papineni

67 %
33 %
Information about papineni
Entertainment

Published on October 15, 2007

Author: Heather

Source: authorstream.com

Cross-lingual Information Access by Natural Language:  Cross-lingual Information Access by Natural Language Kishore Papineni IBM T.J. Watson Research Center Yorktown Heights, NY Slide2:  Doc collection or the Web Do you mean the one in India or New Jersey? Information access by natural language Where is the Taj Mahal Docs in many languages Slide3:  Source: Global Reach English English 2000 2005 Global Internet User Population Slide4:  Type of Content and Answer Structured (DB) Unstructured (docs) Content Answer Structured Unstructured Transactional Systems (demo later) QA (‘demo’..) IR & Search NL Components:  NL Components Statistical parsing NLU, Named Entities, multilinguality Dialog management conversational web/telephony Statistical Machine Translation learning multilingual statistical dictionaries from corpora Named Entities:  Named Entities Eight Nobel Peace Prize winners, after being refused entry to Burma, came to neighboring Thailand to start the campaign for their fellow laureate. Ms. Suu Kyi won the prize in 1991 and has been held under house arrest for over three years by Rangoon's military junta . Numbers, dates, names of locations, people,… Named Entity Detection (shallow parsing):  Location Organization Named Entity Detection (shallow parsing) Date Location Location Statistical named entity system 1st place 2000 DARPA Mandarin Named Entity task Named entity components are used in QA, TDT, labeling clusters/taxonomies, . Machine Translation - 1998:  Machine Translation - 1998 Readable Editable Marginal Chinese Spanish Portuguese Russian Japanese German Korean French Vietnamese The World - 1998* ~228 Countries >6,700 Languages >39,000 Language, dialect, and alternate names * http://www.sil.org/ethnologue/ Polish Arabic Ukrainian Italian Farsi Serbo-Croatian Thai Dutch Indonesian Hungarian Greek Czech Swedish Machine Translation for Search:  Machine Translation for Search High speed Lower quality tolerable Word order and inflection not that important Training data: parallel or comparable corpora Slide12:  Statistical Translation Models: Fertility: pn(n|f-,f,f+ ) Sense translation: p(e|f-,f,f+) X-lingual Retrieval: Query Translation:  X-lingual Retrieval: Query Translation Chinese Docs IR scoring C => E MT E => C MT Ranked Docs Query English Chinese online “English” for gisting online Caveat: MT isn’t perfect and queries tend to be very short. X-lingual Retrieval: Query & Document Translation:  X-lingual Retrieval: Query & Document Translation Chinese Docs IR scoring C => E MT E => C MT Query English Chinese online “English” for gisting offline “English” Docs IR scoring IR scoring Score Merging Ranked Docs English TREC 9:  TREC 9 TransWhiz: commercial MT from Taiwan IBM SMT: statistical MT on HK corpus X-lingual IR on HK corpus IBM SMT Mono- lingual Trans- Whiz Q D Q+D Q+D+T Slide18:  Post-TREC: More data is better Quality/quantity of training data (more data 200k sentences) > 2x performance! Slide19:  Comparable Corpus Example: SDA newswire French, German Stories independently written, not translations Report same events Comparable corpus advantage: Availability Match domain to task Broader coverage Parallel corpus advantages: Match languages to each other Linguistic structure, details Slide21:  IBM Documents & queries can be in any of English, French, German, or Italian. IBM placed 1st in ’98 and ’99 evaluations. Question Answering:  Question Answering Who won the Nobel Peace Prize in 1991? Eight Nobel Peace Prize winners, after being refused entry to Burma, came to neighboring Thailand to start the campaign for their fellow laureate. Ms. Suu Kyi won the prize in 1991 and has been held under house arrest for over three years by Rangoon's military junta . Fast Match :  Fast Match Answer Class Tag Question Who won the Nobel Peace Prize in 1991? PERSON Question-Answering Architecture:  Question-Answering Architecture TREC-9 DB Query Expansion IR Engine Question Answer Who won the Nobel Peace Prize in 1991? ENCYC DB PERSON Eight Nobel Peace Prize winners, after being refused entry to Burma, came to neighboring Thailand to start the campaign for their fellow laureate. Ms. Suu Kyi won the prize in 1991 and has been held under house arrest for over three years by Rangoon's military junta . NE Example:  NE Example The study was conducted by Dr. Aubrey Milunsky of the Center for Human Genetics at the Boston University School of Medicine and colleagues . the study be conduct by Dr. Aubrey Milunsky of the Center for Human Genetics at the Boston University School of Medicine and colleague . DT NN VBD VBN IN NP NP NP IN DT NP IN NP NP IN DT NP NP NP IN NP CC NNS . Other The study was conducted by Dr. <b_enamex TYPE="PERSON"> Aubrey Milunsky <e_enamex> of the <b_enamex TYPE="ORGANIZATION"> Center for Human Genetics <e_enamex> at the <b_enamex TYPE="ORGANIZATION"> Boston University School of Medicine <e_enamex> and colleagues . Answer Selection:  Answer Selection Lowering the granularity of IR Relevant Docs Window Generation  argmax Ranked Answers i-1 i+1 i weight sentence i Word match Thesaurus match Cluster match Dispersion match Missing words NE match Research Issues:  Machine translation broad domain MT key component for Translingual QA single objective measure Monolingual QA deal with “noisy” input from recognized audio or MT output Information Extraction deal with broad domain and noisy input Topic Detection and Tracking explore more detailed document analysis than just clustering/threading of documents Document understanding/Summarization/NLG Research Issues Speech Recognition (WER):  Speech Recognition (WER) Conclusion:  Conclusion Information access by natural language is important: info to everyone from anywhere! Information is multilingual Machine Translation is critical: Bring the world closer! Play with lots of data, supercomputers Learn many languages and have lots of fun! Slide31:  Demos Slide32:  QA response 1 QA response 2

Add a comment

Related presentations

Related pages

Giovanni Papini – Wikipedia

Giovanni Papini (* 9. Januar 1881 in Florenz; † 8. Juli 1956 ebenda) war ein italienischer Schriftsteller. Anfangs dem Pragmatismus und Futurismus ...
Read more

Kishore Papineni | LinkedIn

View Kishore Papineni's professional profile on LinkedIn. LinkedIn is the world's largest business network, helping professionals like Kishore Papineni ...
Read more

Kishore Papineni | LinkedIn

View Kishore Papineni’s professional profile on LinkedIn. LinkedIn is the world's largest business network, helping professionals like Kishore Papineni ...
Read more

Giovanni Papini - Wikipedia, the free encyclopedia

Giovanni Papini (January 9, 1881 – July 8, 1956) [1] was an Italian journalist, essayist, literary critic, poet, and novelist.
Read more

P.Papini

Abonnieren Sie den kostenlosen P.Papini Shop Newsletter und verpassen Sie keine Neuigkeit oder Aktion mehr aus dem P.Papini Shop.
Read more

Sai Papineni | Facebook

Sai Papineni is on Facebook. Join Facebook to connect with Sai Papineni and others you may know. Facebook gives people the power to share and makes the...
Read more

Suchergebnis auf Amazon.de für: Giovanni Papini: Bücher

Online-Shopping mit großer Auswahl im Bücher Shop.
Read more

Customer Login - Papini

How to buy: Papini is a trade only supplier. online Open an account or visit Your Account here. Phone Speak to us on 01977 608 234. Email If you wish to ...
Read more

Laxman Papineni - Personal Blog

It’s that time of the year when the water cooler moments are buzzing with talks concerning the annual employee ratings and who has emerged the boss’s ...
Read more

Laxman Papineni (@LaxmanPapineni) | Twitter

The latest Tweets from Laxman Papineni (@LaxmanPapineni). Co-founder @appvirality Plug & Play Growth Hacking Toolkit for Mobile Apps. Hyderabad
Read more