A Synthesis Proposal: Merging AI in Education with Automata Theory Andre Kenneth Chase Randall University of Massachusetts Amherst (UMass Amherst), Amherst, MA 01003 USA Abstract As classroom sizes expand, formal education faces increased challenges in providing scalable, targeted feedback based on student engagement. 𝐶𝑆250, an undergraduate core course in discrete mathematics, covers topics such as logic, elementary number theory, proof by induction, recursion on trees, search algorithms, regular languages, finite state machines, and computability. These concepts often present challenges due to their abstract nature and the precision required in logical reasoning. Primarily enrolling computer science and related majors, the course benefits from reusable learning objects (RLOs) designed to support concept mastery. In this context, the author proposed a new discussion material on mathematical foundations, specifically targeting regular language expression. He tested a Python tool that allowed students to check their answers’ correctness while mastering regular language expressions. Students completed the Python tool and a survey, which confirmed the tool’s usefulness and provided valuable feedback for iterative design. This paper aims to contribute to the existing body of knowledge on AI in education by shedding light on student perspectives with RLOs. In future iterations, we plan to recruit a more diverse range of educators, including female educators from all-women colleges, to broaden our perspective on instructional effectiveness. Moving forward, we seek to explore the balance between technological and human interventions required for effective course delivery. Although these findings are preliminary, continued research and richer data may reveal organic, inductive themes as this iterative process unfolds. Keywords AI in Education, Automata Theory, Synthesis Proposal, Doctoral Consortium, Regular Expressions, Reusable Learning Objects (RLOs) 1. Introduction The RLO includes several key components: a Python- based interactive tool for answer verification, structured As classroom sizes expand, educators increasingly need practice exercises, and integrated feedback mechanisms scalable, targeted feedback mechanisms to support stu- that guide students through the learning process. De- dent engagement. 𝐶𝑆250, an undergraduate core course signed to actively support students, the RLO delivers in discrete mathematics at UMass Amherst, covers key immediate feedback to help them correct errors and topics such as logic, elementary number theory, proof strengthen their understanding. During the study, the by induction, recursion, search algorithms, regular lan- author assigned a treatment group to use the RLO for guages, finite state machines, and aspects of computabil- regular expression exercises, where students received ity [1]. These concepts challenge students due to their instant feedback as they practiced. The study also in- abstract nature and the logical precision they demand. cluded pre- and post-surveys with the treatment group, To address these challenges, the author developed a capturing changes in understanding and perceptions. Reusable Learning Object (RLO), which refers to “any Survey responses indicated that the RLO was primarily digital resource that can be reused to support learning” used for reinforcing course material and problem-solving [2]. The RLO incorporates modular, adaptable resources skills, with many participants planning to apply these that support specific educational goals. In this context, skills directly to their 𝐶𝑆250 assignments and exams. the RLO focuses on simplifying complex topics in 𝐶𝑆250 Participant #2 noted, “Yes, it was very useful in helping through interactive and reusable digital content, helping check work,” while Participant #5 highlighted the tool’s students engage independently with difficult concepts, benefit in understanding regular expressions, stating, “I especially those involving regular expressions. would recommend it to others because it was helpful in understanding how regex are set up.” However, some EC-TEL 2024: Doctoral Consortium for Nineteenth European Confer- participants felt the tool’s feedback was not sufficiently ence on Technology Enhanced Learning, September 16 - 20, 2024, Krems, personalized, as Participant #6 remarked, “It wasn’t per- Austria * Per application process, Doctoral Consortium paper authored by sonalized, but it was useful,” indicating that while benefi- student himself with acknowledgments of supervisor and collabo- cial, it did not offer individualized guidance. rators at the end of the paper. A number of participants also suggested improvements $ technicalchase@gmail.com (A. K. C. Randall) to make the tool more user-friendly and accessible. For € https://technicalchase.com/ (A. K. C. Randall) instance, Participant #3 recommended “more language  0000-0002-3971-3775 (A. K. C. Randall) © 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License options” and “text-to-speech” functionality, reflecting a Attribution 4.0 International (CC BY 4.0). 1 CEUR ceur-ws.org Workshop ISSN 1613-0073 Proceedings Andre Kenneth Chase Randall CEUR Workshop Proceedings 1–9 desire for inclusive design to meet diverse learning needs. development and user feedback, ensuring that adaptive Initial usability challenges were also reported, with Par- tools like ITS are aligned with real-world educational ticipant #1 mentioning, “I wasn’t sure how to use the tool needs and specific challenges faced by students, instruc- at first,” emphasizing the importance of an intuitive user tors, and administrators [11]. This process aligns with experience. Participant #15 expressed reluctance to rec- AI’s objectives in education by fostering adaptable and ommend the tool, commenting, “I would not recommend scalable solutions suitable for diverse learning contexts. the tool to others; it did not explain very much and took Implementing a SWOT (Strengths, Weaknesses, Op- a long time to run,” highlighting areas for improvement portunities, Threats) analysis further aids in evaluating in efficiency and instructional depth. While the RLO was the feasibility of adaptive learning technologies, identi- seen as helpful for coursework, some participants were fying strengths like personalized learning pathways and uncertain about its long-term applicability, as Participant addressing limitations such as data privacy concerns [? #12 noted, “I do not plan to use it in my personal life,” ]. Together, Steve Blank’s framework and SWOT anal- suggesting that its perceived value was largely limited ysis underscore the importance of developing adaptive to immediate academic goals. educational tools that effectively respond to the evolving Through an iterative design process, the author struc- needs of educational stakeholders [8]. tured the RLO to enhance student engagement by al- Moreover, adaptive algorithms similar to those em- lowing them to interact with content at their own pace. ployed in fields like gaming—where neural networks, In this paper, the author uses the terms "Learning Ob- enhanced by genetic algorithms, refine responses based ject" and "Reusable Learning Object" interchangeably to on real-time performance feedback—illustrate how it- describe these modular educational resources [3]. erative adaptation could similarly benefit educational Beyond promoting independent learning, the RLO contexts [12]. Such an approach could allow educational aims to increase engagement, strengthen problem- AI systems to adjust dynamically to varying learner pro- solving skills, and improve students’ ability to master gressions, supporting engagement and promoting per- abstract mathematical concepts. Prior research highlights sonalized learning paths across diverse skill levels. how learning analytics and feedback within RLOs can enhance learning outcomes, especially in skill-based sub- jects like those in 𝐶𝑆250 [4, 5] 3. Study Design The next sections address the related works, the study The study design happened as an iterative process. First design, development, and testing to ensure the RLO’s the authored created a pilot project as a proof of concept. reusability. Finally, the author outlined a plan for a proof Thereafter, the author became the entrepreneurial lead of concept, drawing from the doctoral program insights, for Team Intelligent Tutoring Systems R Us. Team In- related works and testing within real-world educational telligent Tutoring Systems R Us obtained U. S. National contexts. Future endeavors, especially in the context of Science Foundation’s Innovation Corps (I - Corps™) Cus- a doctoral program, require continuous refinement and tomer Discovery Project funding. optimization of the initial concept as presented herein to achieve the best outcomes. 3.1. Course Pilot Project 2. Related Works As part of teaching assistant preparation course, the au- thor and classmates conducted a pilot study without In- As classroom sizes expand, educators face increased chal- stitution Review Board (IRB) approval. The pilot study lenges in providing scalable, targeted feedback based on ended with 147 participants in March-May 2022. student engagement [6]. The application of AI in educa- In pilot study, they introduced a python script and tion addresses these issues through tools like Intelligent sought to probe students’ reactions to its usefulness as a Tutoring Systems (ITS), adaptive content creation, and au- RLO. tomated administrative tasks [7, 8]. AI in education com- They designed the RLO to enhance the students’ un- bines multiple fields—learning science, human-computer derstanding of a regular expressions. During the pilot interaction (HCI), software engineering, natural language study, they gave participants the Python script along with processing (NLP), and machine learning (ML) [9]. How- instructional materials for completing a series of tasks. ever, developing effective ITS tools requires seamless The study concluded with a survey featuring open-ended integration of advanced algorithms, deep pedagogical questions to collect qualitative data on the students’ ex- knowledge, and user-centered design [10]. periences and feedback. Steve Blank’s Customer Discovery framework, as out- The survey responses varied significantly. Over 31.3% lined in the NSF I-Corps Teaching Handbook, reinforces of participants reported “no challenges,” indicating a this interdisciplinary approach by emphasizing iterative smooth experience with the Python script. In contrast, 2 Andre Kenneth Chase Randall CEUR Workshop Proceedings 1–9 under 15.6% of participants expressed difficulties, with 4. Administrators aim to boost institutional rank- comments such as “I found the instructions hard to follow ing through future-generation technologies while and get the code running.” These responses highlighted addressing academic integrity, accessibility, and areas for instructional materials improvements. privacy concerns. Additionally, 24.7% of participants, particularly those majoring in computer science, showed motivation to- 3.2.2. I-Corps Study Methods wards the concept of using programming in their course- work. These students appreciated the practical appli- 1. Customer Segmentation: The team focused on cation of programming skills and expressed interest in understanding the mental models of STEM stu- further integrating such tools into their studies. dents, professors, and administrators, exploring Overall, the pilot study provided valuable insights into their behaviors, characteristics, and needs. the effectiveness of the RLO and the Python script. The 2. Customer Discovery Interviews: Using a hy- feedback collected not only show improved student learn- brid approach of virtual and in-person interviews, ing experience but also aided the instructional materials the team engaged participants to uncover educa- refinement as part of an IRB protocol (See Figure#1). tional pain points and evaluate the RLO’s com- mercial viability. 3. Data Collection and Analysis: With over 90 3.2. U.S. National Science Foundation’s contacts from 15 colleges and universities, the Innovation Corps (I-Corps™) team conducted 35 detailed interviews with stu- Customer Discovery Project dents, instructors, and administrative profession- The National Science Foundation (NSF) I-Corps Teams als in instructional roles. program provides an intensive seven-week entrepreneur- ship training course with mentorship and funding for 3.2.3. I-Corps Conclusion customer discovery. As part of this program, the author The team initially accepted the null hypothesis but discov- led Team Intelligent Tutoring Systems R Us as the en- ered a strategic pivot by broadening customer segmenta- trepreneurial lead, receiving a travel grant supported by tion beyond the host institution. This pivot enabled the Cornell Tech and the National GEM Consortium to ex- development of a more dynamic business model, contin- plore market potential for the Reusable Learning Object uously revised through customer insights. Aligning with (RLO) project [13]. Woolf et al.’s AI Grand Challenges, future goals include The I-Corps program guided the team in conducting creating intelligent tutoring systems, real-world simu- a hybrid field study to test initial hypotheses and make lation environments, and natural language processing adaptive pivots. Insights from customer discovery in- capabilities to enhance adaptive learning [8, 7]. terviews were essential in refining the RLO’s interface and feedback mechanisms. This process identified key challenges, value propositions, and market opportunities, 4. Study Development structured within a business model framework as illus- trated in the SWOT analysis (See Figure 2). Future work This study employed a mixed-methods research design, includes developing a comprehensive Strengths, Weak- consisting of both quantitative and qualitative data col- nesses, Opportunities, and Threats (SWOT) analysis as lection and analysis. The author conducted three phases: preparation for a national I-Corps application and further expanding the study’s scope. • Phase 1: Literature Review and Case Studies. The first phase of the study involved another compre- hensive review of the literature on the use of AI 3.2.1. I-Corps Research Questions and Hypotheses in Education. This phase helped to identify best 1. What value propositions exist for students, profes- practices and potential challenges associated with sors, and administrators in using adaptive learn- using AI in Education. During this phase, the au- ing tools for instructional support? thor worked with an undergraduate student to 2. STEM students seek adaptive tools to aid in con- complete an Honors Thesis Project, where the tent mastery and skill development, including author served as a committee member. grit, motivation, and soft skills beyond technical • Phase 2: Surveys and Interviews. The second competence. phase of the study involved the administration 3. STEM instructors desire adaptive tools that sim- of surveys and interviews to educators, teaching plify grading, enable course scalability, and intro- assistants, and students. The surveys and inter- duce innovative learning experiences. views assess perceptions of RLOs and its potential impact and other considerations. 3 Andre Kenneth Chase Randall CEUR Workshop Proceedings 1–9 Figure 1: Team Intelligent Tutoring Systems R Us utilized Steve Blank’s Customer Discovery framework and Alexander Osterwalder’s Business Model Generation along with methods used by the National Science Foundation (NSF). • Phase 3: Analysis and Synthesis. The third phase 4.1. IRB Protocol of the study involved the analysis and synthesis of The entire process adhered to ethical guidelines outlined the data collected in phases 1 and 2. Quantitative in the IRB protocols #5139 "Team Chase Undergraduate data analyzed using descriptive and inferential Research Volunteers (URV)" and #5358 "AI in Education statistics, while qualitative data analyzed using and Automata Theory: Synthesis Proposal". content analysis. The results of the analysis syn- thesized to identify best practices and potential recommendations for future research and prac- 4.2. Study Testing tice. 4.2.1. Consent Form After beginning Phase 1, the author collaborated withBefore conducting interviews with educators, partici- an undergraduate student on the Honors Thesis Projectpants were required to provide informed consent. The committee to explore educators’ mental models and per- consent form outlined the purpose of the semi-structured spectives on how RLOs impact learning. Key research interview, which aimed to contribute to an honors thesis questions included: exploring adaptive learning tools for future-generation technologies within the realm of AI in educational 1. How does the RLO compare with traditional Reusable Learning Objects (RLOs). The research study methods in understanding regular expressions? was designed to unpack the mental models surrounding 2. How does the RLO shape students’ problem- AI in education. Eligible participants for the study in- solving skills? cluded instructors and teaching assistants specializing 3. Does RLO use deepen understanding of automata in areas such as proofs, induction, reason, number the- principles? ory, automata theory, regular expressions, finite state 4. What is the connection between RLO engagement machines, and related courses. We informed participants and CS250 performance? that the interview lasted approximately 20 minutes and 5. How do demographics correlate with learning primarily focused on soliciting their opinions and views outcomes? regarding their experience with curriculum development and AI in Education, as well as related research topics. 4 Andre Kenneth Chase Randall CEUR Workshop Proceedings 1–9 Figure 2: SWOT analysis pulled from the SWOT Analysis on AI in Education by Andre Kenneth Chase Randall [14] The interview posed minimal risks such as potential fa- Table 1 tigue or boredom. We disclosed risks and encouraged to Description of our participants for semi structures interviews. adhere to the 20-20-20 rule for eyestrain and to maintain Eligible participants for the study included instructors and teach- hydration throughout the interview. The 20-20-20 rule ing assistants with selfdescribed expertise within the discrete advises taking a 20-second break from looking at screens mathematics area such as proofs, induction, reason, number the- every 20 minutes. During this break, focus on something ory, automata theory, regular expressions, finite state machines, and related courses. at least 20 feet away. This practice helps prevent eye strain and fatigue caused by prolonged screen use. The Institution TA Instructors consent form also covered data privacy, assuring par- University of Massachusetts Amherst 3 1 University of Pennsylvania 0 1 ticipants that personal information would be handled University of California Berkeley 0 2 confidentially with the IRB protocols. Data collection Cornell University 0 1 will be conducted anonymously with transcribed inter- Total 3 5 views in the study’s codebook. Communication on any concerns was encouraged throughout the interview as well. The consent form process and interview proto- 4.2.2. SEMI-STRUCTURED INTERVIEW SCRIPT col were carefully designed to uphold ethical standards, ensure participant confidentiality and comfort, and facil- We designed the questions to explore educators’ mental itate valuable insights into participants’ mental models models regarding their teaching experiences and perspec- surrounding AI in education. tives on adaptive learning tools. Some questions were adapted from the I-Corps customer discovery project to ensure depth in understanding both practical and attitu- dinal dimensions of their insights. The specific questions 5 Andre Kenneth Chase Randall CEUR Workshop Proceedings 1–9 Table 2 learning management systems, and reliance on teaching Usability Testing Participants; 44 participants completed the assistants, were often mentioned as significant factors consent forms, pre survey, and post survey. 65 students opted to influencing tool adoption. use the alternative research for the option to get two extra credit This feedback forms a basis for refining adaptive learn- points with 194 students opting not to partici ing tools to meet real-world teaching demands. Group Number of Students Treatment Group 44 Control Group 65 5. Conclusion and Future Work Non-participants 194 Total 303 One of the most notable advantages of AI in education is its ability to personalize learning, tailoring educational experiences to meet the diverse needs, preferences, and included are as follows: learning styles of students [15]. Using an RLO as the foun- dation for intelligent tutoring systems could enhance the 1. Course Background: Educators describe their educational landscape by providing customized learning teaching context. paths and tailored feedback [9]. Participant feedback in 2. Teaching Challenges: Identification of course- this study highlighted the RLO’s potential for support- specific challenges. ing course material and fostering problem-solving skills, 3. Adaptive Solutions: Educators suggest solu- with Participant #2 noting, “Yes, it was very useful in tions to their teaching challenges. helping check work.” Participant #5 also recommended 4. Solution Drawbacks: Exploration of limitations the tool, sharing that “it was helpful in understanding in current methods. how regex are set up.” 5. Design Preference: Educators explain why pre- Despite these benefits, participants expressed a desire ferred designs are effective. for more personalized feedback. As Participant #6 re- marked, “It wasn’t personalized, but it was useful,” under- 6. Teaching Assistant Roles: Understanding TA scoring the need for individualized support within such responsibilities in supporting adaptive tools. tools. Suggestions for improvement, including “more lan- 7. Adaptive Tools Perspective: Insights on poten- guage options” and “text-to-speech” functionality (Par- tial and challenges of adaptive tools. ticipant #3), point to the importance of accessibility and 8. Incorporation Process: Discuss practical steps adaptability in future RLO iterations. Additionally, feed- for tool implementation. back from Participant #1, who stated, “I wasn’t sure how 9. Tool Features and Functionality: Discuss ex- to use the tool at first,” indicates that enhancing user pectations for tool features to aligned with teach- guidance could improve ease of use. Comments from ing styles. Participant #15, who noted, “I would not recommend the 10. Professional Recommendations: Suggestions tool to others; it did not explain very much and took a for other participants in adaptive learning studies. long time to run,” suggest areas for improving efficiency 11. Additional Insights: Educators provide and instructional depth. thoughts on adaptive learning tools. Future work on this RLO aligns with the AI Grand 12. Open Feedback: Educators can add any other Challenges for Education outlined by Woolf et al., partic- relevant thoughts. ularly the goals of providing "mentors for every learner" and "lifelong and life-wide learning" [8]. In line with AI’s From interview with instructors and teaching assis- strengths, weaknesses, opportunities, and threats in edu- tants, we gain insight into teaching experiences, adaptive cation, as explored by Randall in recent discussions with learning perspectives, and practical integration strate- HBCU faculty [? ], future RLO development will focus gies. Generally, interviewees highlighted several com- on designing adaptable tools to support students from mon themes: varied backgrounds, ensuring inclusivity, accessibility, Challenge in Differentiating Instruction: Educa- and effectiveness. Planned enhancements include: tors often noted difficulty in addressing the varying levels within a single classroom and expressed interest in tools • Adaptive Assessment and Skill Tracking: Im- to support personalized pacing. plementing a system that tracks student perfor- Positive Reception for Adaptive Tools: Many val- mance to provide tailored difficulty levels and ued adaptive tools for their potential in enhancing con- additional exercises based on individual progress. cept retention and student engagement, specifically when • Dynamic Hints and Explanations: Adding tailored to individual learning trajectories. context-aware hints that address specific errors, Implementation Considerations: Practical con- such as syntax misunderstandings in regular ex- straints, including time, ease of integration with current pressions, and provide tailored guidance. 6 Andre Kenneth Chase Randall CEUR Workshop Proceedings 1–9 • Immediate Corrective Feedback: Offering for- Research Volunteers), Jasmine Ngo assisted the deploy- mative feedback that is specific to the type of ment of the semi structured interviews and the survey error made, enabling students to correct misun- instruments as part of an undergraduate Computer Sci- derstandings promptly. ence Department Honors Thesis Project. • Personalized Learning Pathways: Creating Shravan Janga, served as part of Team Intelligent Tu- customized learning pathways based on initial as- toring Systems R Us. He provided invaluable back up sessments or adaptive quizzes, allowing students support during the customery discovery phase. to focus on areas that require reinforcement. The author acknowledges partial funding from the • Progress and Performance Dashboards: De- National GEM Consortium, the Intel Scholars Program, veloping dashboards that give students personal- the NSF I-Corps Hub for Interior Northeast led by Cor- ized insights into their strengths and areas need- nell Tech, Manning College of Information and Com- ing improvement, supporting self-directed learn- puter Science along with the Spaulding- Smith Fellow- ing. ship awarded by the UMass Graduate School. • Natural Language Processing (NLP) for Open- Ended Responses: Using NLP to analyze open- ended responses and provide customized feed- 7. Appendices back based on the semantics of students’ answers. • Gamification and Motivational Feedback: In- 7.1. Appendix A: Automata Theory corporating gamified elements that reward indi- Discussion Objectives vidual achievements and keep students motivated Every week for the length of the course, students met throughout their learning journey. for a 50 minutes discussion group to cover the following • Student Profile Customization: Enabling stu- topics: dents to set preferences or learning goals within the RLO, allowing it to customize feedback based 1. “What is a Proof?” on their unique needs and learning styles. • Objective: To foster an understanding of mathematical proof in real-world scenarios These enhancements aim to make the RLO a more and to practice constructing proofs based flexible, accessible, and impactful educational tool that on definitions, highlighting the role of pre- supports a broad range of learners. By focusing on per- conditions, postconditions, and loop in- sonalization and adapting to diverse learning needs, this variants in validating code. RLO aligns with AI’s grand challenge to democratize ed- 2. "A Murder Mystery” ucational resources and extend individualized learning • Objective: To develop deductive reason- opportunities beyond traditional settings. ing skills using propositional logic, demon- strating the process of narrowing down 6. Acknowledgments possibilities based on given clues, and ap- plying rules of propositional logic to de- The author thanks the study participants for their valu- duce conclusions efficiently. able insights and time. Besides the study participants, 3. “Practicing Proofs” the author extends appreciation to several entities who • Objective: To enhance proficiency in ap- contributed to the development of this paper and its find- plying proof methods to statements about ings. Research professor Beverly P. Woolf serves as his functions and relations, emphasizing pred- research advisor. Professor Woolf specializes generally in icate proof rules and the significance of educational computer science research and more specifi- properties of functions and relations. cally in intelligent tutoring systems. Professor David A. 4. “Infinitely Many Primes” Mix Barrington, the authors teaching assistant supervisor and synthesis reader, focuses on computational complex- • Objective: To apply proof techniques and ity, Boolean circuits, automata, and logic. The combined congruence principles to establish and expertise of both professors bridges diverse fields, offer- understand facts about prime numbers, ing a comprehensive perspective on the project. specifically the infinitude of primes. Samuel Osebe and Sabrina Zaman Ishita assisted in 5. “Practicing Induction Proofs” RLO prototype development as part of a course team • Objective: To cultivate a strong founda- project. They both helped in crafting a proof of concept tion in mathematical induction, focusing for the pilot project. on the structure of induction proofs includ- Under the UMass Human Protect Office (Internal Re- ing base cases, inductive hypotheses, and view Board Protocol #5139: Team Chase Undergraduate inductive steps. 7 Andre Kenneth Chase Randall CEUR Workshop Proceedings 1–9 6. “More Induction Problems” Define the language 𝐸𝐸𝑃 (“even-even-primitive”) of • Objective: To reinforce and expand stu- nonempty strings that are in 𝐸𝐸 and have no proper dents’ skills in mathematical induction, prefix in 𝐸𝐸. (That is, if 𝑤 ∈ 𝐸𝐸𝑃 and 𝑤 = 𝑢𝑣 with challenging them with diverse problems both 𝑢 and 𝑣 ∈ 𝐸𝐸, then either u = or v = .) It turns that require careful proof construction. out that while 𝐸𝐸𝑃 is harder than 𝐸𝐸 to describe in English, it has a simpler regular expression. 7. “Boolean Expressions” • Explain why 𝐸𝐸 = (𝐸𝐸𝑃 ) • Objective: To familiarize students with • Which strings of up to six letters are in 𝐸𝐸𝑃 ? Java-based boolean expressions, emphasiz- • Construct a regular expression for 𝐸𝐸𝑃 , and explain ing the differences between common pro- why this solves the main problem. gramming languages and the structure of tree representations in code. 8. “Course Evaluation Essay Questions” References • Objective: To gather feedback on the [1] U. of Massachusetts Amherst, Compsci 250: Intro- course content, pedagogy, and overall duction to discrete mathematics, n.d. URL: https: learning experience, aiding in future im- //people.cs.umass.edu/~barring/cs250/, accessed: provements and refinements. 2024-10-11. 9. “Designing Regular Expressions” [2] D. A. Wiley, Connecting learning objects to in- structional design theory: A definition, a metaphor, • Objective: To master the art of construct- and a taxonomy, in: The Instructional Use of ing accurate regular expressions for speci- Learning Objects, volume 2830, Agency for Instruc- fied languages, promoting a systematic ap- tional Technology and Association for Educational proach to capture all desired strings while Communications & Technology, Bloomington, In- excluding undesired ones. diana, 2002, pp. 1–35. URL: https://www.scirp.org/ 10. “Minimizing a DFA” reference/referencespapers?referenceid=589633. • Objective: To comprehend the principles [3] P. R. Polsani, Use and abuse of reusable learning behind the Myhill-Nerode Theorem, and to objects, in: E-education: Design and Evaluation, acquire hands-on experience in minimiz- Learning Technology Center, University of Arizona, ing DFAs by leveraging the equivalence 2003. URL: https://jodi-ojs-tdl.tdl.org/jodi/article/ classes of the relation on strings. view/jodi-105. 11. “Applications in Compilers” [4] R. H. Kay, L. Knaack, Exploring the impact of learning objects in secondary school mathemat- • Objective: To understand the foundational ics and science classrooms: A formative analysis, role of deterministic finite automata in Australasian Journal of Educational Technology the lexical analysis phase of compilers, 28 (2012) 775–792. URL: https://eric.ed.gov/?id= underscoring the transition from high- EJ1073840. level programming languages to machine- [5] I. Douglas, Instructional design based on reusable understandable code. learning objects: applying lessons of object- oriented software engineering to learning sys- 7.2. Appendix B: Discussion #8 on Regular tems design, in: 31st Annual Frontiers in Ed- Expressions ucation Conference. Impact on Engineering and Science Education. Conference Proceedings (Cat. Writing Exercise: No.01CH37193), volume 3, 2001, pp. F4E–1. doi:10. Construct a regular expression for the set EE ("even- 1109/FIE.2001.963968. even") of strings in {𝑎, 𝑏} that have both an even number [6] A. Staikopoulos, I. OKeeffe, B. Yousuf, O. Con- of 𝑎′ 𝑠 and an even number of 𝑏′ 𝑠. Justify your answer lan, E. Walsh, V. Wade, Enhancing stu- carefully – explain why your expression generates only dent engagement through personalized motiva- even-even strings and why it generates all even-even tions, in: ICALT, IEEE, Hualien, Taiwan, strings. 2015, pp. 340–344. URL: https://ieeexplore.ieee.org/ Note that all even-even strings have even length, so abstract/document/7265345. doi:10.1109/ICALT. you may think of the whole string as being broken up 2015.116. into two-letter blocks. [7] B. P. Woolf, Ai and education: Celebrating 30 years Here are some more hints. You are not required to use of marriage, in: B. Jesus, M. Kasia (Eds.), AIED them to solve the main problem, but they will probably Workshops, volume 1432 of CEUR Workshop Pro- be useful. 8 Andre Kenneth Chase Randall CEUR Workshop Proceedings 1–9 ceedings, CEUR-WS.org, Madrid, Spain, 2015, pp. 38–47. URL: https://ceur-ws.org/Vol-1432/ai_ed_ pap5.pdf. [8] B. Woolf, H. C. Lane, V. K. Chaudri, J. L. Kolodner, Ai grand challenges for education, AI Magazine 34 (2013) 66–84. URL: https://ojs.aaai.org/aimagazine/ index.php/aimagazine/article/view/2490. doi:10. 1609/aimag.v34i4.2490. [9] O. Gutierrez, A strategy for the development of reusable learning objects in the boston area ad- vanced technological education connections part- nership, in: Issues in Information Systems (IIS), vol- ume 4, International Association for Computer In- formation Systems (IACIS), Las Vegas, 2003, pp. 143– 149. URL: https://iacis.org/iis/2003-!/Gutierrez.pdf. [10] B. P. Woolf, Artificial Intelligence in Edu- cation, John Wiley & Sons, Inc., 1992. URL: https://web.cs.umass.edu/publication/docs/1991/ UM-CS-1991-037.pdf. [11] S. Blank, J. Engel, The national science founda- tion innovation corps™ teaching handbook, 2018. URL: https://venturewell.org/wp-content/uploads/ NSF_Handbook_Web.pdf. [12] M. Weeks, D. Binnion, A. C. Randall, V. Patel, Ad- venture game with a neural network controlled non-playing character, 16th IEEE International Conference on Machine Learning and Applications (ICMLA) (2017). URL: https://ieeexplore.ieee.org/ abstract/document/8260663. [13] C. E. Hub, Interior northeast i-corps hub, 2023. URL: https://eship.cornell.edu/item/ interior-northeast-i-corps-hub/, accessed: 2024-10- 11. [14] A. K. C. Randall, Ai in education: A swot anal- ysis with chase randall, 2024. URL: https://www. youtube.com/watch?v=OsO9nFa5rvQ. [15] I. Celik, M. Dindar, H. Muukkonen, The promises and challenges of artificial intelligence for teach- ers: A systematic review of research, TechTrends 66 (2022) 616–630. URL: https://doi.org/10.1007/ s11528-022-00715-y. 9