Automatic identification of discourse moves in scientific article introductions

Pendar, Nick; Cotos, Elena

doi:10.3115/1631836.1631844

Cited by 21 publications

(22 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The F1 score, or the harmonic mean of precision and recall, measures the overall performance of the system for a category (Van Rijsbergen, 1979) and is calculated as: Table 4 shows that the move classifier predicted Move 1 and Move 3 with higher precision than Move 2. This result is in agreement with our earlier experimentation where we found that Move 2 is most difficult to identify and that it tends to be misclassified as Move 1 (Pendar & Cotos, 2008). This is not surprising since this time the training data for Move 2 was also considerably sparser than the data for the other two moves (6,039 sentences for Move 1; 1,609 for Move 2; and 2,352 for Move 3).…”

Section: Evaluation and Discussionsupporting

confidence: 82%

“…The main features used for the identification of moves and steps were sets of word unigrams and trigrams (i.e., single words and three word sequences) from the annotated corpus. In our earlier work, we found that bigrams (two word sequences) had a negative effect on the classifier (Pendar & Cotos, 2008), which is why we did not experiment with the bigrams here. We also found that our extracted features were not discipline-dependent.…”

Section: Movementioning

confidence: 99%

“…Previously, we approached a very similar genre classification task when we developed a feedback application called IADE (Pendar & Cotos, 2008). Both the Mover and IADE used a classification schema based on Swales' (1990) Create a Research Space (CARS) framework for Introduction sections.…”

Section: Automated Categorization Of Discourse and Genrementioning

confidence: 99%

See 2 more Smart Citations

Discourse Classification into Rhetorical Functions for AWE Feedback

Cotos

Pendar

2016

CALICO Journal

Self Cite

View full text Add to dashboard Cite

This paper reports on the development of the analysis engine for the Research Writing Tutor (RWT), an AWE program designed to provide genre and discipline-specific feedback on the functional units of research article discourse. Unlike traditional NLP-based applications that categorize complete documents, RWT's analyzer categorizes every sentence in the text as both a communicative move and a rhetorical step. We describe the construction of a cascade of two support vector machine classifiers trained on a multi-disciplinary corpus of annotated Introduction texts. This work not only demonstrates the usefulness of NLP for automated genre analysis, but also paves the road for future AWE endeavors and forms of automated feedback that could facilitate construction of functional meaning in writing.

show abstract

Section: Evaluation and Discussionsupporting

confidence: 82%

Section: Movementioning

confidence: 99%

See 1 more Smart Citation

Discourse Classification into Rhetorical Functions for AWE Feedback

Cotos

Pendar

2016

CALICO Journal

Self Cite

View full text Add to dashboard Cite

show abstract

“…2 Then, with the help of preprogrammed scripts, percentages for the move distribution in the student's draft are automatically calculated and compared with the distribution of moves in the corpus of his/her academic field (see Pendar & Cotos, 2008). The classification into moves and the information about the distribution of moves, both in the student draft and in the corpus, are the sources of two forms of feedback-color-coded and numerical (see sample feedback in Appendix C).…”

Section: Intelligent Academic Discourse Evaluator (Iade)mentioning

confidence: 99%

Potential of Automated Writing Evaluation Feedback

Cotos¹

2011

CALICO Journal

Self Cite

View full text Add to dashboard Cite

This paper presents an empirical evaluation of automated writing evaluation (AWE) feedback used for L2 academic writing teaching and learning. It introduces the Intelligent Academic Discourse Evaluator (IADE), a new web-based AWE program that analyzes the introduction section to research articles and generates immediate, individualized, and discipline-specific feedback. The purpose of the study was to investigate the potential of IADE's feedback. A mixed-methods approach with a concurrent transformative strategy was employed. Quantitative data consisted of responses to Likert-scale, yes/no, and open-ended survey questions; automated and human scores for first and final drafts; and pre-/posttest scores. Qualitative data contained students' first and final drafts as well as transcripts of think-aloud protocols and Camtasia computer screen recordings, observations, and semistructured interviews. The findings indicate that IADE's colorcoded and numerical feedback possesses potential for facilitating language learning, a claim supported by evidence of focus on discourse form, noticing of negative evidence, improved rhetorical quality of writing, and increased learning gains.

show abstract

“…Using the annotated corpus data and the statistical properties of n-grams in our corpus, probabilistic language models for predicting the occurrence of moves and steps were built to generate rhetorical feedback (Babu, 2013;Cotos, Gilbert, & Sinapov, 2014). RWT's analysis engine approaches the identification of these discourse units as a supervised classification problem (see Burstein, Marcu, & Knight, 2003;Pendar & Cotos, 2008), where each sentence in a text is considered an independent unit of analysis to be classified into two categories -one corresponding to a move and the second corresponding to a step within the identified move.…”

Section: Computational Operationalization For Pedagogical Usementioning

confidence: 99%

Furthering and applying move/step constructs: Technology-driven marshalling of Swalesian genre theory for EAP pedagogy

Cotos

Huffman

Link

2015

Journal of English for Academic Purposes

Self Cite

View full text Add to dashboard Cite

John Swales' seminal work has inspired a wealth of research with important pedagogical implications for genre-based writing instruction. Continuing the prolific move analysis tradition in EAP research, this article presents empirically devised and validated cross-disciplinary IMRD move/step frameworks for the research article genre and demonstrates how Swales' move and step concepts underlying these frameworks formed the foundation of innovative genre-based automated writing evaluation technology. Overall, this paper makes the relationships between genre theory, genre analysis, and genre instruction explicit, demonstrating that move analysis is a powerful and promising theoretical, analytic, and teaching construct. With that, we take Swales' vision to a new dimension of conceptualizing EAP. Keywordsgenre analysis, move, step, genre-based writing pedagogy, automated writing evaluation, research writing Disciplines Curriculum and Instruction | Curriculum and Social Inquiry | Educational Assessment, Evaluation, and Research | Educational MethodsComments NOTICE: this is the author's version of a work that was accepted for publication in Journal of English for Academic Purposes. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Journal of English for Academic Purposes, [19, (2015) AbstractJohn Swales' seminal work has inspired a wealth of research with important pedagogical implications for genre-based writing instruction. Continuing the prolific move analysis tradition in EAP research, this article presents empirically devised and validated cross-disciplinary IMRD move/step frameworks for the research article genre and demonstrates how Swales' move and step concepts underlying these frameworks formed the foundation of innovative genre-based automated writing evaluation technology. Overall, this paper makes the relationships between genre theory, genre analysis, and genre instruction explicit, demonstrating that move analysis is a powerful and promising theoretical, analytic, and teaching construct. With that, we take Swales' vision to a new dimension of conceptualizing EAP.

show abstract

Automatic identification of discourse moves in scientific article introductions

Cited by 21 publications

References 29 publications

Discourse Classification into Rhetorical Functions for AWE Feedback

Discourse Classification into Rhetorical Functions for AWE Feedback

Potential of Automated Writing Evaluation Feedback

Furthering and applying move/step constructs: Technology-driven marshalling of Swalesian genre theory for EAP pedagogy

Contact Info

Product

Resources

About