Background Background Text Complexity Text Complexity Text - - PowerPoint PPT Presentation

background background
SMART_READER_LITE
LIVE PREVIEW

Background Background Text Complexity Text Complexity Text - - PowerPoint PPT Presentation

On Measures of On Measures of On Measures of Background Background Text Complexity Text Complexity Text Complexity Sowmya V.B., Sowmya V.B., Sowmya V.B., What do we mean by measures of text complexity ? Why would anyone want to do


slide-1
SLIDE 1

On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

On Measures of Text Complexity

Sowmya V.B. Detmar Meurers

Universit¨ at T¨ ubingen

Second T¨ ubingen-Berlin Meeting on Analyzing Learner Language T¨ ubingen, December 5-6, 2011

1 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Background

What do we mean by “measures of text complexity” ?

◮ Measuring how difficult it is to read a text, ◮ given a purpose, e.g.,

◮ general comprehension of key ideas of text ◮ identification of specific information looked for

◮ based on properties of the text using criteria which are

◮ theory-driven (e.g., difficult syntactic constructions) ◮ data-induced (e.g., corpora with graded texts), and a lot ◮ in-between (e.g., derived frequency information for words)

◮ and information about the user (e.g., language ability,

age, working memory)

◮ obtained directly (e.g., questionnaire), or ◮ indirectly (e.g, inferred from nature of a query) 2 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Background

Why would anyone want to do this?

◮ To evaluate the quality of (manually written) texts, e.g.,

◮ for articles, manuals, books to be accessible to the

intended readership

◮ for reading and writing assessment in language teaching

◮ To evaluate the quality of natural language generation

systems (e.g., in text summarization)

◮ To track first and second language acquisition and

language attrition

◮ Analysis of complexity in Kobalt-DaF network ◮ Criterial features for language development (MERLIN) 3 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

A concrete “real-life” challenge

◮ Develop a search engine ranking web-search results

based on complexity.

◮ support a range of complexity features ◮ first prototype of a Language-Aware Search Engine

(Ott & Meurers 2010)

◮ Which measures of complexity should we use? ◮ Which gold-standards can the resulting approach be

evaluated against?

4 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

How do we measure text complexity?

Traditional readability formulas and lexical measures

◮ Clearly different aspects of linguistic complexity play a

role in determining the readability of a text.

◮ But traditional readability formulas use only shallow

quantiative features (e.g., lengths of words, sentences).

◮ (e.g., Flesch 1948; Coleman & Liau 1975; Kincaid et al.

1975; DuBay 2004)

◮ Others are exclusively based on lexical measures,

such as occurrence in specific word lists (Dale & Chall

1948; Chall & Dale 1995; Coxhead 2000; Bauer & Nation 1993).

5 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Some recent CL research on text complexity

◮ Language n-gram models (Collins-Thompson & Callan

2004; Si & Callan 2001)

◮ Machine learning approaches, with several lexical and

syntactic features (Heilman et al. 2007; Petersen &

Ostendorf 2009; Lijun Feng & Elhadad 2010)

◮ What kind of features are relevant here? Insights from

◮ Language Acquisition ◮ Psycholinguistics 6 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Complexity in language acquisition

Automated L1 acquisition measures:

◮ Measures based on identifying specific syntactic patterns:

◮ Index of Productive Syntax (IPSyn, Scarborough 1990;

Sagae et al. 2005)

◮ Developmental Level (D-Level, Rosenberg & Abbeduto

1987; Covington et al. 2006; Lu 2009)

◮ Some others include (cf., Cheung & Kemper 1992):

◮ Developmental Sentence Scoring (DSS) ◮ Directional Complexity (D-Complexity) ◮ Frazier’s node count, Yngve’s depth

Automated Second-Language Acquisition measures:

◮ Lexical richness (Lu 2011b) ◮ Syntactic complexity in second language writing

(Lu 2010, 2011a; Vyatkina 2012)

◮ Measures mostly based on general counts of phrases,

T-units, clauses, . . .

7 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Complexity and psycholinguistics

◮ long and colorful history, cf. Derivational Theory of

Complexity (DTC, Fodor et al. 1974)

◮ meaning: propositional idea density (Kintsch 1974;

Turner & Greene 1977; Brown et al. 2008)

◮ form: complexity in human sentence processing (e.g.,

surprisal Boston et al. 2008, 2011)

◮ discourse: text coherence and cohension (Coh-Metrix

project, McNamara et al. 2002)

◮ Link to cognition also relevant for applications:

◮ Cognitively motivated readability assessment for adults

with intellectual disabilities (Feng et al. 2009; Feng 2010)

◮ Using Syntactic Complexity measures to detecting

cognitive impairment (Roark et al. 2007)

8 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

How do we evaluate complexity measures?

◮ Measures of readability typically are evaluated against

a gold standard classification of graded readers, which are written with the traditional measures in mind.

◮ What can serve as independently motivated gold

standard for evaluating complexity?

◮ Correlating complexity with cognitive measures

◮ online eye tracking measures identifying processing

difficulty in human sentence processing

◮ working memory decrease in language attrition

(Cheung & Kemper 1992)

◮ Analyzing complexity of the language produced at

different times in first language acquisition

9 / 12

slide-2
SLIDE 2

On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

So, what are we currently doing?

◮ Mining complexity research in linguistics, psychology,

and language acquisition to analyze texts on the web.

◮ Building and evaluating different classifier models. ◮ Acquiring “gold-standard corpora” to test the validity of

these feature sets.

10 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References 11 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Thank you for your attention

. . . and references follow too!

12 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

References

Bagga, A. (2000). Analyzing the Reading Comprehension Task. In Bagga et al. (2000), pp. 35–41. URL http://aclweb.org/anthology/W00-0106. Bagga, A., J. Pustejovsky & W. Zadrozny (eds.) (2000). Syntactic and semantic complexity in natural language processing systems : NAACL-ANLP 2000 workshop, April 30, 2000, Seattle, Washington. New Brunswick, NJ: Assoc. for Computational Linguistics. URL http://www.aclweb.org/anthology/W/W00/#0100. Bauer, L. & P . Nation (1993). Word Families. International Journal of Lexicography 6(4), 253–279. Birch, G. (2005). Balancing fluency, accuracy and complexity. In C. Edwards &

  • J. Willis (eds.), Teachers Exploring Tasks in English Language Teaching,

Palgrave Macmillan, pp. 228–239. Blevins, J. P . (1990). Syntactic complexity: evidence for discontinuity and

  • multidomination. Ph.D. thesis, University of of Massachusetts.

Boston, M. F., J. T. Hale, U. Patil, R. Kliegl & S. Vasishth (2008). Parsing costs as predictors of reading difficulty: An evaluation using the Potsdam Sentence

  • Corpus. Journal of Eye Movement Research 2(1), 1–12. URL

http://www.jemr.org/online/2/1/1. Boston, M. F., J. T. Hale, S. Vasishth & R. Kliegl (2011). Parallel processing and sentence comprehension difficulty. Language and Cognitive Processes 26(3), 301–349. Brown, C., T. Snodgrass, S. J. Kemper, R. Herman & M. A. Covington (2008). Automatic measurement of propositional idea density from part-of-speech

  • tagging. Behavior Research Methods 40(2), 540–545.

12 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Chall, J. S. & E. Dale (1995). Readability Revisted: The New Dale-Chall Readability Formula. Brookline Books. Chen, M. & K. Zechner (2011). Computing and Evaluating Syntactic Complexity Features for Automated Scoring of Spontaneous Non-native speech. In Proceedings of the 49th Annual Meeting of the Association for Computational

  • Linguistics. ACL, Portland, Oregon: ACL, pp. 722–731. URL

http://aclweb.org/anthology/P/P11/P11-1073.pdf. Cheung, H. & S. Kemper (1992). Competing complexity metrics and adults’ production of complex sentences. Applied Psycholinguistics 13(01), 53–76. URL http://dx.doi.org/10.1017/S0142716400005427. Coleman, M. & T. Liau (1975). A Computer Readability Formula Designed for Machine Scoring. Journal of Applied Psychology 60, 283–284. Collins-Thompson, K. & J. Callan (2004). A language modeling approach to predicting reading difficulty. In Proceedings of HLT/NAACL 2004. Boston, USA. URL http://www.cs.cmu.edu/∼callan/Papers/hlt04-kct.pdf. Collins-Thompson, K. & J. Callan (2005). Predicting reading difficulty with statistical language models. Journal of the American Society for Information Science and Technology 56(13), 1448–1462. Covington, M. A., C. He, C. Brown, L. Nac ¸i & J. Brown (2006). How complex is that sentence? A proposed revision of the Rosenberg and Abbeduto D-Level Scale. Computer Analysis of Speech for Psychological Research (CASPR) Research Report 2006-01, The University of Georgia, Artificial Intelligence Center, Athens, GA. URL http://www.ai.uga.edu/caspr/2006-01-Covington.pdf. Coxhead, A. (2000). A New Academic Word List. Teachers of English to Speakers

  • f Other Languages 34(2), 213–238. URL http://www.ingentaconnect.com/

content/tesol/tq/2000/00000034/00000002/art00002?token= 004112cb2e405847447b492b2f5f737b6f2c2b67217d3375686f236efca0b0d66.

12 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Dale, E. & J. S. Chall (1948). A Formula for Predicting Readability. Educational research bulletin; organ of the College of Education 27(1), 11–28. DuBay, W. H. (2004). The Principles of Readability. Costa Mesa, California: Impact Information. URL http://www.impact-information.com/impactinfo/readability02.pdf. Evetts, J. & M. Gauthier (2005). Literacy Task Assessment Guide. Canada: National Literacy Secretariat. Feng, L. (2010). Automatic Readability Assessment. Ph.D. thesis, CUNY. Feng, L., N. Elhadad & M. Huenerfauth (2009). Cognitively Motivated Features for Readability Assessment. In Proceedings of the 12th Conference of the European Chapter of the ACL (EACL 2009). Athens, Greece: Association for Computational Linguistics, pp. 229–237. URL http://www.aclweb.org/anthology/E09-1027. Flank, S. (2000). Sentences vs. Phrases: Syntactic Complexity in Multimedia Information Retrieval. In Bagga et al. (2000), pp. 1–5. URL http://aclweb.org/anthology/W00-0101. Flesch, R. F. (1948). A New Readability Yardstick. Journal of Applied Psychology 32(3), 221–233. Fodor, J. A., T. G. Bever & M. Garrett (1974). The psychology of language. New York: McGraw-Hill. Heilman, M., K. Collins-Thompson, J. Callan & M. Eskenazi (2007). Combining Lexical and Grammatical Features to Improve Readability Measures for First and Second Language Texts. In Human Language Technologies 2007: The Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL ’07). Rochester, New York: Association for Computational Linguistics, pp. 460–467. URL http://aclweb.org/anthology/N07-1058.

12 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Heilman, M., K. Collins-Thompson & M. Eskenazi (2008). An Analysis of Statistical Models and Features for Reading Difficulty Prediction. In Proceedings of the 3rd Workshop on Innovative Use of NLP for Building Educational Applications. Columbus, Ohio. URL http://aclweb.org/anthology/W08-0909. Kemper, S., K. Rice & Y.-J. Chen (1995). Complexity metrics and growth curves for measuring grammatical development from five to ten. First Language 15, 151–166. URL http://fla.sagepub.com/cgi/reprint/15/44/151.pdf. Kincaid, J. P ., R. P . J. Fishburne, R. L. Rogers & B. S. Chissom (1975). Derivation

  • f new readability formulas (Automated Readability Index, Fog Count and

Flesch Reading Ease Formula) for Navy enlisted personnel. Research Branch Report 8-75, Naval Technical Training Command, Millington, TN. Kintsch, W. (1974). Toward a Model of Text Comprehension and Productions. Erlbaum. Lijun Feng, Martin Jansche, M. H. & N. Elhadad (2010). A Comparison of Features for Automatic Readability Assessment. In In Proceedings of the 23rd International Conference on Computational Linguistics (COLING 2010), Beijing, China.. URL http://www.aclweb.org/anthology/C10-2032. Lu, X. (2009). Automatic measurement of syntactic complexity in child language

  • acquisition. International Journal of Corpus Linguistics 14(1), 3–28.

Lu, X. (2010). Automatic analysis of syntactic complexity in second language

  • writing. International Journal of Corpus Linguistics 15(4), 474–496.

Lu, X. (2011a). A Corpus-Based Evaluation of Syntactic Complexity Measures as Indices of College-Level ESL Writers’ Language Development. TESOL Quarterly 45(1), 36–62. Lu, X. (2011b). The Relationship of Lexical Richness to the Quality of ESL Learners’ Oral Narratives. The Modern Languages Journal In press.

12 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

McNamara, D. S., M. M. Louwerse & A. C. Graesser (2002). Coh-Metrix: Automated Cohesion and Coherence Scores to Predict Text Readability and Facilitate Comprehension. Proposal of Project funded by the Office of Educational Research and Improvement, Reading Program. URL http://cohmetrix.memphis.edu/cohmetrixpr/archive/Coh-MetrixGrant.pdf. Monkhouse, K. M. (1972). Measures of Syntactic Complexity and Their Relationship to Psychological Scale Values of Systematically Varied

  • Sentences. Ph.D. thesis, The University of Iowa, Speech Pathology.

Nezhad, G. R. H. P . (2001). Reading complexity judgments: Episode 1. Shiken: JALT Testing & Evaluation SIG Newsletter 5(3), 2–5. URL http://www.jalt.org/test/haj 1.htm. Nezhad, G. R. H. P . (2002a). Reading complexity judgments: Episode 2. Shiken: JALT Testing & Evaluation SIG Newsletter 6(1), 7–13. URL http://www.jalt.org/test/haj 2.htm. Nezhad, G. R. H. P . (2002b). Reading complexity judgments: Episode 3. Shiken: JALT Testing & Evaluation SIG Newsletter 6(2), 8–11. URL http://www.jalt.org/test/haj 3.htm. Ott, N. & D. Meurers (2010). Information Retrieval for Education: Making Search Engines Language Aware. Themes in Science and Technology Education. Special issue on computer-aided language analysis, teaching and learning: Approaches, perspectives and applications 3(1–2), 9–30. URL http://purl.org/dm/papers/ott-meurers-10.html. Petersen, S. E. (2007). Natural Language Processing Tools for Reading Level Assessment and Text Simplification for Bilingual Education. Ph.D. thesis, University of Washington. URL http://sarahpetersen.net/sarah petersen dissertation.pdf.

12 / 12 On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Petersen, S. E. & M. Ostendorf (2006a). Assessing the Reading Level of Web

  • Pages. In Ninth International Conference on Spoken Language Processing

(Interspeech-ICSLP). Pittsburgh, Pennsylvania. URL http://sarahpetersen.net/portfolio/petersen is2006.pdf. Petersen, S. E. & M. Ostendorf (2006b). Assessing the Reading Level of Web Pages (Poster). In Proceedings of Interspeech 2006. Petersen, S. E. & M. Ostendorf (2007). Text Simplification for Language Learners: A Corpus Analysis. In Speech and Language Technology for Education (SLaTE). URL http://sarahpetersen.net/portfolio/Petersen Ostendorf SLaTE2007 final.pdf. Petersen, S. E. & M. Ostendorf (2009). A machine learning approach to reading level assessment. Computer Speech and Language 23, 86–106. Pollard, S. & A. W. Biermann (2000). A Measure of Semantic Complexity for Natural Language Systems. In Bagga et al. (2000), pp. 42–46. URL http://aclweb.org/anthology/W00-0107. Roark, B., M. Mitchell & K. Hollingshead (2007). Syntactic complexity measures for detecting mild cognitive impairment. In Proceedings of the Workshop on BioNLP 2007: Biological, Translational, and Clinical Language Processing. Stroudsburg, PA, USA: Association for Computational Linguistics, BioNLP ’07,

  • pp. 1–8. URL http://dl.acm.org/citation.cfm?id=1572392.1572394.

Rosenberg, S. & L. Abbeduto (1987). Indicators of linguistic competence in the peer group conversational behavior of mildly retarded adults. Applied Psycholinguistics 8, 19–32. URL http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid= 2607480&fulltextType=RA&fileId=S0142716400000047.

12 / 12

slide-3
SLIDE 3

On Measures of Text Complexity

Sowmya V.B., Detmar Meurers

Background

What is it? Why is it relevant?

Real-life challenge Some measures

Traditional formulas and lexical measures Some recent CL approaches Language acquisition Psycholinguistics

How to evaluate Work in progress References

Sagae, K., A. Lavie & B. MacWhinney (2005). Automatic measurement of syntactic development in child language. In Proceedings of the 42nd Meeting of the Association for Computational Linguistics (ACL ’05). Ann Arbor, MI. URL http://aclweb.org/anthology/P05-1025. Scarborough, H. S. (1990). Index of Productive Syntax. Applied Psycholinguistics 11(1), 1–22. URL http://journals.cambridge.org/action/displayAbstract? fromPage=online&aid=2747016. Schwarm, S. & M. Ostendorf (2005). Reading Level Assessment Using Support Vector Machines and Statistical Language Models. In Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL ’05). Ann Arbor, Michigan, pp. 523–530. URL http://aclweb.org/anthology/P05-1065. Si, L. & J. Callan (2001). A Statistical Model for Scientific Readability. In Proceedings of the 10th International Conference on Information and Knowledge Management (CIKM). ACM, pp. 574–576. Poster description. Tarone, E., M. Bigelow & B. Swierzbin (2006). Impact of literacy level on features of interlanguage in oral narratives. Rivista di psicolinguistica applicata 6, 65–77. Turner, A. & E. Greene (1977). The construction and use of a propositional text

  • base. Tech. Rep. 63, University of Colorado, Institute for the Study of

Intellectual Behavior, Boulder. URL http://ics.colorado.edu/techpubs/pdf/77-63.pdf. Vyatkina, N. (2012). The development of second language writing complexity in groups and individuals: A longitudinal learner corpus study. The Modern Language Journal To appear.

12 / 12