paraphrasing 4 microblog normalization
play

Paraphrasing 4 Microblog Normalization Wang Ling Carnegie Mellon - PowerPoint PPT Presentation

Dec 13, 2023 •25 likes •885 views

Paraphrasing 4 Microblog Normalization Wang Ling Carnegie Mellon University Chris Dyer Instituto Superior Tcnico Alan Black INESC-ID L2F Isabel Trancoso In a nutshell... In a nutshell... Automatically create a Normalization corpora

  1. Paraphrasing 4 Microblog Normalization Wang Ling Carnegie Mellon University Chris Dyer Instituto Superior Técnico Alan Black INESC-ID L2F Isabel Trancoso

  2. In a nutshell...

  3. In a nutshell... Automatically create a Normalization corpora Original Normalization gonna b nutz ! ! going to be nuts!! sunday morn . wrking Sunday morning. Working I went 2 da national army I went to the national army

  4. In a nutshell... Automatically create a Build a Paraphrase Normalization corpora model for Normalization Normalization Model Original Normalization gonna b nutz ! ! going to be nuts!! sunday morn . wrking Sunday morning. Working I went 2 da national army I went to the national army

  5. In a nutshell... Normalization Model Leavin canada goin bac to cali

  6. In a nutshell... Normalization Model Leaving Canada, Leavin canada going back to goin bac to cali California

  7. Why Normalize? msg 4 Warren G his cday is today 1 yr older.

  8. Why Normalize? msg 4 Warren G his cday is today 1 yr older. Bing Translator

  9. Why Normalize? msg 4 Warren G his cday is today 1 yr older. Bing Translator 味精 4 沃 伦 G 他 cday 今天是 较 旧的 1 年。

  10. Why Normalize? msg 4 Warren G his cday is today 1 yr older. Bing Translator monosodium glutamate 4 味精 4 沃 伦 G 他 cday 今天是 较 warren G he cday today is 旧的 1 年。 older 1 year.

  11. Related Work: Lexical Normalization ● msg → message ● 4 → for ● yr → year

  12. Related Work: Lexical Normalization ● Yang and Eisenstein, 2013; Hassan and Menezes, 2013; Liu et al., 2012; Han et al., 2013; Han et al., 2012; Han and Baldwin, 2011; Gouws et al., 2011; Beaufort et al., 2010; Liu et al., 2010; Contractor et al, 2010; Kaufmann, 2010; Cook and Stevenson, 2009; Kobuset al., 2008; Choudhury et al., 2007; Aw et al., 2006... and many many many more

  13. Lexical Normalization ● Finding lexical normalizations ● Is lexical normalization enough???

  14. Is lexical normalization enough??? msg 4 Warren G his cday is today 1 yr older. Bing Translator 味精 4 沃 伦 G 他 cday 今天是 较 旧的 1 年。

  15. Is lexical normalization enough??? ● Lexical normalization message for Warren G his birthday is today 1 year older. Bing Translator

  16. Is lexical normalization enough??? ● Lexical normalization message for Warren G his birthday is today 1 year older. Bing Translator 他的生日今天是大 1 岁 的沃 伦 G His birthday today is 1 year older Warren G message. 消息。

  17. How to make it better??? message for Warren G his birthday is today 1 year older.

  18. How to make it better??? ● Punctuation A message for Warren G , his ● Pro-drop birthday is today , he is 1 year older. ● Missing articles Bing Translator A message to/from Warren 沃 伦 G 的消息，他的生日是今天， G, because his birthday is 他是大 1 岁 。 today, he is one year older.

  19. Previous work ● Beam Search Decoder combining different models (Wang and Ng, 2013) Model Example Pronunciation 2day → today Missing word “Be” I late → I am late Retokenization u.where → u . where Apostrophe im → I’m Abbreviation lol → laughing out loud Time 1130am → 11:30 am

  20. Previous work ● Other possible models/features: Model Example Swapping letters concsience →conscience Letter Repetition gooooood → good Context rly → really Visual g00d → good Measure Units 100k → 100000

  21. Problem

  22. Problem ● I am too lazy

  23. Problem ● I am too lazy ○ Need to look at the data and look for stuff to normalize

  24. Problem ● I am too lazy ○ Need to look at the data and look for stuff to normalize ○ Implement models/features to address each case

  25. Problem ● I am too lazy ○ Need to look at the data and look for stuff to normalize ○ Implement models/features to address each case ○ Repeat again for every new language to normalize

  26. Problem ● I am too lazy ○ Need to look at the data and look for stuff to normalize ○ Implement models/features to address each case ○ Repeat again for every new language to normalize ● How to build a good normalizer without working that hard???

  27. Dream

  28. Dream ● Obtain data with of tweets paired with normalizations Original Normalization gonna b nutz ! ! going to be nuts!! sunday morn . wrking Sunday morning. Working I went 2 da national army I went to the national army

  29. Dream ● Obtain data with of tweets paired with normalizations ● Build a model that learns to normalize based on these examples Data Normalization Model Original Normalization gonna b nutz ! ! going to be nuts!! sunday morn . wrking Sunday morning. Working I went 2 da national army I went to the national army

  30. But how can we get the data? Original Normalization gonna b nutz ! ! going to be nuts!! sunday morn . wrking Sunday morning. Working I went 2 da national army I went to the national army

  31. But how can we get the data? ● Annotate (Wang and Ng, 2013) ○ Not scalable Original Normalization gonna b nutz ! ! going to be nuts!! sunday morn . wrking Sunday morning. Working I went 2 da national army I went to the national army

  32. But how can we get the data? ● Annotate (Wang and Ng, 2013) ○ Not scalable ● Extract Paraphrases in Twitter (Xu et al, 2013) ○ No distinction between original tweet and normalization ○ Works in practice by using a formal language model Original Normalization gonna b nutz ! ! going to be nuts!! sunday morn . wrking Sunday morning. Working I went 2 da national army I went to the national army

  33. Do people normalize text online?

  34. Do people normalize text online? ● Claim: Yes , but they do it in another language .

  35. Translationese ● Standardization during translations (Laviosa, 1998;Volansky et al., 2013) ● No equivalent idiosyncrasies msg 4 Warren G his cday is today 1 yr older. happy cday may god bless u and the... - 发信息给 Warren G , 今天是他的生日,又 老了一岁了。生日快乐,愿上帝保佑 你和 ...

  36. Translationese msg 4 Warren G his cday is today 1 yr older. happy cday may god bless u and the... Bing Translator 味精 4 沃 伦 G 他 cday 今天是 较 旧的 1 年。快 乐 cday 5 月上帝保 佑你和 ...

  37. Translationese msg 4 Warren G his cday is 发信息给 Warren G ,今天是他的 today 1 yr older. happy cday 生日,又 老了一岁了。生日快乐, may god bless u and the... 愿上帝保佑你和 ... Bing Translator 味精 4 沃 伦 G 他 cday 今天是 较 旧的 1 年。快 乐 cday 5 月上帝保 佑你和 ...

  38. Translationese msg 4 Warren G his cday is 发信息给 Warren G ,今天是他的 today 1 yr older. happy cday 生日,又 老了一岁了。生日快乐, may god bless u and the... 愿上帝保佑你和 ... Bing Translator Bing Translator 味精 4 沃 伦 G 他 cday 今天是 较 旧的 1 年。快 乐 cday 5 月上帝保 佑你和 ...

  39. Translationese msg 4 Warren G his cday is 发信息给 Warren G ,今天是他的 today 1 yr older. happy cday 生日,又 老了一岁了。生日快乐, may god bless u and the... 愿上帝保佑你和 ... Bing Translator Bing Translator Send a message to Warren g, 味精 4 沃 伦 G 他 cday 今天是 较 today is his birthday, old and a 旧的 1 年。快 乐 cday 5 月上帝保 one year old. Happy birthday and 佑你和 ... may God bless you and ...

  40. Translationese msg 4 Warren G his cday is today 1 yr older. happy cday may god bless u and the... Send a message to Warren g, today is his birthday, old and a one year old. Happy birthday and may God bless you and ...

  41. Translationese Send a message to Warren g, msg 4 Warren G today is his birthday, his cday is today old and a one year old. 1 yr older. Happy birthday and may God happy cday may god bless u and the... bless you and ...

  42. Translationese abbreviation, punctuation, completion Send a message to Warren g , msg 4 Warren G today is his birthday, his cday is today old and a one year old. 1 yr older. Happy birthday and may God happy cday may god bless u and the... bless you and ...

  43. Translationese abbreviation, punctuation, completion Send a message to Warren g, msg 4 Warren G jargon, punctuation, phrasing today is his birthday, his cday is today old and a one year old. 1 yr older. Happy birthday and may God happy cday may god bless u and the... bless you and ...

  44. Translationese abbreviation, punctuation, completion Send a message to Warren g, msg 4 Warren G jargon, punctuation, phrasing today is his birthday, his cday is today abbreviation, machine translation sucks old and a one year old. 1 yr older. Happy birthday and may God happy cday may god bless u and the... bless you and ...

  45. Translationese abbreviation, punctuation, completion Send a message to Warren g, msg 4 Warren G jargon, punctuation, phrasing today is his birthday, his cday is today abbreviation, machine translation sucks old and a one year old. 1 yr older. jargon, abbreviation, conjunction Happy birthday and may God happy cday may god bless u and the... bless you and ...

  46. Where do these come from?

Recommend

Paraphrasing MS. STRAUSSS EPS CLASS What is paraphrasing? Taking An effective paraphrase

Paraphrasing MS. STRAUSSS EPS CLASS What is paraphrasing? Taking An effective paraphrase

Paraphrasing MS. STRAUSSS EPS CLASS What is paraphrasing? Taking An effective paraphrase keeps the same core meaning someone elses ideas as the original passage. But the meaning is expressed and putting them in in your own personal

278 views • 13 slides
Paraphrasing vs. Plagiarism rev ised : 0 3.0 4.13 | | English 1301: Com position I || D. Glen Sm

Paraphrasing vs. Plagiarism rev ised : 0 3.0 4.13 | | English 1301: Com position I || D. Glen Sm

Paraphrasing vs. Plagiarism rev ised : 0 3.0 4.13 | | English 1301: Com position I || D. Glen Sm ith, instructor 1 Paraphrasing vs. Plagiarism Research papers m ake students nervous. There are m any w ays to unintentionally create im proper

259 views • 12 slides
Automatic extraction of paraphrasing rules: A survey and plans for future work Prodromos

Automatic extraction of paraphrasing rules: A survey and plans for future work Prodromos

Automatic extraction of paraphrasing rules: A survey and plans for future work Prodromos Malakasiotis, Ph.D. student Department of Informatics Athens University of Economics and Business rulller@aueb.gr 1 What is Paraphrasing? X is

497 views • 24 slides
Semantic Parsing via Paraphrasing Mateusz Malinowski Based on: J. Berant and P. Liang

Semantic Parsing via Paraphrasing Mateusz Malinowski Based on: J. Berant and P. Liang

Semantic Parsing via Paraphrasing Mateusz Malinowski Based on: J. Berant and P. Liang Semantic Parsing via Paraphrasing ACL 2014 Outline Abstract view on the semantic parser ! What party did Clay establish? paraphrase model What

200 views • 17 slides
Paraphrasing controlled English texts Kaarel Kaljurand CNL 2009, Marettimo, Italy 2009-06-09

Paraphrasing controlled English texts Kaarel Kaljurand CNL 2009, Marettimo, Italy 2009-06-09

Paraphrasing controlled English texts Kaarel Kaljurand CNL 2009, Marettimo, Italy 2009-06-09 Outline What is a paraphrase? Usage and requirements Paraphrasing ACE by DRS verbalization DRS Core ACE DRS NP ACE

573 views • 25 slides
Normal forms and normalization An example of normalization using normal forms We assume we have

Normal forms and normalization An example of normalization using normal forms We assume we have

Normal forms and normalization An example of normalization using normal forms We assume we have an enterprise that buys products from different supplying companies, and we would like to keep track of our data by means of a database. We would like

205 views • 9 slides
Database Normalization Asst. Prof. Dr. Kanda Runapongsa Saikaew (krunapon@kku.ac.th)

Database Normalization Asst. Prof. Dr. Kanda Runapongsa Saikaew (krunapon@kku.ac.th)

Electricite Du Laos (EDL) Database Normalization Asst. Prof. Dr. Kanda Runapongsa Saikaew (krunapon@kku.ac.th) Department of Computer Engineering Khon Kaen University 1 Overview What and why normalization Background to normalization

526 views • 13 slides
Predic'ng Responses to Microblog Posts Yoav Artzi 1 , Patrick

Predic'ng Responses to Microblog Posts Yoav Artzi 1 , Patrick

Predic'ng Responses to Microblog Posts Yoav Artzi 1 , Patrick Pantel 2 and Michael Gamon 2 1 University of Washington, MicrosoC Research Work conducted at

598 views • 45 slides
Normalization Redundancy causes several anomalies : insert, delete and update

Normalization Redundancy causes several anomalies : insert, delete and update

What and Why Normalization? To remove potential redundancy in design Normalization Redundancy causes several anomalies : insert, delete and update Normalization uses concept of dependencies Functional Dependencies Idea used:

73 views • 5 slides
Normalization-Invariant Fuzzy Logic Need for Normalization Operations Explain Empirical Success

Normalization-Invariant Fuzzy Logic Need for Normalization Operations Explain Empirical Success

Traditional . . . Need for Heavy-Tailed . . . What We Do Normalization-Invariant Fuzzy Logic Need for Normalization Operations Explain Empirical Success of How to Combine Degrees Student Distributions in Describing Deriving Student . . .

322 views • 13 slides
Translations Requiring Paraphrasing A student who studies hard will learn to tango. Mark Criley

Translations Requiring Paraphrasing A student who studies hard will learn to tango. Mark Criley

Translations Requiring Paraphrasing A student who studies hard will learn to tango. Mark Criley tango. 6 Read It Back! But now, lets check our work: Translations Requiring Paraphrasing Mark Criley 5 Substitute Translations 4 Translate

274 views • 11 slides
Normalization by Evaluation for Martin-Lf Type Theory with One Universe Peter Dybjer,

Normalization by Evaluation for Martin-Lf Type Theory with One Universe Peter Dybjer,

Normalization Martin-Lf type theory Normalization algorithm Decidability of equality Normalization by Evaluation for Martin-Lf Type Theory with One Universe Peter Dybjer, Gteborg (with Andreas Abel, Munich, and Klaus Aehlig, Swansea)

616 views • 28 slides
Talk Overview Paraphrases Paraphrasing and Translation What theyre useful for How

Talk Overview Paraphrases Paraphrasing and Translation What theyre useful for How

1 Talk Overview Paraphrases Paraphrasing and Translation What theyre useful for How other people generate them Chris Callison-Burch How we do it 16 March 2006 Applying Paraphrases to Translation Problem of unseen words

264 views • 3 slides
Genealogical Place Name Normalization Bob Leaman (bob.leaman@asu.edu) What is meant by

Genealogical Place Name Normalization Bob Leaman (bob.leaman@asu.edu) What is meant by

Genealogical Place Name Normalization Bob Leaman (bob.leaman@asu.edu) What is meant by Normalization? Enforcing a standardized representation Increases accuracy Data shared over e-mail can be very hard to correct Easier

471 views • 15 slides
A Generate and Rank Approach to Sentence Paraphrasing Prodromos Malakasiotis * Ion Androutsopoulos

A Generate and Rank Approach to Sentence Paraphrasing Prodromos Malakasiotis * Ion Androutsopoulos

A Generate and Rank Approach to Sentence Paraphrasing Prodromos Malakasiotis * Ion Androutsopoulos * * NLP Group, Department of Informatics, Athens University of Economics and Business, Greece Digital Curation Unit IMIS, Research Centre

308 views • 17 slides
Diverse Paraphrasing and its Effectiveness in Data Augmentation Ashutosh Kumar*, Satwik

Diverse Paraphrasing and its Effectiveness in Data Augmentation Ashutosh Kumar*, Satwik

Diverse Paraphrasing and its Effectiveness in Data Augmentation Ashutosh Kumar*, Satwik Bhattamishra*, Manik Bhandari, Partha Talukdar Machine and Language Learning Lab (MALL) Indian Institute of Science, Bangalore *Equal Contributions

905 views • 87 slides
Placing images on the world map: a microblog- based enrichment approach Claudia Hau ff &

Placing images on the world map: a microblog- based enrichment approach Claudia Hau ff &

Placing images on the world map: a microblog- based enrichment approach Claudia Hau ff & Geert-Jan Houben, TU Delft (NL) SIGIR 2012 1 Claudia Hauff, 2012 Problem & Motivation autumn, leaves, red, Latitude/Longitude: 52.4/-3.2 sad,

251 views • 23 slides
A Semi-Supervised Bayesian Network Model for Microblog Topic Classification Yan Chen 1 , 2 Zhoujun

A Semi-Supervised Bayesian Network Model for Microblog Topic Classification Yan Chen 1 , 2 Zhoujun

A Semi-Supervised Bayesian Network Model for Microblog Topic Classification Yan Chen 1 , 2 Zhoujun Li 1 Liqiang Nie 2 Xia Hu 3 Xiangyu Wang 2 Tat-seng Chua 2 Xiaoming Zhang 1 1 State Key Laboratory of Software Development Environment, Beihang

528 views • 49 slides
wPod Weibo Public Opinion (Polarity) Detection Haotian He & Sanae Sato Microblog is

wPod Weibo Public Opinion (Polarity) Detection Haotian He & Sanae Sato Microblog is

wPod Weibo Public Opinion (Polarity) Detection Haotian He & Sanae Sato Microblog is always updating, and new issues and terms come out . There are not many valuable handy labeled data. Co-training is a semi-supervised learning

453 views • 11 slides
Relational Normalization Theory Chapter 6 1 Limitations of E-R Designs Provides a set of

Relational Normalization Theory Chapter 6 1 Limitations of E-R Designs Provides a set of

Relational Normalization Theory Chapter 6 1 Limitations of E-R Designs Provides a set of guidelines, does not result in a unique database schema Does not provide a way of evaluating alternative schemas Normalization theory

610 views • 33 slides
The Velocity of Censorship: High-Fidelity Detection of Microblog Post Deletions Tao Zhu

The Velocity of Censorship: High-Fidelity Detection of Microblog Post Deletions Tao Zhu

The Velocity of Censorship: High-Fidelity Detection of Microblog Post Deletions Tao Zhu Independent Researcher David Phipps Bowdoin College Adam Pridgen Rice University Jedidiah R. Crandall University of New Mexico Dan S. Wallach Rice University

547 views • 28 slides
Segment-Phrase Table for Semantic Segmentation, Visual Entailment and Paraphrasing Hamid

Segment-Phrase Table for Semantic Segmentation, Visual Entailment and Paraphrasing Hamid

Segment-Phrase Table for Semantic Segmentation, Visual Entailment and Paraphrasing Hamid Izadinia, Fereshteh Sadeghi, Santosh K. Divvala, Hannaneh Hajishirzi, Yejin Choi, Ali Farhadi Presentated by Edward Banner Outline What is a SPT?

958 views • 43 slides
Strong normalization for the parameter-free Strong polymorphic lambda calculus based on the

Strong normalization for the parameter-free Strong polymorphic lambda calculus based on the

Strong Normalization Akiyoshi and Terui Introduction Our Results Strong normalization for the parameter-free Strong polymorphic lambda calculus based on the -rule Normalization Theorem Previous Results Buchholz -Rule Ryota

706 views • 35 slides
Formalizing Strong Normalization Proofs Kazuhiko Sakaguchi College of Information Science,

Formalizing Strong Normalization Proofs Kazuhiko Sakaguchi College of Information Science,

Formalizing Strong Normalization Proofs Kazuhiko Sakaguchi College of Information Science, University of Tsukuba 2014/12/3 TPP2014 Strong Normalization Theorem In typed -calculi, strong normalization (SN) theorem is as follows. , t ,

1.59k views • 129 slides

More recommend