| dc.description.abstract | Due to the rapid development of technology and the popularity of the Internet, e-learning becomes one of the learning mediums for many people today. Compare to traditional education, which is limited by time, location, and teacher resources, e-learning can enhance learning efficiency. We noticed that there has not been much research on these multiple learning platforms that specifically recognize and evaluate primary school students′ handwriting problems, indicating potential for improvement, such as systems that may not automatically recognize stroke order and evaluate the handwriting quality, or systems that may require additional devices to collect handwriting. Currently, the e-learning environment and resources for primary school students are relatively unfriendly. For primary school students, learning to write has a significant impact on the complex cognitive behaviors associated with text production in the future, so learning good handwriting habits is very important for the development of primary school students. Therefore, in this study, we proposed an e-learning system that uses a bidirectional long short-term memory to automatically recognize the correct stroke order and handwriting quality of primary school students′ written English letters. By using a smartpen to write on a special paper, the data is transmitted to the server for evaluation and return of results. In the experimental results, we achieved outstanding accuracy in recognizing the correctness of stroke order. Additionally, our system′s designed scoring criteria for handwriting quality effectively enables students to self-assess their learning results, thus enhancing the efficiency of e-learning. | en_US |