2014-11-10 - 版本历史

2014年11月11日 (二) 08:06 Liuc

2014-11-11T08:06:17Z

2014-11-11T08:02:02Z

2014-11-11T00:38:13Z

‎Domain specific LM

2014-11-11T00:36:56Z

‎Domain specific LM

2014-11-11T00:31:58Z

‎Text Processing

2014-11-11T00:30:34Z

‎QA

2014-11-11T00:25:37Z

‎tag LM

2014-11-11T00:24:55Z

‎tag LM

2014-11-11T00:23:14Z

‎tag LM

2014-11-11T00:22:46Z

‎tag LM

@@ 第9行： / 第9行： @@
 * Performance improvement found when pruned slightly
 * The result of AURORA 4 will be available soon.
-* details at http://liuc.cslt.org/pages/obd.html
+* details at http://liuc.cslt.org/pages/sparse.html
 ==== RNN AM====
@@ 第20行： / 第20行： @@
 ==== A new nnet training scheduler ====
 * Initial code done. No better than original one considering of taking much more iterations.
 * details at http://liuc.cslt.org/pages/nnet-sched.html

@@ 第7行： / 第7行： @@
 ==== Sparse DNN ====
 * Performance improvement found when pruned slightly
 * The result of AURORA 4 will be available soon.
 ==== RNN AM====
 * Initial nnet seems not very well, need to be pre-trained or test lower learn-rate.
 * For AURORA 4 1h/epoch, model train done.
@@ 第18行： / 第17行： @@
 * Adjusting the learning rate.(+)
 * Trying toolkit of Microsoft.(+)
 ==== A new nnet training scheduler ====
-[link here]
+* full history at http://liuc.cslt.org/pages/nnet-sched.html
 * Initial code done. No better than original one considering of taking much more iterations.
 ====Noise training====

@@ 第203行： / 第203行： @@
 ====Domain specific LM====
 * domain lm
-:* merger weibo、baiduhi and baiduzhidao lm and test (need result)
+:* merger weibo、baiduhi and baiduzhidao lm and test ('''need result''')
 :* confirm the size of alpa with xiaomin for business application.(like e-13)
 :* get the general test data from miaomin .this test set may get from online.

@@ 第203行： / 第203行： @@
 ====Domain specific LM====
 * domain lm
-:* weibo lm with pruning 0 10 10 20 20 testing done. weibo lm with pruning 0 10 8 8 8 under testing. weibo lm without pruning 4/8 done.
+:* merger weibo、baiduhi and baiduzhidao lm and test (need result)
-:* merger weibo、baiduhi and baiduzhidao lm and test ('''this week''')
 :* confirm the size of alpa with xiaomin for business application.(like e-13)
 :* get the general test data from miaomin .this test set may get from online.
 *  new dict.

@@ 第274行： / 第274行： @@
 ===QA===
 deatil:[http://cslt.riit.tsinghua.edu.cn/mediawiki/index.php/Hulan-2014-11-06]
-===Spell mistake===
+====Spell mistake====
 :* retrain the ngram model('''caoli''')
 :* prepare the test and development set('''caoli''')
-===improve fuzzy match===
+====improve fuzzy match====
 * add Synonyms similarity using MERT-4 method
-===improve lucene search===
+====improve lucene search====
 * our vsm method
 {| border="2px"
@@ 第311行： / 第311行： @@
 :* using MERT-4 method to get good value of multi-feature.like IDF,NER,baidu_weight,keyword etc.('''liurong this month''')
-===Multi-Scene Recognition===
+====Multi-Scene Recognition====
 * add the triples search to QA engine
 :* discuss the detail and give a report.('''liurong''')

@@ 第273行： / 第273行： @@
 ===QA===
-* lucene Optimization
+===improve fuzzy match===
-:* rewrite the method to select the 50 standard question not same template.('''this week''')
+* add Synonyms similarity using MERT-4 method
-:* test the boost keyword weight and extract the synonyms word.('''this week''')
-:* check the word segment for template.('''this week''')
+===improve lucene search===
-:* min-segment method improve the accuracy.(0.61->0.66)
+* our vsm method
-:* check the query method for getting lucene information and to rewrite the score method like the idf value.
+{| border="2px"
-* test
+|+ different result in lucene
-:* test the different idf vale from baidu sougou in fuzzymatch.('''this week''')
+|-
-:* need to check the other 10% error.('''this week''')
+! method !!lucene  !! vsm_idf(haiguan) !! VSM_idf(baidu) !! vsm_idf(tain) !! vsm_idf(calculate)
-* spell check
+|-
-:* simple demo done.
+! Accary
 * new inter will install SEMPRE

@@ 第222行： / 第222行： @@
 :*
 {| border="2px"
-|+ result with tag-lm
+|+ no "北京" in corpus with tag-lm
 |-
 ! method !!baeline  !! weight0.1 !! weight0.5 !! weight1 !! weight2 !! weight3

@@ 第231行： / 第231行： @@
 !"北京"
 |6/10||4/10||4/10||2/10||1/10||0
 |-
 |}

@@ 第220行： / 第220行： @@
 * set new test
-:* {| border="2px"
+:*
 |+ result with tag-lm
 |-