Excerpt |
---|
The language detector stage uses OpenNLP (https://opennlp.apache.org/) and its language detector model to identify the language of a text block. |
Operates On: Lexical Items with TEXT_BLOCK flag.
Library: saga-lang-detector-stage
Saga_is_recognizer Recognizer false
Tip |
---|
It can detect 103 languages outputting ISO 639-3 language codes. (https://opennlp.apache.org/news/model-langdetect-183.html) |
...
Note |
---|
The model works better with longer texts |
...
containing at least |
...
two sentences. |
...
It is important to configure this stage earlier in the pipeline and before tokenizing the text. |
Include Page | ||||
---|---|---|---|---|
|
No configuration parameters are needed.
Parameter | ||||||
---|---|---|---|---|---|---|
|
Code Block | ||||||
---|---|---|---|---|---|---|
| ||||||
"atLeastOneFlag": []
"boundaryFlags": ["SENTENCE_SPLIT", "TEXT_BLOCK_SPLIT"]
"confidenceAdjustment": 1
"debug": false
"langModel": "langdetect-183.bin"
"dontProcessFlags": []
"requiredFlags": []
"skipFlags": [] | ||||||
Code Block | ||||||
| ||||||
{
"type":"LangDetectorStage",
} |
As you can see, the first sentence is tagged with "LANG_ENG" and the second sentence with "LANG_SPA". For
In this case, a sentence breaker stage was configured before the language detector stage. This way As a result, language identification could happen can occur at the sentence level.
...
LANG_??? - Flags all text blocks where a language was identified.
Note |
---|
Notice '???' at the end of the Flag. This is replaced by |
an ISO |
three letter language code. |
For example, if Spanish is detected, |
the three letter code is SPA, |
and the Flag will be "LANG_SPA" |
Vertex Flags
...
Info |
---|
No vertices are created in this stage |