[17:35:07] <ToAruShiroiNeko>	 ello Amir1
[17:35:16] <ToAruShiroiNeko>	 want to work on something in the hack session? :)
[18:27:42] <ToAruShiroiNeko>	 O_o
[18:27:47] <ToAruShiroiNeko>	 O_o
[18:52:07] <Amir1>	 hey ToAruShiroiNeko, I was afk
[18:52:25] <ToAruShiroiNeko>	 I noticed :)
[18:53:01] <Amir1>	 What Can I do? I can't promise
[18:53:27] <ToAruShiroiNeko>	 well, I was hoping you to generate stop words locally
[18:53:34] <ToAruShiroiNeko>	 say 1000 or 1500 of them per wiki
[18:53:57] <ToAruShiroiNeko>	 once you have a library of 1000-1500 of them per wiki you can run tf-idf on this new list
[18:54:03] <ToAruShiroiNeko>	 to find words common in all wikis
[18:54:28] <ToAruShiroiNeko>	 1500 because we will get local stop words as well as many many on wiki specific ones
[18:54:42] <ToAruShiroiNeko>	 on the final list we would take the top 250 or 500 or something
[18:55:07] <ToAruShiroiNeko>	 you wanted to work on this on tuesday
[18:55:23] <ToAruShiroiNeko>	 but the tf=idf approach is already there I think
[18:55:36] <ToAruShiroiNeko>	 and would take uite a bit of time to run them
[18:56:22] <Amir1>	 hmm
[18:56:22] <Amir1>	 ok
[18:59:32] <ToAruShiroiNeko>	 I am holding back on generating language models just yet
[18:59:39] <ToAruShiroiNeko>	 since I think this will have good signal gain
[19:15:34] <ToAruShiroiNeko>	 oh I am reading through your wikidata mail
[19:21:01] <ToAruShiroiNeko>	 kian looks like somehting uite exciting. :)
[19:26:01] <Amir1>	 :)