Python REST Service Install and Configuration

This document details the Python REST Service, and the Python enrichment steps that comes included with Aiimi Insight Engine.

The following enrichment steps are shipped with the Python REST Service:

Classification – This step uses Aiimi’s clustering and classification framework to classify documents using a pre-trained model. Please see the additional documentation for model training, which can be found in the InsightMaker.Python\DocumentClassification\docs folder.
Entity Mapper – This step maps values found in one entity to synonyms of that value and stores them into another entity.
Huggingface Named Entity Recognition – This step extracts named entities from text or documents using statistical methods. This step is more accurate than Spacy but slower.
Language Detection – This step will detect 54 different languages from text.
Phase and Topic Detection – This step extracts repeating phrases from a document or text that are said to be ‘left right complete’. Usually when people are writing about things, they will repeat the core concepts and topics several times. This step extracts these from the text and gives the user a list of core concepts, themes, and topics in the text.
Spacy Named Entity Recognition – This step extracts named entities from text or documents using statistical methods. This step is faster than Huggingface, but less accurate.
Document Summaries – This step summarizes a document into a short multi-sentence summary that can then be used by users to quickly understand what the document is about. There are several algorithms provided, each with different merits.

Note that there may be other services in the endpoints folder – these will be alpha enrichment steps and are unsupported.

This document details the Python REST Service, and the Python enrichment steps that comes included with Aiimi Insight Engine.

The following enrichment steps are shipped with the Python REST Service:

Classification – This step uses Aiimi’s clustering and classification framework to classify documents using a pre-trained model. Please see the additional documentation for model training, which can be found in the InsightMaker.Python\DocumentClassification\docs folder.
Entity Mapper – This step maps values found in one entity to synonyms of that value and stores them into another entity.
Huggingface Named Entity Recognition – This step extracts named entities from text or documents using statistical methods. This step is more accurate than Spacy but slower.
Language Detection – This step will detect 54 different languages from text.
Phase and Topic Detection – This step extracts repeating phrases from a document or text that are said to be ‘left right complete’. Usually when people are writing about things, they will repeat the core concepts and topics several times. This step extracts these from the text and gives the user a list of core concepts, themes, and topics in the text.
Spacy Named Entity Recognition – This step extracts named entities from text or documents using statistical methods. This step is faster than Huggingface, but less accurate.
Document Summaries – This step summarizes a document into a short multi-sentence summary that can then be used by users to quickly understand what the document is about. There are several algorithms provided, each with different merits.

Note that there may be other services in the endpoints folder – these will be alpha enrichment steps and are unsupported.