Annif: Research APIs

[hortongn]

Explore the API documentation. Note how we might use it to connect Annif to Scholar.

https://github.com/NatLibFi/Annif/wiki/REST-API

[haitzlm]

Preliminary Work- references

• https://github.com/NatLibFi/Annif-tutorial/blob/master/exercises/OPT_rest_api.md
• https://github.com/NatLibFi/Annif-client ("This is a minimal Python 3.x client library for accessing the Annif REST API which can be used for automated subject indexing and classification of text documents."
• Jupyter Notebook: https://colab.research.google.com/drive/1S0JX_befWeECKCJEZYSvcGzCnJfDePF0?usp=sharing#scrollTo=wW2j36Wad194

[haitzlm]

• REST API: The Annif REST API provides a way to interact with Annif over HTTP. It allows users to train models, run predictions, and manage projects and vocabularies. The API is designed to be easy to use and can be accessed from any programming language that supports HTTP requests.

• Command Line Interface (CLI): The Annif CLI provides a command-line interface for interacting with Annif. It can be used to train models, run predictions, and manage projects and vocabularies. The CLI is particularly useful for users who prefer to work with command-line tools or who need to automate their workflows.

• Python API: The Annif Python API provides a Python interface for interacting with Annif. It allows users to train models, run predictions, and manage projects and vocabularies directly from Python code. The Python API is particularly useful for users who want to integrate Annif into a larger Python-based application.

• R API: The Annif R API provides an R interface for interacting with Annif. It allows users to train models, run predictions, and manage projects and vocabularies directly from R code. The R API is particularly useful for users who prefer to work with R or who want to integrate Annif into an R-based application.

[scherztc]

https://api.annif.org/

[scherztc]

We might want to look at into this:

This service is for demonstration and testing purposes; for integrations in production systems use the API service of Finto AI.

https://www.kiwi.fi/display/Finto/Finto+AI%2C+a+service+for+automated+subject+indexing

https://ai.finto.fi/v1/ui/