1 Introduction

Artificial intelligence technology, especially machine learning and natural language processing, plays an important role in the development of automated writing evaluation (AWE) tools. AWE tools originated in the 1960s, initially used for grading essays in large-scale exams, and expanded to formative assessment in language teaching in the early 21st century. The development of these tools is changing traditional writing teaching, providing more intelligent and personalized instant feedback. However, the effectiveness of AWE error correction feedback is controversial, and some scholars are concerned that it may shift students' attention from writing content to formal attributes, which could have a negative impact on writing behavior. The book "Automatic Written Error Correction Feedback in Research Paper Revision" is a research achievement by Professor Guo Qian's team at Tsinghua University, based on the doctoral English course "Academic Paper Writing and Publishing". It explores the application of automatic error correction technology in the process of research paper revision and is the latest progress in this field.

2 Content Overview

Chapter One discusses the application of Automatic Writing Evaluation System (AWE) in English writing, emphasizing the importance of Automatic Written Error Correction Feedback (AWCF) for academic writing and pointing out the insufficient research on AWE tools in research paper writing. Chapter 2 reviewed relevant literature on the effectiveness, student engagement, and feedback quality of AWE, and found that the AWE system can improve student writing, but its effectiveness is influenced by multiple factors. Chapter 3 provides an overview of the background, design, and methodology of two empirical studies, including a systematic review and user evaluation of Grammarly. Chapter 4 reports on Grammarly's accuracy and recall in research paper revisions, and finds that its labeling and recommendation accuracy is high, but the recall rate is low. Chapter 5 focuses on the impact of AWE on the revision results of research papers, and finds that using Grammarly can improve language accuracy, but has limited help for missing errors. Chapter 6 describes the participation and influencing factors of students when using Grammarly, and finds that students actively participate in behavior, cognition, and emotion, but social participation is not significant. Chapter 7 summarizes the purpose, findings, contributions, and future directions of the research, pointing out the application effects of Grammarly in paper revision and the characteristics of student participation.

3 Brief Reviews

A brief evaluation was conducted on "Automatic Written Error Correction Feedback in Research Paper Revision" from both research methods and AWE application methods. In terms of research methods, the book adopts quantitative and qualitative methods, but does not combine them. The sample size is small and concentrated in the field of engineering, which limits the generalizability of the results. In terms of AWE application methods, the book focuses on the Grammarly system, paying attention to error correction and accuracy of suggestions, but using the free version may result in a lower recall rate. Future research can use a paid version to compare different systems, focus on generative AI tools, and the application of AWE in different genres, error types, and feedback methods. This book fills the gap in AWE's research on advanced academic English writing, but does not explore higher-level issues such as logic, rhetoric, and citation. Future research can combine AWE with manual feedback to fully leverage its advantages.

4 Conclusion

The book explores the application of automatic writing error correction feedback in academic writing, analyzes the impact of AI technology on language learning, and provides insights for teachers, educational technology developers, and researchers. The author evaluated the potential of automatic error correction feedback in improving students' writing ability and quality, and provided a basis for improving the Automatic Written Error Correction (AWE) system. The book points out limitations in research design, sample size, and generalizability of results, and proposes directions for future research. The author provides case studies for the application of AI technology in foreign language education through empirical research, while emphasizing the importance of combining AI technology with human teachers in language teaching. The book also discusses the challenges faced by automatic feedback technology in language teaching, such as inaccurate feedback and excessive reliance on machine feedback that may reduce learners' autonomy. These issues need further exploration to promote the cultivation of student participation and autonomy, and to leverage the complementary advantages of AI and human teachers.