a digital toolkit for analysing and visualizing bibliographic metadata and texts

What is AVOBMAT?

The development of AVOBMAT began in 2017 with the objective of creating a tool that is capable of performing critical and interactive analysis of bibliographic metadata AND texts with data-driven and NLP (Natural Language Processing) methods supported by AI (Artificial Intelligence) techniques in a number of languages. 

The Features of AVOBMAT


Text and data mining large library & research databases

content analysis

Content analyis

Metadata analysis and visualization

Interactive metadata analysis & visualizations

preprocessing options

8 Preprocessing options

Topic modeling, correlations, visualizations

Network analysis

Network analysis

52 available languages

N-gram viewer


Over 100 metadata fields

Individual configuration

Individual configuration of all analytical tools

TagSpheres (context of a word)

Metadata enrichment & cleaning

Google Drive import

Live Google Drive import

Significant text analysis

Significant text analysis (comparing corpora)

language detection

Automatic language detection

search modes

Fast advanced, fuzzy, proximity and commandline searches

Keyword in Context / Concordance

gender analysis

Gender analysis of authors

Reproducibility (import/export)

lexical diversity metrics

8 Lexical diversity metrics

What can AVOBMAT offer?

How can it help your research?

  • Explore your large digital collections in innovative & interactive ways with customizable preprocessing, analysis & visualization tools
  • Discover new insights, unveil overlooked connections, themes, trends & patterns 
  • Critically analyze & interpret texts, (meta)data & visualizations 
  • Identify biases and errors in your databases at scale (e.g. selection, metadata, classification) to make more informed decisions about your research (questions)
  • Discover novel type of evidence & test old hypotheses


Zsolt Szántó

Data Scientist
University of Szeged

József Seres

Data Scientist
University of Szeged

Miklós Csiky

Web Design
Corvinus University of Budapest

Gábor Berend

Senior Assistant Professor
Research Group on Artificial Intelligence
University of Szeged & Hungarian Academy of Sciences

Róbert Péter

Project Coordinator
Associate professor
Department of English Studies
University of Szeged

Close & distant reading

Close reading

Focus on the interpretation of few, individual documents.


Distant reading

Exploring the great unread: unveiling and analysing repeated patterns, hidden connections, trends, themes and parallels in large quantity of texts.



The combination of close and distant reading.

Big data in the humanities

Big (meta)data and texts are dirty, noisy, situated and relational. 
They reflect the biases and assumptions of their creators and the power structures involved.
That’s why they should be cleaned, contextualized and critically interpreted with appropriate methods and tools, bearing in mind the strengths and limitations of the latter.

Get in touch

If you have any questions or want to use AVOBMAT in your project or library, please drop us a message:

Stay in touch

If you would like to be notified about the development and release of AVOBMAT, sign up below: