Skip to content

Latest commit

 

History

History
259 lines (190 loc) · 11.2 KB

README.md

File metadata and controls

259 lines (190 loc) · 11.2 KB

Annif tutorial

The tutorial includes short video presentations and hands-on exercises. Two example data sets are provided to be used in the exercises.

The tutorial was initially organized at SWIB19 and later updated for other conferences and occasions, the materials are freely available for self-study; see the excercises below.

Prerequisites

You will need a computer with sufficient resources (at least 8GB RAM and 20GB free disk space) to be able to install Annif and complete the exercises on your own computer. Installation of Annif is one topic of the exercises.

Note also that it might be convenient to have either Docker or VirtualBox installed beforehand.

However, if you cannot install Annif on your own computer, you can use the GitHub Codespaces installation, which is described in Exercise 1.2. In this setup Annif will be running in a GitHub-hosted machine, which you will access via a remote terminal running in your browser. For this you are required to have an account on GitHub.

Getting the tutorial materials

To complete the exercises of this tutorial, you will need a local copy of the materials, especially the data sets (unless you use the pre-built VirtualBox VM, which includes them). The easiest way to get them is to either clone this repository or download it as a zip archive from GitHub (click the green "Code" button near the top for clone and download options).

When you have the files locally, you also need to download the example full text documents for either or both data sets. The downloads are automated using make - see the README files for both data sets (yso-nlf, stw-zbw) for details.

Upcoming help sessions

From time to time, we organize (online or in-person) help sessions for people working on the tutorial exercises. To register, you should have watched the videos and at least attempted to complete the exercises. Info will be posted here.

  • TBA

Past help sessions

See here

Exercises and videos

Welcome to the actual Annif tutorial content. There are video-only lectures that are prefixed with 🎞️. Exercises marked with 💻 require some coding, and those with 📖 are for reading only. NB! If you have problems with viewing the pdf files on GitHub, you can download them or e.g. try a different browser (see also discussion about this problem).

🎞️ Introduction and overview

Video

The exercises drawn with thick borders and a blue background are core, the others are optional extras.

%%{
    init: {
       "flowchart": {
           "nodeSpacing": 20,
           "rankSpacing": 45,
           "useMaxWidth": true,
           "curve": "linear"
        }
    }
}%%
flowchart TD
    classDef core fill:#ADD8E6,stroke:#000000,stroke-width:2px;
    classDef optional fill:#ffffff,stroke:#000000,stroke-width:1px;

    install([install]) --> tfidf([TFIDF])
    tfidf --> webui([Web UI])
    webui --> eval([evaluate])
    eval --> mllm([MLLM])
    mllm --> ensemble([ensemble])
    ensemble --> nn_ensemble([NN ensemble])
    ensemble --> custom([Custom corpus])
    ensemble --> dvc([DVC])
    ensemble --> huggingfacehub([Hugging Face Hub])
    mllm --> ft([Hogwarts/fastText])
    mllm --> lang_filter([Languages & filtering])
    webui --> rest([REST API])
    rest --> production([Production use])
    eval --> omikuji([Omikuji])
    omikuji --> classification([Classification])
    class install core
    class tfidf core
    class webui core
    class eval core
    class mllm core
    class lang_filter optional
    class ensemble core
    class dvc optional
    class rest optional
    class production optional
    class omikuji optional
    class classification optional
    class ft optional
    class custom optional
    class huggingfacehub optional
    class nn_ensemble optional
Loading

💻 1. Installation

Select your installation type. If you don’t know what to choose, we suggest using VirtualBox.

🎞️ Data sets

This tutorial provides two example data sets; one of them should be chosen to be used in the exercises.

💻 2. TFIDF project

The basic functionality of Annif is introduced by setting up and training a project using a TFIDF model.

🎞️ Algorithms

The principles of the algorithm types used by Annif models are presented.

💻 3. Web UI

The web user interface of Annif allows quick testing of projects.

💻 [Optional] REST API

The REST API of Annif can be used for integrating Annif with other systems.

📖 [Optional] Production use

Here is described aspects to consider when going from testing and development phase to a production-ready deployment of Annif.

💻 4. Metrics & evaluation

Quantitative testing and comparison of projects against standard metrics can be done using the eval command.

💻 [Optional] Omikuji project

Omikuji is a tree-based associative machine learning model that often produces very good results, but requires more resources than the TFIDF model. This exercise is optional, because training an Omikuji model on the full datasets can take around 40 minutes.

💻 [Optional] Automated classification

Annif can also be used for multiclass classification, where the goal is to choose the correct class among mutually exclusive classes. This exercise demonstrates automated classification using the well known "Twenty Newsgroups" data set.

💻 5. MLLM project

MLLM is a lexical algorithm for matching terms in document text to terms in a controlled vocabulary.

💻 [Optional] Hogwarts Sorting Hat using fastText

Yet another algorithm you can try is fastText, which can also work on the level of individual characters.

📖 [Optional] Languages and filtering

The ability of Annif to process text in a given language depends on the choice of the analyzer, which performs text preprocessing. Sometimes it might be useful to filter out parts of the document that are not in the main language of the document.

💻 6. Ensemble project

An ensemble project combines results from the projects set up in previous exercises.

💻 [Optional] Neural network ensemble project

A neural network ensemble can be trained to intelligently combine the results from the base projects.

💻 [Optional] Custom corpus

A big challenge in applying Annif to own data is gathering documents and converting them to form a corpus in suitable format. In this exercise metadata from arXiv articles are used to form a corpus, which can be used to train Annif models.

💻 [Optional] Data Version Control

Data Version Control (DVC) eases maintaining machine learning projects. In this exercise a DVC pipeline is used to set up, train and evaluate Annif projects.

💻 [Optional] Hugging Face Hub

🤗 Hugging Face is an ecosystem and collaboration platform for AI use and development. In this exercise ready-to-use Annif projects are downloaded from a Hugging Face Hub repository.

🎞️ Closing

Summary of the material in the tutorial and some pointers to further information.

Video


Authors

The tutorial material was created by:

  • Osma Suominen, National Library of Finland
  • Mona Lehtinen, National Library of Finland
  • Juho Inkinen, National Library of Finland
  • Anna Kasprzik, ZBW - Leibniz Information Centre for Economics
  • Moritz Fürneisen, ZBW - Leibniz Information Centre for Economics

License

The materials created for this tutorial (presentations and exercises) are licensed under a Creative Commons Attribution 4.0 International License.

CC BY 4.0

The data sets were collected from other sources and have their own licensing; see each individual data set for details.