Legacy Product

Fusion 5.10
    Fusion 5.10

    Detect Language Index Stage

    Table of Contents

    The Detect Language index stage (called the Language Detection stage in versions earlier than 3.0) operates over one or more fields in the Pipeline Document. The contents of each field are analyzed using the Language Detection Library for Java, which is an open-source project hosted on GitHub. The analyzer returns the ID of the language which best matches the contents of that field, if any. These IDs can be returned as an annotation on the Pipeline Document context, or as annotation on each field analyzed.

    The language identification algorithm breaks the text in each source field into ngrams and compares them to sets of ngrams compiled from all the different language versions of the Wikipedia. This library will only produce reasonable results for document fields which are comparable in length, vocabulary, and style to the known texts compiled from the Wikipedia. Caveats are discussed below.

    If a positive language identification is made, that information is added to the Pipeline Document according to the choice of configuration property "Output Type". If the language annotation is added to the PipelineDocument context object, the name of the context key string is specified by configuration property "Output Key". For Output Type configuration property "Document", per-field language annotations are added to the document using a parallel naming convention where the name of the language identification field starts with the name of the analyzed field and has an additional suffix string, default value "_lang". For example, if a document contains fields named "plot_summary_txt" and "user_reviews_txt" to be analyzed, if the software can detect the language, it will add fields "plot_summary_txt_lang" and "user_reviews_txt_lang".

    There is also an option to allow detection of multiple languages. This can be achieved by setting "Return all detected languages and their confidence scores." to true. In this case, the detected languages will be either set as document fields in a form of "Field Name_Document Postfix.Language:Confidence", or as a field with name "Output Key" in the Context having a dictionary of following form "{ "language":"probability" }" as a value. Example Document fields could look like this: "plot_summary_txt_lang.pl_: [0.99]", "plot_summary_txt_lang.en_: [0.99]" when languages pl and en would be detected.

    Languages

    The Language Detection Library for Java has build-in profiles for many languages. If there is a set of Wikipedia entries written in a language, it is likely that the Language Detection Library can identify texts written in this language.

    Caveats

    This library should produce reasonable results on document fields which are comparable in length, vocabulary, and style to the known texts compiled from the Wikipedia.

    The documentation lists the following challenges:

    • This software does not work as well when the input text to analyze is short, or unclean. For example tweets.

    • When a text is written in multiple languages, the default algorithm of this software is not appropriate. You can try to split the text (by sentence or paragraph) and detect the individual parts. Running the language guesser on the whole text will just tell you the language that is most dominant, in the best case.

    • This software cannot handle it well when the input text is in none of the expected (and supported) languages.

    • Detection of unwanted languages (for example the stage might detect some language that is not even used in the input data because of some language similarities). By default, the stage uses a full array of available languages for detection (List here). If one wants to only use selected languages, this can be configured

    Configuration

    When entering configuration values in the UI, use unescaped characters, such as \t for the tab character. When entering configuration values in the API, use escaped characters, such as \\t for the tab character.

    Detect the language of the input source fields using https://github.com/optimaize/language-detector. If the output is stored to the document, there will be a new field created for each source field with the language

    skip - boolean

    Set to true to skip this stage.

    Default: false

    label - string

    A unique label for this stage.

    <= 255 characters

    condition - string

    Define a conditional script that must result in true or false. This can be used to determine if the stage should process or not.

    source - array[string]required

    The fields/context keys to detect on. May be a String Template. See https://github.com/antlr/stringtemplate4/blob/master/doc/index.md

    languages - array[object]

    The language profiles to use for language detection (in a form of language codes)

    object attributes:{code required : {
     display name: Language code
     type: string
    }    }

    outputKey - string

    The name of the key to insert into the context if the output type is 'context'. The value is a map of source name to language. May be a String Template. See https://github.com/antlr/stringtemplate4/blob/master/doc/index.md

    Default: languages

    documentPostfix - string

    The postfix to add to the source name when storing the results on the document (via the output type).

    Default: _lang

    outputType - stringrequired

    Select whether the flag should be set on the document or in the Pipeline Context.

    Default: document

    Allowed values: documentcontext

    minimumConfidence - number

    Minimum confidence score (in range 0 - 1) for a language to be detected. This filters out cases where languages are detected with low confidence.

    Default: 0.5

    returnAllMatchedWithConfidenceScores - boolean

    Return all languages whose confidence scores exceed the minimum and their corresponding confidence scores. By default (unchecked) only the language with the highest confidence score is returned.

    Default: false