A script was written to extract the lexicon data and the rules required for SuMor to generate correct analysis. The endings were integrated in the flexion file and the words were added to the dict.tsv.
| # | Objectives | Associated Deliverables |
|---|---|---|
| 1 | Get all the CDLI data (rules and words) | Write a script that scraps all the required information |
| 2 | Integrate the data to the generation files | Add rules to the flexion file and words to the dictionary |
| 3 | Test if the given outputs are correct | Write tests in Makefiles generating 22 correct analysis on 36 words |
Learnt about Sumerian grammar and SFST.
The unusual difficulties in the scraping the files and trouble-shooting
In future, I plan to have a more polished design for SuMor as the code right now is spaghetti.