Bots From Extension: cfxai_nlp
CloudFabrix NLP Engine
This extension provides 6 bots.
Bot @nlp:analyze-sentiment
Bot Position In Pipeline: Sink
Do sentiment analysis on a specified input descriptive column
This bot expects a Restricted CFXQL.
Each parameter may be specified using '=' operator and AND logical operation
Following are the parameters expected for this Bot
| Parameter Name | Type | Default Value | Description |
|---|---|---|---|
| description_col* | Text | Name of the input description column which carries text | |
| neg_polarity_col | Text | neg_polarity | Output column name which stores worst negative sentiment score (-1.0 to 0.0) |
| neg_subjectivity_col | Text | neg_subjectivity | Output column name which stores worst negative sentiment subjectivity (0 to 1.0) |
| neg_sentence_col | Text | neg_sentence | Output column name which stores worst negative sentiment sentence |
| pos_polarity_col | Text | pos_polarity | Output column name which stores best positive sentiment score (0.0 to 1.0) |
| pos_subjectivity_col | Text | pos_subjectivity | Output column name which stores best positive sentiment subjectivity (0 to 1.0) |
| pos_sentence_col | Text | pos_sentence | Output column name which stores best positive sentiment sentence |
Bot @nlp:extract-keywords
Bot Position In Pipeline: Sink
Extract keyword phrases from an input descriptive column using Rapid Automatic Keyword Extraction (RAKE) algorithm
This bot expects a Restricted CFXQL.
Each parameter may be specified using '=' operator and AND logical operation
Following are the parameters expected for this Bot
| Parameter Name | Type | Default Value | Description |
|---|---|---|---|
| description_col* | Text | Name of the input description column which carries text | |
| keywords_col | Text | keywords | Output column name which stores the extracted Keywords |
| n | Text | 10 | Max number keyword phrases in the output |
| min_length | Text | 1 | Minimum number of words in the phrase |
| max_length | Text | 4 | Maximum number of words in the phrase |
| lang | Text | NlTK Language code (ex: 'en') |
Example Pipelines Using this Bot
Bot @nlp:extract-named-entities
Bot Position In Pipeline: Sink
Extract named entities from an input descriptive column
This bot expects a Restricted CFXQL.
Each parameter may be specified using '=' operator and AND logical operation
Following are the parameters expected for this Bot
| Parameter Name | Type | Default Value | Description |
|---|---|---|---|
| description_col* | Text | Name of the input description column which carries text | |
| named_entities_col | Text | named_entities | Output column name which stores all the extracted Named Entities |
| custom_named_entities_dict | Text | Optional dictionary dataset that has 'label' and 'pattern' columns to list any custom named entities |
|
| extract_ner_labels | Text | Comma separated list of NER labels. If specified, additional columns with 'ner_' prefix are added for each type of NER label |
Example Pipelines Using this Bot
Bot @nlp:extract-nouns
Bot Position In Pipeline: Sink
Extract noun phrases from an input descriptive column
This bot expects a Restricted CFXQL.
Each parameter may be specified using '=' operator and AND logical operation
Following are the parameters expected for this Bot
| Parameter Name | Type | Default Value | Description |
|---|---|---|---|
| description_col* | Text | Name of the input description column which carries text | |
| nouns_col | Text | nouns | Output column name which stores comma separated list of noun phrases |
Bot @nlp:scrape-content
Bot Position In Pipeline: Sink
Scrape the HTML content from specified 'input_col' and retrieve plain text content
This bot expects a Restricted CFXQL.
Each parameter may be specified using '=' operator and AND logical operation
Following are the parameters expected for this Bot
| Parameter Name | Type | Default Value | Description |
|---|---|---|---|
| input_col* | Text | Name of the input column with HTML text | |
| output_col* | Text | Name of the output column with plain text |
Bot @nlp:scrape-url
Bot Position In Pipeline: Sink
Scrape link specified in 'url' column and retrieve title, and plain text content
This bot expects a Restricted CFXQL.
Each parameter may be specified using '=' operator and AND logical operation
Following are the parameters expected for this Bot
| Parameter Name | Type | Default Value | Description |
|---|---|---|---|
| url_col* | Text | Name of the input column with URL links |