Monali

EAP 507

Summary SF3 Comment by Shannon D Tumanut: You indicated below one place where you chose your own writing over that of AI. Does this hold true for the entire summary? Remember that if you use AI or a combination, there should be appropriate citations.

Cho and Kim (2021) created a revolutionary text interface for mobile English language education applications using deep learning. The primary objective was to design an accessible and intuitive interface suitable for mobile platforms based on touch inputs while eliminating sophisticated graphical user interface (GUI) interactions. Their approach began with the creation of a touch-based text interface that allowed users to generate handwritten inputs right on the mobile screen. They then used a convolutional neural network (CNN) model trained on the Extended Modified National Institute of Standards and Technology (EMNIST) dataset to recognize the handwritten texts and link them to appropriate actions. The ability to analyze handwriting inputs via deep learning for more realistic human-computer interactions is one of the fundamental services. The interface consists of a virtual joystick for movement control, a render texture to display the user’s handwriting in real-time, and a button to perform behaviors based on the detected text. By identifying handwritten words and letters using the CNN, tasks can be executed immediately without the need for additional laborious GUI interfaces. This enables the user to make intuitive decisions and trigger behavior. The author’s requirement is to enable easy yet highly immersive educational interactions on widely available mobile platforms. Their text interface promises to improve student engagement and learning results by allowing users to control behaviors and make selections through handwriting, which is then mapped to actions via deep learning. The author generated prototype English language education applications centered on writing practice to test the interface. A survey experiment revealed that respondents were extremely satisfied with the interface’s ease of use and learning. User testing of the instructional applications revealed an overall pleasant experience and improved attentiveness while utilizing the handwriting interaction method. The author introduced a novel deep learning-powered text interface for mobile platforms that can perform tasks via handwriting detection. This interface was successfully used to create intuitive English teaching programs with little GUI distractions. The text interface attempts to improve immersion and learning while increasing accessibility compared to advanced mobile GUI interactions. The initial user evaluations suggest that users are pleased with the interface and the overall learning experience given by applications built with it. Further work planned involves increasing the capabilities of the teaching applications and undertaking more rigorous evaluations of the educational outcomes offered by this new handwriting-driven interaction strategy. Comment by Monali Chowdary Hanumara: This outlines the study’s goal, which was to develop an easy handwriting-based text interface for mobile education apps that removes challenging GUI operations. I wrote this on my way own because AI didn’t generate a clear response Comment by Monali Chowdary Hanumara: This explains the methodology: creating a text interface for handwriting and training a CNN using EMNIST data to identify handwriting and associate it with actions. Comment by Monali Chowdary Hanumara: This discusses the main features of the text interface itself, such as the ability to analyze handwriting and correlate it with behaviors using deep learning. Thus, it defines the study’s object. Comment by Shannon D Tumanut: I’m not sure that this is the object of study. Could it be more related to exigence? It’s not clear because the sentence itself isn’t clear, but you do need to acknowledge the study’s exigence somewhere. Comment by Shannon D Tumanut: One of the fundamental services of what? Comment by Shannon D Tumanut: What does this mean? Comment by Shannon D Tumanut: This is unclear. 1) There are 2 authors, and 2) what do you mean by ‘requirement’? Comment by Shannon D Tumanut: Again, there are 2 authors. Comment by Shannon D Tumanut: All of this seems to repeat what was written above. Do you need all of it? Some editing could be in order. Comment by Shannon D Tumanut: See previous comments. Comment by Shannon D Tumanut: Immersion in what?

The Extended Modified National Institute of Standards and Technology (EMNIST) dataset served as the basis for the deep learning model that the authors utilized, together with a convolutional neural network (CNN), to identify and deduce handwritten sentences. Additionally, they used the Unity3D engine and the TensorFlow Sharp 1.15.1 plug-in to construct the mobile English language teaching application. Comment by Shannon D Tumanut: Okay, but how is this related to your KC?

The authors continue by going into detail on the experimental procedure, the hardware and software that were employed, the techniques for gathering and analyzing data, and the end outcomes of the trials. They clarify that they carried out several tests, including ones on user satisfaction, inference speed, and recognition accuracy, to assess how well the suggested text interface and interaction performed. Additionally, the authors offer a thorough analysis of the experiment outcomes. They talk about the benefits and drawbacks of the suggested text interface and interactivity and offer suggestions for further development. In general, this section offers a thorough summary of the experimentation procedure and the analysis of the outcomes, which is crucial for comprehending the project’s technical elements and the reliability of the study findings. Comment by Shannon D Tumanut: You should cite the authors here since it’s a new, different paragraph. Comment by Shannon D Tumanut: This paragraph doesn’t explain a key concept, and it doesn’t synthesize two articles. You’ve provided a general summary of the methodological process, but it only applies to Cho & Kim (2021).

Reference: Cho, Y., & Kim, J. (2021). Production of mobile english language teaching application based on text interface using deep learning. 
Electronics, 
10(15), 1809. Comment by Shannon D Tumanut: Check this.