Báo cáo khoa học: "Mining Refinements to Online Instructions from User Generated Content" doc

... 2012. c 2012 Association for Computational Linguistics Spice it Up? Mining Refinements to Online Instructions from User Generated Content Gregory Druck Yahoo! Research gdruck@gmail.com Bo Pang Yahoo! ... revealed that “how to questions are the most popular (Pang and Kumar, 2011). People consult online resources to answer technical questions like “how to put music on my i...

Ngày tải lên: 07/03/2014, 18:20

... according to their dis- tributional similarity to the nouns “method” and “model”. Subsequently, the noun “method” is used to find transitive verbs and rank them according to their similarity to “introduce” ... experiments to in- directly evaluate the quality of the automatically generated cue phrase variants. Given an abstract of an article and a sentence extracted from the art...

Ngày tải lên: 08/03/2014, 02:21

... plausible tokenization; in that case, the plausible token is useless to us. Thus, we define a strongly plausible token, abbre- viated sp-token, which is a token that is induced by some plausible tokenization. ... need to rely on the (possibly tiny) query log of the single user at hand, due to privacy or security concerns; moreover, as noted earlier about kohli, the statistics of one u...

Ngày tải lên: 17/03/2014, 00:20

... quotes (“A B”) and submitted to Google to narrow the results to texts with exact phrases. The Web crawl yielded 17657 web pages, stripped from HTML and other web tags to filter out non-textual content. ... as for example not enough or too much. 3 English translations (morphosyntactic tags in parenthe- ses): [0] seem to (inf), [1] seemed to (sg,pri,perf,m), [2] seemed to (sg,pri...

Ngày tải lên: 24/03/2014, 03:20

... researchers tried to encode do- 2 http://www.nlm.nih.gov/mesh/ 3 http://www.nlm.nih.gov/research/umls/ Total radiology records 1,954 Total tokens 51,940 Total ICD-9-Codes 45 Total code instances ... typical radiology report is shown below: 68 786 Symptoms involving respiratory system and other chest symptoms (0/698) 786.0 Dyspnea and respiratory abnormalities (0/98) 786.1 Stridor (0/0)...

Ngày tải lên: 31/03/2014, 00:20

... aliphatics, often recalcitrant to degradation. Among these molecules examples are toluene, that is converted to p-hydroxytoluene by toluene-4-mono-oxygenase in Pseudo- monas mendocina KR1 [5]; xylene, ... T3MO, toluene-3-monooxygenase from Pseudomonas pickettii PKO1; T4MO, toluene-4-monooxygenase from Pseudomonas mendocina KR1; Xyl/TMO, xylene/toluene monooxygenase from Pseudomonas...

Ngày tải lên: 31/03/2014, 01:20

... and an extension to interac- tive real-time NLG. 760 Segment123 From: Node1 To: Node2 On: “Main Street” Segment124 From: Node2 To: Node3 On: “Main Street” Segment125 From: Node3 To: Node4 On: “Park ... system monitors the user s position and computes new, corrective instructions when the user leaves the intended path. We evaluate our system using a driving simulator, and com- par...

Ngày tải lên: 31/03/2014, 21:20

... features. 1 Introduction Review mining and summarization aims to extract users’ opinions towards specific products from reviews and provide an easy -to- understand sum- mary of those opinions for potential ... Wang Key Laboratory of Computational Linguistics (Peking University) Ministry of Education, China wanghf@pku.edu.cn Abstract This paper proposes a method to ex- tract product featu...

Ngày tải lên: 20/02/2014, 09:20

... metalinguistic information from text, the first is- sue to tackle is how to obtain a reliable set of can- didate sentences from free text for input into the next phases of extraction. From our initial ... receptor Markers/ Operators: termed Table 4. Sample entry of MID The final processing stage presents metrics shown in Figure 4, using a ß factor of 1.0 to esti- mate F-m...

Ngày tải lên: 20/02/2014, 15:20

... we could get many factors of human judg- ments, machine learning will be a good method to combine these factors together. As proved in the recent literature, learning from regression is of ... character- istic of news is its timeliness. News come from the year 2002 are nearly totally unrelated to that from the year 2003. It can be seen from Table 3 that we have got the expect...

Ngày tải lên: 31/03/2014, 00:20