Page History
| Excerpt |
|---|
Part Of Speech tags a word in a text (corpus) as corresponding to a particular part of speech such as noun, verb, adjective, etc., based on its definition, as well as its context. Using OpenNLP (https://opennlp.apache.org/) and its POS Tagger |
The tagging of each token is done with flags, meaning that no semantic tag is created with this stage.
Operates On: Lexical Items with TOKEN and possibly other flags as specified below.
Library: saga-parts-of-speech-stage
Saga_is_recognizer
| Warning |
|---|
Currently only English is supported |
| Include Page | ||||
|---|---|---|---|---|
|
Configuration Parameters
Parameter summary Threshold within a part of speech is accepted as one default 0.7 name prob type double Parameter summary prefix of the model to use as part of the speech model. Currently only English is supported default en name language Parameter summary Path to the folder where the models are stored name modelPath
| Code Block | ||||||
|---|---|---|---|---|---|---|
| ||||||
{
| ||||||
| Saga_config_stage | ||||||
| ||||||
"prob": 0.7, "language": "en", "modelPath": null, } |
Example Output
| Code Block | ||||
|---|---|---|---|---|
| ||||
V-----------------------------------[Pierre Vinken , 61 years old , will join the board as a nonexecutive director Nov. 29 .]-----------------------------------V ^-[Pierre]-V-[Vinken]-V-[,]-V-[61]-V-[years]-V-[old]-V-[,]-V-[will]-V-[join]-V-[the]-V-[board]-V-[as]-V-[a]-V-[nonexecutive]-V-[director]-V-[Nov.]-V-[29]-V-[.]-^ Item [as] - [TOKEN,ORIGINAL,POS_TOKEN,POS_IN] Item [years] - [TOKEN,ORIGINAL,POS_TOKEN,POS_NNS] Item [old] - [TOKEN,ORIGINAL,POS_TOKEN,POS_JJ] Item [director] - [TOKEN,ORIGINAL,POS_TOKEN,POS_NN] Item [Pierre] - [TOKEN,ORIGINAL,POS_NNP,POS_TOKEN] Item [the] - [TOKEN,ORIGINAL,POS_TOKEN,POS_DT] Item [Nov.] - [TOKEN,ORIGINAL,POS_NNP,POS_TOKEN,HAS_PUNCTUATION] Item [61] - [ALL_DIGITS,TOKEN,ORIGINAL,POS_TOKEN,POS_CD] Item [29] - [ALL_DIGITS,TOKEN,ORIGINAL,POS_TOKEN,POS_CD] Item [will] - [TOKEN,ORIGINAL,POS_TOKEN,POS_MD] Item [,] - [TOKEN,ORIGINAL,ALL_PUNCTUATION,POS_TOKEN,POS_,] Item [,] - [TOKEN,ORIGINAL,ALL_PUNCTUATION,POS_TOKEN,POS_,] Item [join] - [TOKEN,ORIGINAL,POS_TOKEN,POS_VB] Item [board] - [TOKEN,ORIGINAL,POS_TOKEN,POS_NN] Item [.] - [TOKEN,ORIGINAL,ALL_PUNCTUATION,POS_TOKEN,POS_.] Item [nonexecutive] - [TOKEN,ORIGINAL,POS_TOKEN,POS_JJ] Item [a] - [TOKEN,ORIGINAL,POS_TOKEN,POS_DT] Item [Vinken] - [TOKEN,ORIGINAL,POS_NNP,POS_TOKEN] |
Output Flags
Lex-Item Flags:
- TOKEN - All tokens produced are tagged as TOKEN
- POS_TOKEN - Identifies the token as recognized as a part of speech
POS_??? - Flags all TOKENs where a part of speech was recognized.
Note Notice '???' at the end of the Flag. This is replaced by an acronym of the part-to-speech identified.
For example, if a base form verb is detected, the acronym is VB, and the Flag will be "POS_VB"
Vertex Flags:
| Info |
|---|
No vertices are created in this stage |
| Flag | Definition |
|---|---|
POS_CC | Coordinating conjunction |
POS_CD | Cardinal number |
POS_DT | Determiner |
POS_EX | Existential there |
POS_FW | Foreign word |
POS_IN | Preposition or subordinating conjunction |
POS_JJ | Adjective |
POS_JJR | Adjective, comparative |
POS_JJS | Adjective, superlative |
POS_LS | List item marker |
POS_MD | Modal |
POS_NN | Noun, singular or mass |
POS_NNS | Noun, plural |
POS_NNP | Proper noun, singular |
POS_NNPS | 'Proper noun, plural |
POS_PDT | Predeterminer |
POS_POS | Possessive ending |
POS_PRP | Personal pronoun |
POS_PRP$ | Possessive pronoun |
POS_RB | Adverb |
POS_RBR | Adverb, comparative |
POS_RBS | Adverb, superlative |
POS_RP | Particle |
POS_SYM | Symbol |
POS_TO | to |
POS_UH | Interjection |
POS_VB | Verb, base form |
POS_VBD | Verb, past tense |
POS_VBG | Verb, gerund or present participle |
POS_VBN | Verb, past participle |
POS_VBP | Verb, non-3rd person singular present |
POS_VBZ | Verb, 3rd person singular present |
POS_WDT | Wh-determiner |
POS_WP | Wh-pronoun |
POS_WP$ | Possessive wh-pronoun |
POS_WRB | Wh-adverb |
Overview
Content Tools