...
Operates On: Lexical Items with TOKEN within TEXT_BLOCK_SPLIT and SENTENCE_SPLIT vertex flags.
Library: saga-name-trainer-stage
Tip |
---|
Models can be trained directly with OpenNLP tools or from the Saga UI. See Name Entity Recognizer in the User Manual for more information on how to create a model using Saga. |
...
Parameter |
---|
summary | if used with automatic pipeline creation, it assigns the tag to which the recognizer belongs to. |
---|
default | match |
---|
name | tagWith |
---|
|
Parameter |
---|
summary | Probability threshold. Will only tag sentences that match better or equal to prob. |
---|
default | 0.95 |
---|
name | prob |
---|
type | double |
---|
|
Parameter |
---|
summary | File location of the model. |
---|
name | model |
---|
|
Parameter |
---|
summary | List of Tags used to normalize the text |
---|
name | normalize |
---|
type | string array |
---|
|
- For example, let's say you want to normalize all different numbers in the text. You can create a "Numeric" tag using the numeric recognizer, that way each different number will me normalized to "{Numeric}".
Saga_config_stage |
---|
boundaryFlags | text block split, sentence split |
---|
stage | NamePredictorStage |
---|
requiredFlags | token, semantic tag |
---|
skipFlags | skip |
---|
|
"tagWith": "component", //NAME-OF-OUTPUT-TAG
"prob": "0.95",
"model": ".\model-file.bin",
"normalize": [] |
Example Output
Saga_graph |
---|
V---------------------------------[ONE MAIN ROTOR BLADE CONTACTED A WIRE WHILE GOING THROUGH MOUNTAINS ON TRAFFIC WATCH.]---------------------------------V
^-------------------------------[ONE MAIN ROTOR BLADE CONTACTED A WIRE WHILE GOING THROUGH MOUNTAINS ON TRAFFIC WATCH]-------------------------------V-[]-^
^-[ONE]-V-[MAIN]-V-----[ROTOR]-----V-[BLADE]-V-[CONTACTED]-V-[A]-V-[WIRE]-V-[WHILE]-V-[GOING]-V-[THROUGH]-V-[MOUNTAINS]-V-[ON]-V-[TRAFFIC]-V-[WATCH]-^
^-[one]-^-[main]-^-----[rotor]-----^-[blade]-^-[contacted]-^-[a]-^-[wire]-^-[while]-^-[going]-^-[through]-^-[mountains]-^-[on]-^-[traffic]-^-[watch]-^
^--[{component}]--^ |
...