Unsupervised Concept-to-text Generation with Hypergraphs Ioannis - PowerPoint PPT Presentation

Unsupervised Concept-to-text Generation with Hypergraphs Ioannis Konstas, Mirella Lapata Institute for Language, Cognition and Computation University of Edinburgh NAACL 2012, Montréal Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 1 / 25

Introduction Introduction Concept-to-text generation refers to the task of automatically producing textual output from nonlinguistic input (Reiter and Dale, 2000) Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 2 / 25

Introduction Introduction Concept-to-text generation refers to the task of automatically producing textual output from nonlinguistic input (Reiter and Dale, 2000) Temperature Cloud Sky Cover Time Min Mean Max Time Percent (%) 06:00-21:00 9 15 21 06:00-09:00 25-50 09:00-12:00 50-75 Wind Speed Wind Direction Time Mode Time Min Mean Max 06:00-21:00 S 06:00-21:00 15 20 30 Cloudy, with a low around 10. South wind between 15 and 30 mph. Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 2 / 25

Introduction Introduction Concept-to-text generation refers to the task of automatically producing textual output from nonlinguistic input (Reiter and Dale, 2000) Flight Day direction from to day dep/ar/ret oneway edinburgh montreal saturday departure Search of type what fare argmin flight Show me the cheapest one way flights from Edinburgh to Montreal leaving on Saturday Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 2 / 25

Introduction Traditional NLG Pipeline Communicative Goal Input Data Content Selection Surface Realisation Text Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 3 / 25

Introduction Traditional NLG Pipeline Communicative Goal Input Data Content Selection Surface Realisation Text Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 4 / 25

Introduction Our Approach Temperature Cloud Sky Cover Cloudy, with a low around 10. South wind between 15 and 30 Time Min Mean Max Time Percent (%) mph. 06:00-21:00 9 15 21 06:00-09:00 25-50 09:00-12:00 50-75 Partly cloudy, with a low Wind Speed Wind Direction around 9. Time Mode Time Min Mean Max Breezy, with a south wind be- 06:00-21:00 S 06:00-21:00 15 20 30 tween 15 and 30 mph. Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 5 / 25

Introduction Our Approach Temperature Cloud Sky Cover Cloudy, with a low around 10. South wind between 15 and 30 Time Min Mean Max Time Percent (%) mph. 06:00-21:00 9 15 21 06:00-09:00 25-50 09:00-12:00 50-75 Partly cloudy, with a low Wind Speed Wind Direction around 9. Time Mode Time Min Mean Max Breezy, with a south wind be- 06:00-21:00 S 06:00-21:00 15 20 30 tween 15 and 30 mph. Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 5 / 25

Introduction Our Approach Temperature Cloud Sky Cover Cloudy, with a low around 10. South wind between 15 and 30 Time Min Mean Max Time Percent (%) mph. 06:00-21:00 9 15 21 06:00-09:00 25-50 09:00-12:00 50-75 Partly cloudy, with a low Wind Speed Wind Direction around 9. Time Mode Time Min Mean Max Breezy, with a south wind be- 06:00-21:00 S 06:00-21:00 15 20 30 tween 15 and 30 mph. Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 5 / 25

Introduction Our Approach Temperature Cloud Sky Cover Cloudy, with a low around 10. South wind between 15 and 30 Time Min Mean Max Time Percent (%) mph. 06:00-21:00 9 15 21 06:00-09:00 25-50 09:00-12:00 50-75 Partly cloudy, with a low Wind Speed Wind Direction around 9. Time Mode Time Min Mean Max Breezy, with a south wind be- 06:00-21:00 S 06:00-21:00 15 20 30 tween 15 and 30 mph. Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 5 / 25

Introduction Our Approach Temperature Cloud Sky Cover Cloudy, with a low around 10. South wind between 15 and 30 Time Min Mean Max Time Percent (%) mph. 06:00-21:00 9 15 21 06:00-09:00 25-50 09:00-12:00 50-75 Partly cloudy, with a low Wind Speed Wind Direction around 9. Time Mode Time Min Mean Max Breezy, with a south wind be- 06:00-21:00 S 06:00-21:00 15 20 30 tween 15 and 30 mph. Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 5 / 25

Introduction Our Approach Temperature Cloud Sky Cover Cloudy, with a low around 10. South wind between 15 and 30 Time Min Mean Max Time Percent (%) mph. 06:00-21:00 9 15 21 06:00-09:00 25-50 09:00-12:00 50-75 Partly cloudy, with a low Wind Speed Wind Direction around 9. Time Mode Time Min Mean Max Breezy, with a south wind be- 06:00-21:00 S 06:00-21:00 15 20 30 tween 15 and 30 mph. Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 5 / 25

Introduction Our Approach Temperature Cloud Sky Cover Cloudy, with a low around 10. South wind between 15 and 30 Time Min Mean Max Time Percent (%) mph. 06:00-21:00 9 15 21 06:00-09:00 25-50 09:00-12:00 50-75 Partly cloudy, with a low Wind Speed Wind Direction around 9. Time Mode Time Min Mean Max Breezy, with a south wind be- 06:00-21:00 15 20 30 06:00-21:00 S tween 15 and 30 mph. Key idea: recast generation as a parsing problem 1 Describe the structure of the input with a PCFG 2 Convert PCFG to a hypergraph 3 Goal: Find the most fluent and grammatical derivation Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 5 / 25

Introduction Related Work Angeli et al., 2010 Unified content selection and surface realisation Obtain alignments from Liang et al. (2009) Sequence of discriminative (log-linear) local decisions (records - fields - templates) Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 6 / 25

Introduction Related Work Angeli et al., 2010 Unified content selection and surface realisation Obtain alignments from Liang et al. (2009) Sequence of discriminative (log-linear) local decisions (records - fields - templates) Our Approach Unsupervised generative model Joint content selection and surface realisation, breaks the traditional NLG pipeline Domain independent, trainable end-to-end system Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 6 / 25

Introduction Input Cloud Sky Cover Input: database records d Output: words w corresponding Time Percent (%) to some records of d 06:00-09:00 25-50 09:00-12:00 50-75 Each record r ∈ d has a type r . t and fields f Fields have values f . v and mostly cloudy, types f . t (integer, categorical) Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 7 / 25

Introduction Input Cloud Sky Cover Input: database records d Output: words w corresponding Time Percent (%) to some records of d 06:00-09:00 25-50 09:00-12:00 50-75 Each record r ∈ d has a type r . t and fields f Fields have values f . v and mostly cloudy, types f . t (integer, categorical) Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 7 / 25

Introduction Input Cloud Sky Cover Input: database records d Output: words w corresponding Time Percent (%) to some records of d 06:00-09:00 25-50 09:00-12:00 50-75 Each record r ∈ d has a type r . t and fields f Fields have values f . v and mostly cloudy, types f . t (integer, categorical) Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 7 / 25

Introduction Input Cloud Sky Cover Input: database records d Output: words w corresponding Time Percent (%) to some records of d 06:00-09:00 25-50 09:00-12:00 50-75 Each record r ∈ d has a type r . t and fields f Fields have values f . v and mostly cloudy, types f . t (integer, categorical) Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 7 / 25

Introduction Input Cloud Sky Cover Input: database records d Output: words w corresponding Time Percent (%) to some records of d 06:00-09:00 25-50 09:00-12:00 50-75 Each record r ∈ d has a type r . t and fields f Fields have values f . v and mostly cloudy, types f . t (integer, categorical) Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 7 / 25

Introduction Grammar Definition 1 S → R ( start ) 2 R ( r i . t ) → FS ( r j , start ) R ( r j . t ) 3 R ( r i . t ) → FS ( r j , start ) 4 FS ( r , r . f i ) → F ( r , r . f j ) FS ( r , r . f j ) 5 FS ( r , r . f i ) → F ( r , r . f j ) 6 F ( r , r . f ) → W ( r , r . f ) F ( r , r . f ) 7 F ( r , r . f ) → W ( r , r . f ) 8 W ( r , r . f ) → α 9 W ( r , r . f ) → g( f . v ) Konstas, Lapata (ILCC) Unsupervised Concept-to-text Generation NAACL 2012, Montréal 8 / 25