# mysql
mysql -u hcai6 -p

check single day
python news_clus_from_file.py /home/wtong8/NewsTwitter/news/2014-11-18.txt /home/wtong8/NewsTwitter/tweets/2014-11-18/ 20 50  text out1118_t50_redund.20
compare two days news
python news_clus_2day.py /home/wtong8/NewsTwitter/news/2014-11-17.txt /home/wtong8/NewsTwitter/news/2014-11-18.txt 50 50 text out1117_1118.50.50
python news_clus_from_file_lda.py /home/wtong8/NewsTwitter/news/2014-11-17.txt tmp 40 1 all tmp.txt
python news_clus_from_file_aggl_multiday.py /home/wtong8/NewsTwitter/news/ 2015-01-01 2015-01-04 tmp 50 1 all tmp50.txt
time python clus_from_file_aggl_multiday.py /home/wtong8/NewsTwitter/news/ 2014-12-20 2015-01-04 /home/wtong8/NewsTwitter/tweets/ 20 1000 text outauto1225_0104.txt
python2.7 gen_matlab_ent.py /home/wtong8/NewsTwitter/news/ 2014-12-20 2015-01-04 /home/wtong8/NewsTwitter/tweets/ 50 1000 text outentsurf.txt 5
load.py /home/wtong8/NewsTwitter/news/ 2014-12-20 2015-01-04 /home/wtong8/NewsTwitter/tweets/ 50 1000 text outentsurf.txt
python testIndex.py "query OR query1" 50
python gen_matlab_tweet.py tweetSony ~/Dropbox/linux_buffer/testX_thresh_10K_50.mat 2014-12-20 ~/Dropbox/linux_buffer/tweetX.mat ~/Dropbox/linux_buffer/tweetCleanSony

--generate pickle
python runPLSA.py /home/wtong8/NewsTwitter/news/ 2014-12-20 2015-01-04 tmp 30 0 text outentsurf12200104.txt 5
--load pickle
python loadPLSA.py 30 0

-- kmeans reduce K + plsa reduce K
python dataSplitJesse.py 30 0 jesse30/


select a.ID, b.news_id, b.tweet_id, raw_text, hash_tags from overviews_tweet a join overviews_tweet_related_news b on b.news_id = 6385 and a.ID = b.tweet_id;


awk -F\\t '{print $3". "$9}'  news/2015-04-13.txt|sed 's/::::::::/ /g'> ~/projects/tweet/tweetNews/data/news_2015-04-13.txt
cut -f3 news/2015-04-13.txt > ~/projects/tweet/tweetNews/data/news_title_2015-04-13.txt

# SegPhrase
./train_custom.sh /home/jwang112/projects/embedding/trunk/news.2012.en.shuffled-norm0 RUN_ID  1(auto_label) 0(wordnet_noun) 10(support_threshold) 0(need_unigram)
./parse_custom.sh test_file RUN_ID top_n