-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathMakefile
234 lines (205 loc) · 15.7 KB
/
Makefile
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
SHELL=/bin/bash
LC_ALL=C
CP=bin:jar/mysql-connector-java-5.1.20-bin.jar:jar/JFlex.jar:jar/ark-1.0-SNAPSHOT.jar:jar/commons-cli-1.2.jar:jar/commons-math3-3.0.jar:jar/opencsv-2.3.jar
DIR=/mnt/freddy
EVENT_LIST=obamacare japan bp wallstreet
#EVENT_LIST=facebook_ipo obamacare japan bp wallstreet
MODEL_LIST=lda np_lda decay_lda gauss_lda
prec_recall1:
cat $(DIR)/results/facebook_ipo/K=8/lda/retrieved_tweets_cleaned.txt | java -cp $(CP) CalculatePrecRecall | uniq > $(DIR)/results/facebook_ipo/K=8/lda/prec_recall.txt &
cat $(DIR)/results/facebook_ipo/K=8/np_lda/retrieved_tweets_cleaned.txt | java -cp $(CP) CalculatePrecRecall | uniq > $(DIR)/results/facebook_ipo/K=8/np_lda/prec_recall.txt &
cat $(DIR)/results/facebook_ipo/K=8/decay_lda/retrieved_tweets_cleaned.txt | java -cp $(CP) CalculatePrecRecall | uniq > $(DIR)/results/facebook_ipo/K=8/decay_lda/prec_recall.txt &
cat $(DIR)/results/facebook_ipo/K=8/gauss_lda/retrieved_tweets_cleaned.txt | java -cp $(CP) CalculatePrecRecall | uniq > $(DIR)/results/facebook_ipo/K=8/gauss_lda/prec_recall.txt &
prec_recall2:
cat $(DIR)/results/facebook_ipo/K=8/lda/prec_recall.txt | java -cp $(CP) CalculatePrecRecall2 | uniq > $(DIR)/results/facebook_ipo/K=8/lda/prec_recall_2.txt &
cat $(DIR)/results/facebook_ipo/K=8/np_lda/prec_recall.txt | java -cp $(CP) CalculatePrecRecall2 | uniq > $(DIR)/results/facebook_ipo/K=8/np_lda/prec_recall_2.txt &
cat $(DIR)/results/facebook_ipo/K=8/decay_lda/prec_recall.txt | java -cp $(CP) CalculatePrecRecall2 | uniq > $(DIR)/results/facebook_ipo/K=8/decay_lda/prec_recall_2.txt &
cat $(DIR)/results/facebook_ipo/K=8/gauss_lda/prec_recall.txt | java -cp $(CP) CalculatePrecRecall2 | uniq > $(DIR)/results/facebook_ipo/K=8/gauss_lda/prec_recall_2.txt &
prec_recall3:
cat $(DIR)/results/facebook_ipo/K=8/lda/prec_recall_2.txt | java -cp $(CP) CalculatePrecRecall3 > results/facebook_ipo/K=8/lda/prec_recall_3.txt &
cat $(DIR)/results/facebook_ipo/K=8/np_lda/prec_recall_2.txt | java -cp $(CP) CalculatePrecRecall3 > results/facebook_ipo/K=8/np_lda/prec_recall_3.txt &
cat $(DIR)/results/facebook_ipo/K=8/decay_lda/prec_recall_2.txt | java -cp $(CP) CalculatePrecRecall3 > results/facebook_ipo/K=8/decay_lda/prec_recall_3.txt &
cat $(DIR)/results/facebook_ipo/K=8/gauss_lda/prec_recall_2.txt | java -cp $(CP) CalculatePrecRecall3 > results/facebook_ipo/K=8/gauss_lda/prec_recall_3.txt &
ngrams_corpus:
n=1 ; while [[ $$n -le 3 ]]; do \
cut -d $$'\t' -f 3 $(DIR)/data/tweets/sfbay_facebook_ipo.txt | bin/opennlp SimpleTokenizer | java -cp $(CP) GetNgram $${n} | sort | java -cp $(CP) Reduce > $(DIR)/data/tweets/sfbay_facebook_ipo.$${n}grams ; \
(( n = n + 1 )); \
done
ngrams_wiki:
n=1 ; while [[ $$n -le 3 ]]; do \
cat testset/$(EVENT).wiki | bin/opennlp SentenceDetector bin/en-sent.bin | bin/opennlp SimpleTokenizer | java -cp $(CP) GetNgram $${n} | sort | java -cp $(CP) Reduce > testset/$(EVENT).wiki.$${n}grams; \
(( n = n + 1 )); \
done
excel:
echo "Summary 1" > results/$(EVENT)/K=$(K)/summary.txt
cut -d $$'\t' -f 2,3 $(DIR)/results/$(EVENT)/K=$(K)/lda/coherent.txt >> results/$(EVENT)/K=$(K)/summary.txt
echo $$'\n' >> results/$(EVENT)/K=$(K)/summary.txt
echo "Summary 2" >> results/$(EVENT)/K=$(K)/summary.txt
cut -d $$'\t' -f 2,3 $(DIR)/results/$(EVENT)/K=$(K)/np_lda/coherent.txt >> results/$(EVENT)/K=$(K)/summary.txt
echo $$'\n' >> results/$(EVENT)/K=$(K)/summary.txt
echo "Summary 3" >> results/$(EVENT)/K=$(K)/summary.txt
cut -d $$'\t' -f 2,3 $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/coherent.txt >> results/$(EVENT)/K=$(K)/summary.txt
echo $$'\n' >> results/$(EVENT)/K=$(K)/summary.txt
echo "Summary 4" >> results/$(EVENT)/K=$(K)/summary.txt
cut -d $$'\t' -f 2,3 $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/coherent.txt >> results/$(EVENT)/K=$(K)/summary.txt
echo $$'\n' >> results/$(EVENT)/K=$(K)/summary.txt
count_ngrams:
n=1 ; while [[ $$n -le 3 ]]; do \
java -cp $(CP) CountNgram $(DIR)/results/$(EVENT)/K=$(K)/lda/coherent.$${n}grams testset/$(EVENT).wiki.$${n}grams > $(DIR)/results/$(EVENT)/K=$(K)/coherent_$${n}grams.txt ; \
java -cp $(CP) CountNgram $(DIR)/results/$(EVENT)/K=$(K)/np_lda/coherent.$${n}grams testset/$(EVENT).wiki.$${n}grams >> $(DIR)/results/$(EVENT)/K=$(K)/coherent_$${n}grams.txt ; \
java -cp $(CP) CountNgram $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/coherent.$${n}grams testset/$(EVENT).wiki.$${n}grams >> $(DIR)/results/$(EVENT)/K=$(K)/coherent_$${n}grams.txt ; \
java -cp $(CP) CountNgram $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/coherent.$${n}grams testset/$(EVENT).wiki.$${n}grams >> $(DIR)/results/$(EVENT)/K=$(K)/coherent_$${n}grams.txt ; \
(( n = n + 1 )); \
done
ngrams_tweets:
n=1 ; while [[ $$n -le 3 ]]; do \
cat $(DIR)/results/$(EVENT)/K=$(K)/lda/coherent.txt | cut -d $$'\t' -f 2 | bin/opennlp SimpleTokenizer | java -cp $(CP) GetNgram $${n} | sort | java -cp $(CP) Reduce > $(DIR)/results/$(EVENT)/K=$(K)/lda/coherent.$${n}grams ; \
cat $(DIR)/results/$(EVENT)/K=$(K)/np_lda/coherent.txt | cut -d $$'\t' -f 2 | bin/opennlp SimpleTokenizer | java -cp $(CP) GetNgram $${n} | sort | java -cp $(CP) Reduce > $(DIR)/results/$(EVENT)/K=$(K)/np_lda/coherent.$${n}grams ; \
cat $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/coherent.txt | cut -d $$'\t' -f 2 | bin/opennlp SimpleTokenizer | java -cp $(CP) GetNgram $${n} | sort | java -cp $(CP) Reduce > $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/coherent.$${n}grams ; \
cat $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/coherent.txt | cut -d $$'\t' -f 2 | bin/opennlp SimpleTokenizer | java -cp $(CP) GetNgram $${n} | sort | java -cp $(CP) Reduce > $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/coherent.$${n}grams ; \
(( n = n + 1 )); \
done
summarize_all:
k=4 ; while [[ $$k -le 12 ]]; do \
make summarize EVENT=$(EVENT) K=$${k} ; \
(( k = k + 1 )) ; \
done
#summarize the tweets using all the models and find their scores
summarize: ngrams_tweets count_ngrams
mkdir -p results/$(EVENT)/K=$(K)
java -cp $(CP) Bleu $(DIR) $(EVENT) $(K) > results/$(EVENT)/K=$(K)/coherent_bleu.txt
#delete deprecated files
rm_files:
for event in $(EVENT_LIST); do \
k=8 ; while [[ $${k} -le 8 ]]; do \
for model in $(MODEL_LIST); do \
mv $(DIR)/results/$${event}/K=$${k}/$${model}/distribution.txt results/$${event}/K=$${k}/$${model}/ ; \
mv $(DIR)/results/$${event}/K=$${k}/$${model}/time.txt results/$${event}/K=$${k}/$${model}/ ; \
done ; \
(( k = k + 1 )); \
done ; \
done
#Look for more tweets from the original corpus using the "trained" model
search_gauss_lda:
cat $(TWEETS) | java -cp $(CP) gauss_lda.StreamSearch --prefix_dir=$(DIR) --num_topics=$(K) --decay=$(DECAY) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)" > $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/retrieved_tweets.txt
#Look for more tweets from the original corpus using the "trained" model
search_decay_lda:
cat $(TWEETS) | java -cp $(CP) decay_lda.StreamSearch --prefix_dir=$(DIR) --num_topics=$(K) --decay=$(DECAY) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)" > $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/retrieved_tweets.txt
#Look for more tweets from the original corpus using the "trained" model
search_np_lda:
cat $(TWEETS) | java -cp $(CP) np_lda.StreamSearch --prefix_dir=$(DIR) --num_topics=$(K) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)" > $(DIR)/results/$(EVENT)/K=$(K)/np_lda/retrieved_tweets.txt
#Look for more tweets from the original corpus using the "trained" model
search_lda:
cat $(TWEETS) | java -cp $(CP) lda.StreamSearch --prefix_dir=$(DIR) --num_topics=$(K) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)" > $(DIR)/results/$(EVENT)/K=$(K)/lda/retrieved_tweets.txt
infer_all: infer_lda infer_np_lda infer_decay_lda infer_gauss_lda
load_all: load_lda load_np_lda load_decay_lda load_gauss_lda
#Loads the model without additional training
load_lda:
mkdir -p $(DIR)/results/$(EVENT)/K=$(K)/lda/
cat $(DIR)/data/tweets/$(EVENT)_K=$(K).lda | java -cp $(CP) lda.InferModel --iterations=0 --prefix_dir=$(DIR) --num_topics=$(K) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)"
k=1 ; while [[ $${k} -le $(K) ]]; do \
sort -r -g -t$$'\t' -k1,1 $(DIR)/results/$(EVENT)/K=$(K)/lda/topic\_$${k}.txt > $(DIR)/results/$(EVENT)/K=$(K)/lda/tmp ; \
mv $(DIR)/results/$(EVENT)/K=$(K)/lda/tmp $(DIR)/results/$(EVENT)/K=$(K)/lda/topic\_$${k}.txt ; \
((k = k + 1)) ; \
done
cut -d $$'\t' -f 1 $(DIR)/results/$(EVENT)/K=$(K)/lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/lda/distribution.txt
cut -d $$'\t' -f 4 $(DIR)/results/$(EVENT)/K=$(K)/lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/lda/time.txt
#Loads the model without additional training
load_np_lda:
mkdir -p $(DIR)/results/$(EVENT)/K=$(K)/np_lda/
cat $(DIR)/data/tweets/$(EVENT)_K=$(K).np_lda | java -cp $(CP) np_lda.InferModel --iterations=0 --prefix_dir=$(DIR) --num_topics=$(K) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)"
k=1 ; while [[ $${k} -le $(K) ]]; do \
sort -r -g -t$$'\t' -k1,1 $(DIR)/results/$(EVENT)/K=$(K)/np_lda/topic\_$${k}.txt > $(DIR)/results/$(EVENT)/K=$(K)/np_lda/tmp ; \
mv $(DIR)/results/$(EVENT)/K=$(K)/np_lda/tmp $(DIR)/results/$(EVENT)/K=$(K)/np_lda/topic\_$${k}.txt ; \
((k = k + 1)) ; \
done
cut -d $$'\t' -f 1 $(DIR)/results/$(EVENT)/K=$(K)/np_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/np_lda/distribution.txt
cut -d $$'\t' -f 4 $(DIR)/results/$(EVENT)/K=$(K)/np_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/np_lda/time.txt
#Loads the model without additional training
load_decay_lda:
mkdir -p $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/
cat $(DIR)/data/tweets/$(EVENT)_K=$(K).decay_lda | java -cp $(CP) decay_lda.InferModel --iterations=0 --prefix_dir=$(DIR) --num_topics=$(K) --decay=$(DECAY) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)"
k=1 ; while [[ $${k} -le $(K) ]] ; do \
sort -r -g -t$$'\t' -T$(HOME)/data -k1,1 $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/topic\_$${k}.txt > $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/tmp ; \
mv $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/tmp $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/topic\_$${k}.txt ; \
((k = k + 1)) ; \
done
cut -d $$'\t' -f 1 $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/distribution.txt
cut -d $$'\t' -f 4 $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/time.txt
#Loads the model without additional training
load_gauss_lda:
mkdir -p results/$(EVENT)/K=$(K)/gauss_lda/
mkdir -p $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/
cat $(DIR)/data/tweets/$(EVENT)_K=$(K).gauss_lda | java -cp $(CP) gauss_lda.InferModel --iterations=0 --prefix_dir=$(DIR) --num_topics=$(K) --decay=$(DECAY) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)"
k=1 ; while [[ $${k} -le $(K) ]] ; do \
sort -r -g -t$$'\t' -T$(HOME)/data -k1,1 $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/topic\_$${k}.txt > $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/tmp ; \
mv $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/tmp $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/topic\_$${k}.txt ; \
((k = k + 1)) ; \
done
cut -d $$'\t' -f 1 $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/distribution.txt
cut -d $$'\t' -f 4 $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/time.txt
#Inference for LDA (Topic Model)
infer_lda:
mkdir -p results/$(EVENT)/K=$(K)/lda/
mkdir -p $(DIR)/results/$(EVENT)/K=$(K)/lda/
cat $(DIR)/data/tweets/$(EVENT)_init.lda | java -cp $(CP) lda.InferModel --prefix_dir=$(DIR) --num_topics=$(K) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)" > results/$(EVENT)/K=$(K)/lda/log.txt
cut -d : -f 3 results/$(EVENT)/K=$(K)/lda/log.txt > results/$(EVENT)/K=$(K)/lda/mle.txt
k=1 ; while [[ $${k} -le $(K) ]]; do \
sort -r -g -t$$'\t' -k1,1 $(DIR)/results/$(EVENT)/K=$(K)/lda/topic\_$${k}.txt > $(DIR)/results/$(EVENT)/K=$(K)/lda/tmp ; \
mv $(DIR)/results/$(EVENT)/K=$(K)/lda/tmp $(DIR)/results/$(EVENT)/K=$(K)/lda/topic\_$${k}.txt ; \
((k = k + 1)) ; \
done
cut -d $$'\t' -f 1 $(DIR)/results/$(EVENT)/K=$(K)/lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/lda/distribution.txt
cut -d $$'\t' -f 4 $(DIR)/results/$(EVENT)/K=$(K)/lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/lda/time.txt
#Inference for NounPhrases + Topic Model
infer_np_lda:
mkdir -p results/$(EVENT)/K=$(K)/np_lda/
mkdir -p $(DIR)/results/$(EVENT)/K=$(K)/np_lda/
cat $(DIR)/data/tweets/$(EVENT)_init.np | java -cp $(CP) np_lda.InferModel --prefix_dir=$(DIR) --num_topics=$(K) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)" > results/$(EVENT)/K=$(K)/np_lda/log.txt
cut -d : -f 3 results/$(EVENT)/K=$(K)/np_lda/log.txt > results/$(EVENT)/K=$(K)/np_lda/mle.txt
k=1 ; while [[ $${k} -le $(K) ]]; do \
sort -r -g -t$$'\t' -k1,1 $(DIR)/results/$(EVENT)/K=$(K)/np_lda/topic\_$${k}.txt > $(DIR)/results/$(EVENT)/K=$(K)/np_lda/tmp ; \
mv $(DIR)/results/$(EVENT)/K=$(K)/np_lda/tmp $(DIR)/results/$(EVENT)/K=$(K)/np_lda/topic\_$${k}.txt ; \
((k = k + 1)) ; \
done
cut -d $$'\t' -f 1 $(DIR)/results/$(EVENT)/K=$(K)/np_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/np_lda/distribution.txt
cut -d $$'\t' -f 4 $(DIR)/results/$(EVENT)/K=$(K)/np_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/np_lda/time.txt
#Inference for the Decay Topic Model
infer_decay_lda:
mkdir -p results/$(EVENT)/K=$(K)/decay_lda/
mkdir -p $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/
cat $(DIR)/data/tweets/$(EVENT)_init.np | java -cp $(CP) decay_lda.InferModel --prefix_dir=$(DIR) --num_topics=$(K) --decay=$(DECAY) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)" > results/$(EVENT)/K=$(K)/decay_lda/log.txt
cut -d : -f 3 results/$(EVENT)/K=$(K)/decay_lda/log.txt > results/$(EVENT)/K=$(K)/decay_lda/mle.txt
k=1 ; while [[ $${k} -le $(K) ]] ; do \
sort -r -g -t$$'\t' -T$(HOME)/data -k1,1 $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/topic\_$${k}.txt > $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/tmp ; \
mv $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/tmp $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/topic\_$${k}.txt ; \
((k = k + 1)) ; \
done
cut -d $$'\t' -f 1 $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/distribution.txt
cut -d $$'\t' -f 4 $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/decay_lda/time.txt
#Inference for the Gaussian Topic Model
infer_gauss_lda:
mkdir -p results/$(EVENT)/K=$(K)/gauss_lda/
mkdir -p $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/
cat $(DIR)/data/tweets/$(EVENT)_init.np | java -cp $(CP) gauss_lda.InferModel --prefix_dir=$(DIR) --num_topics=$(K) --decay=$(DECAY) --experiment_name=$(EVENT) --ignore_words="$(IGNORE)" > results/$(EVENT)/K=$(K)/gauss_lda/log.txt
cut -d : -f 3 results/$(EVENT)/K=$(K)/gauss_lda/log.txt > results/$(EVENT)/K=$(K)/gauss_lda/mle.txt
k=1 ; while [[ $${k} -le $(K) ]] ; do \
sort -r -g -t$$'\t' -T$(HOME)/data -k1,1 $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/topic\_$${k}.txt > $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/tmp ; \
mv $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/tmp $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/topic\_$${k}.txt ; \
((k = k + 1)) ; \
done
cut -d $$'\t' -f 1 $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/distribution.txt
cut -d $$'\t' -f 4 $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/transition.txt > $(DIR)/results/$(EVENT)/K=$(K)/gauss_lda/time.txt
#This rule is for tokenizing and finding noun phrases in tweets
find_np:
cat $(DIR)/data/tweets/$(DATA)_$(EVENT).txt | java -cp $(CP) np_lda.TagTweets > $(DIR)/data/tweets/$(EVENT)_init.np
cat $(DIR)/data/tweets/$(DATA)_$(EVENT).txt | java -cp $(CP) lda.TagTweets2 > $(DIR)/data/tweets/$(EVENT)_init.lda
compile:
mkdir -p bin/
javac -d bin/ -cp $(CP) src/org/json/*.java
javac -d bin/ -cp $(CP) src/*.java
javac -d bin/ -cp $(CP) src/lda/*.java
javac -d bin/ -cp $(CP) src/np_lda/*.java
javac -d bin/ -cp $(CP) src/decay_lda/*.java
javac -d bin/ -cp $(CP) src/gauss_lda/*.java
javac -d bin/ -cp $(CP) src/visualize/*.java