OrganicPad: A Tablet PC-Based Interactivity Tool for Organic Chemistry

该项目将开发和测试基于平板电脑的教学和评估系统，该系统使用自然的用户界面，让学生学习绘制化学结构。其他结构绘图程序通常使用繁琐的工具或基于拖放的系统，无法模仿学生学习绘制化学结构的正常方式。该软件工具名为OrganicPad，使教师能够通过发送问题来解决问题，从而使学生参与课堂。学生们在软件提供的提示和提示的指导下开发问题的答案。 所有的通信都是通过网络，使用HTTP通信协议。在任何时候，教师都可以评估学生提交的内容，并快速了解学生遇到的困难。或者，该系统可以用于教程模式。学生的每个输入（原子，键，电子，电荷等）将被存储在数据库中，以供以后进行数据分析。使用与以前工作类似的分析系统，他们将使用神经网络来识别类似的学生解决问题的方法，并将其聚类为策略。随着时间的推移，学生的进步将通过使用隐马尔可夫模型来开发学生如何实现构建化学结构的能力的模型。一旦我们有了这些模型，我们将能够针对学生在学习过程中可能遇到的特定问题制定干预措施。这个项目的目标是a）将OrganicPad开发成一个功能齐全的软件包，既可以作为评估，也可以作为指导，帮助学生学习绘制化学结构; B）建立数据模型（笔画，输入顺序，时间线），并确定学生使用的成功和不成功的策略;以及c）使用建模数据的结果来开发干预措施，以提高学生的结构绘图策略。智力优点本提案中概述的工作将为学生如何培养构建结构表征的能力提供新的见解。抽象概念。以前，我们已经开发和使用类似的方法来探索学生的问题解决，涉及复杂的基于Web的问题。然而，这些问题给学生带来了结构性的影响，因为他们的行为只涉及做出选择，要么通过点击行动项目，要么通过拖动图标。我们现在建议将这一过程扩展到使用平板电脑允许的自由形式输入，这对学生施加了更少的结构。我们期望通过这项工作所获得的见解，使我们能够开发出更有效的教学材料，适用于更广泛的应用。更广泛的影响随着平板电脑在教室中的使用越来越多，能够利用其灵活性和功能的教学方法和材料将变得越来越重要。该项目将在整个项目过程中影响成千上万的学生，并使他们能够参与新的教学方法和材料的开发。该提案侧重于绘制和表示不同的化学结构，但我们提出的方法最终适用于许多其他领域，如数学方程，几何图形和计算机科学数据结构。

This project will develop and test a tablet PC based instructional and assessment system that uses the natural user interface to allow students to learn to draw chemical structures. Other structure drawing programs often use cumbersome tools or drag and drop based systems that do not mimic the normal way students learn to develop draw chemical structures.The software tool, called OrganicPad, enables an instructor to engage her students in class by sending them problems to solve. The students develop answers to the problems guided by tips and hints provided by the software. All communication is through the web, using the HTTP communication protocol. At any time, the instructor can evaluate student submissions and can quickly understand where students are having difficulty. Alternatively the system may be used in a tutorial mode.Each input by the student (atom, bond, electron, charge etc.) will be stored in a database for later data analysis. Using a similar analysis system to previous work, they will usel use neural nets to identify similar student approaches to the problem and cluster them into strategies. Student progress will be followed over time by using Hidden Markov Modeling to develop models for how students achieve competence in constructing chemical structures. Once we have these models we will be able to develop interventions specifically aimed at the particular problem a student may have with the process. The goals of this project are to a) develop OrganicPad into a fully functional software package that can be used as both an assessment and a tutorial to help students learn to draw chemical structures; b) model the data (ink strokes, sequence of inputs, timelines) and identify both successful and unsuccessful strategies that students use; and c) To use the results from the modeling data to develop interventions that will improve student structure drawing strategies.Intellectual Merit The work outlined in this proposal will give a new insight into how students develop competence in constructing representations of abstract concepts. Previously we have developed and used similar methods for probing student problem solving, that involved complex web based problems. However, these problems impose structure on students, since their actions only involve making choices, either by clicking on action items, or by dragging icons. We now propose to extend this process to the free form input allowed by using tablet PCs, which impose much less structure on the student. We anticipate that the insights gained by this work will allow us to develop more effective teaching and learning materials for a wide range of applications.Broader Impacts As the use of tablet PCs increases in classrooms, teaching methods and materials that are able to take advantage of their flexibility and power, will become increasingly important. This project will impact thousands of students over the course of the project and will allow them to become involved in the development of new teaching methods and materials. This proposal focuses on drawing and representing different chemical structures, but the methods we propose are ultimately applicable to many other domains, such as mathematical equations, geometric figures, and computer science data structures.