I am a bit ahead of the posts with the table of contents. I have posts scheduled up to April 1 and hope to reach April 6 before I leave on a brief holiday April 2. Each week sees a bit of progress. This past week is measured by the graph.
Domain-subdomain will correct itself for guesses longer than 1 word but will have to be thoroughly examined at the end of all things. Stems are simply ambiguous at times so the software can't guess them. Like לך, could be to you or could be go, the imperative of הלך. I could probably resolve some of these if I took vowels into account in the algorithms. But I didn't because at the time I wrote them, ths idea would have buried me.
Overall status is always on the table of contents. As of today 195 chapters to go, 30 to 40 weeks estimate.