Article Preview
TopIntroduction
Big data technology and deep learning algorithm have been widely used in industry and economy, this study aims to explore the application of big data technology and deep learning algorithm in English writing scoring system (Wang & Alexander, 2016). A subjective test is an effective way to measure students’ language proficiency in English learning (Gultom, 2016). At present, it has been used in various kinds of English examinations, such as Chinese CET-4 and CET-6, and international examinations such as IELTS and TOEFL. The subjective question is to allow the examinee to answer questions on his own to express his understanding of the test questions, which has a unique function in examining the candidate's language expression ability, thinking innovation ability, and so on (Wang, 2009). In English learning, as a test of candidate's language expression ability, essays are especially representative (Weir, 2005). Correction of essay is not only a heavy workload but also easy to be influenced by subjective factors of markers. Therefore, how to effectively and fairly correct the essay is a big problem facing the education industry. Consuming a lot of manpower and material resources, different scoring standards will cause errors, and high requirements for professionals are the three major shortcomings of traditional human judgment. And using artificial intelligence technology, in combination with Internet technology, communication technology, 5G technology and other popular technologies (Zhang et al., 2022), the system developed can automatedally correct it, which is a great progress compared to traditional methods.
Artificial intelligence technology is a technology developed based on the combination of computer science, information transmission, psychology, and philosophy (Lv et al., 2020). The research and development of the Automated Essay Scoring (AES) method is the concrete embodiment of this trend (Balfour, 2013; Taghipour & Ng, 2016; Uto et al., 2020). The system is to use the computer to score the essay; it can overcome the shortcomings of many people's work paper scoring methods. First, reliability, many studies have shown that the computer scoring system works well. Second, validity, the validity of the scorer is the most influential link in the essay scoring, different marking teachers have certain differences in the scoring of the same essay. However, the use of computer scoring system is affected by human factors (Attali, 2013). Finally, economical, large-scale examinations such as the Chinese college entrance examination, CET-4 and CET-6 need to organize a large number of reviewers to correct the test papers and need personnel guarantee and suitable place, which are based on a lot of resources (Zheng & Cheng, 2008). The automated essay scoring system researched and developed on the basis of artificial intelligence is not only fast, but also saves a lot of manpower and material resources. In addition, the automated essay correction system can also record each candidate's words, grammar and other errors, providing scientific guidance data for teachers' teaching and students' progress (Dikli, 2006). One of the research fields in machine learning is called deep learning. It can obtain information in machine learning and find out the laws and appearances of data or information, such as images, texts, sounds and other data for interpretation analysis. The deep learning discrimination training algorithm differentiates the acquired learning on the basis of machine learning and conducts training analysis for different types of data (Shinde & Shah, 2018).