Skip to content
Various (linguistic) applications of the BHSA data
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
bits-and-pieces ordinary chapters Nov 19, 2018
easter/tf/c ordinary Nov 19, 2018
excel prep Feb 5, 2018
heads update work Dec 14, 2018
learning before pull Nov 14, 2018
minimalPairs vocab Jan 16, 2018
trees trees Apr 14, 2018
vocabulary question by stephen ku Mar 14, 2018
voyant readme Jan 26, 2018
.gitignore trees as TF-features Jan 21, 2018
LICENSE Initial commit Jan 16, 2018 update readme Apr 7, 2018


Various (linguistic) applications of the BHSA data:

Minimal pairs

Compute pairs of words that have a limited edit distance. If the distance is 1, both words differ in exactly one character.

Various decisions have to be taken:

  • do we feed lexemes (lemmas) or word occurrences (both of course)
  • do we retain accents?
  • do use unicode normalization or denormalization?

We do not provide all options here. But the code is reasonably clear, so, clone it and go your own way!

Done for Hebrew and Greek.


Generate syntax trees from the Hebrew data. The notebook does all BHSA versions. De trees are exported in plain text in bracket notation. The output is in the repo.


A simple notebook to produce list of Hebrew and Aramaic lexemes in order of decreasing frequency.


If you want to put the data in a corpus form, you can use the notebook prepare.ipynb.

You can choose a text format and a granularity, and then the notebook will create a corpus for that in a local _temp directory. Then you can zip it and upload to text processing services such as Voyant.


Export edge relationships between phrases and their head words. A head word is defined as "the word that determines the syntactic category of that phrase" (wikipedia).

You can’t perform that action at this time.