apparatus and method of indexing

    ERICKSON JORDAN S / LOUIE MICHAEL K | European Patent Office | 2024

    Free access