frankier / finnvocabcollect Goto Github PK

This code was used in the collection of TallVocabL2Fi.

Robertson, F., Chang & L., Söyrinki, S. (2022).
TallVocabL2Fi: An Extensive Mapping of 15 Finnish L2 Learners' Vocabulary.
In Language Resources and Evaluation Conference (LREC 2022).

This code is not actively developed and is provided as-is for the purposes of providing complete information about TallVocabL2Fi. You are welcome to use use the code in your own research but it comes unsupported. If you want help developing something similar, I may be able to help. Please contact me and we can discuss collaboration.

Apache v2

This project has two parts, the word list pipeline, which is a Snakemake project, and the second is the self assessment website itself which is a Quart (similar to Flask) project.

Instructions for the website, which has it's own set of dependencies/pyproject.toml/Dockerfile are in selfassess.

The Snakemake project is backed by the Python module in finnwordlist. To build the Snakemake project run:

$ poetry install
$ snakemake -j1 -C WIKIPARSE_DB=/path/to/wikiparse/defns.db

Where WIKIPARSE_DB is a parsed sqlite database from wikiparse.

Other things here:

• proc_interest.py: Process the CSV of the expressions of interest CSV.
• plot_en_wordlist_freqs.py: Plot a frequency histogram of the words in SVL12K and all English words known to wordfreq for comparison with the plots for the Finnish wordlist. You can run this with the Snakemake rule plot_en_wordlist_freqs.