Comparative Analysis of Adaptive Learning and Fast for Word Programs for ASD Students in Learning English, and Mathematics, and Predicting Future Academic Performance Using Machine Learning Algorithms

Aisha Ahmed1*, Abdullahi Abdu Ibrahim2

1Department of Electrical and Computer Engineering, Bani Waleed University, Bani Waleed, Libya.
2Department of Electrical and Computer Engineering, Altinbas University, Istanbul, Turkey.

Introduction

Autism Spectrum Disorder (ASD) is a developmental disorder characterized by difficulties with social interaction, communication, and repetitive behaviors [1]. According to the Centers for Disease Control and Prevention (CDC), approximately 1 in 54 children in the United States is diagnosed with ASD [2]. These students often face significant challenges in traditional educational settings, where standard teaching methods may not cater to their unique learning needs. Academic interventions for students with ASD have evolved over the years, with increasing emphasis on personalized and technology-driven approaches. Adaptive learning methods utilize data analytics and machine learning algorithms to tailor educational content to the individual learner, adjusting in real time to provide a customized learning experience [3]. The Fast ForWord program is a computer-based intervention designed to enhance cognitive skills related to language and reading, leveraging neuroplasticity to improve cognitive function [4]. Understanding the most effective educational strategies for students with ASD is crucial for several reasons. First, students with ASD often experience lower academic achievement compared to their neurotypical peers due to the misalignment between their learning needs and traditional teaching methods [5]. By identifying effective interventions, educators can provide more supportive and effective learning environments, leading to better educational outcomes and quality of life for these students. Second, the increasing prevalence of ASD necessitates scalable and effective educational solutions. As more students are diagnosed with ASD, schools and educators face the challenge of meeting diverse learning needs with limited resources. Adaptive learning methods and the Fast ForWord program offer scalable solutions that can be implemented across various educational settings [4]. Lastly, this study contributes to the growing body of research on technology-enhanced learning. By comparing adaptive learning methods with the Fast ForWord program and integrating machine learning models to predict student scores, this study aims to provide evidence-based recommendations for integrating technology into special education, thereby informing policy and practice. Adaptive Learning Methods General Approach: Adaptive learning methods are a broad category of educational techniques that use technology to customize learning experiences for individual students [6]. Technology Use: These methods leverage various algorithms, data analytics, and sometimes artificial intelligence to continuously adjust the content and difficulty based on student performance [7]. Adaptive learning can be applied across multiple subjects and educational levels, from elementary education to higher education and professional training [8]. Personalization: The key feature is the continuous personalization of learning paths based on real-time data, providing individualized support and resources [9]. Fast ForWord Program: Fast ForWord is a specific program designed to improve language and literacy skills through a series of computer-based exercises [10]. The program specifically targets cognitive skills such as memory, attention, processing speed, and sequencing, which are essential for reading and learning [11]. User Adaptation: Exercises within Fast ForWord adapt to the user’s performance, but the program remains focused on language and literacy improvement rather than a broad educational scope [12]. Fast ForWord is an evidence-based intervention with specific studies supporting its efficacy in improving language skills in children with language impairments [13]. The primary objective of this study is to predict students’ scores using machine learning methods and evaluate the efficacy of adaptive learning methods in instructing English and Mathematics to students diagnosed with ASD, compared to the Fast ForWord program. Specific objectives include: 1. Assessing Academic Performance: To compare the improvement in English and Mathematics performance between students using adaptive learning methods and those using the Fast ForWord program. 2. Evaluating Engagement and Behavior: To examine the levels of student engagement and behavioral outcomes associated with each intervention. 3. Predicting Student Scores: To utilize machine learning models (SVM, KNN, GPR, LR) to predict student scores and identify factors contributing to educational outcomes. Literature Review Adaptive Learning Methods Adaptive learning methods have emerged as a transformative approach in the education sector, leveraging technology to provide personalized learning experiences tailored to individual student needs. These methods utilize data analytics, machine learning, and artificial intelligence to continuously assess and adapt the learning process, aiming to enhance student engagement and improve learning outcomes. Adaptive learning systems are designed to modify the presentation of material in response to student performance. These systems utilize various data points, such as quiz results, interaction patterns, and time spent on tasks, to dynamically adjust content and instructional methods [14]. Personalized learning, a broader concept encompassing adaptive learning, refers to educational approaches that tailor learning experiences to meet the diverse needs of students. While adaptive learning focuses on real-time content adaptation, personalized learning may also include strategies beyond real-ti adjustments, such as project-based learning and student choice [15]. The historical development of adaptive learning methods reveals a progression from early rule-based systems to modern data-driven approaches. Early systems used decision trees and expert-de????ined pathways, requiring significant manual input to create adaptive learning paths. In contrast, contemporary adaptive learning leverages sophisticated algorithms and large datasets, enabling more nuanced and effective adaptations based on real-time data [16]. This shift has been driven by advances in machine learning techniques, which have enhanced the ????lexibility and scalability of adaptive learning systems [17]. Adaptive learning technologies have gained significant attention in educational research for their potential to provide personalized and engaging learning experiences. These systems use algorithms and data analytics to adjust the content and dif????iculty level of educational material in real-time, based on individual learner’s performance and progress [18]. This personalized approach is particularly bene????icial for students with diverse learning needs, such as those diagnosed with autism spectrum disorder (ASD). Adaptive learning platforms, such as Dream Box Learning for Mathematics and Smart Sparrow for various subjects, have demonstrated positive impacts on student outcomes. A study by [19] found that students using adaptive learning technologies showed signi????icant improvements in their academic performance compared to those receiving traditional instruction. These platforms often incorporate elements of gami????ication and interactive content, which can enhance engagement and motivation among learners [20]. In the context of ASD, adaptive learning methods are especially promising due to their ability to tailor instruction to the speci????ic needs and preferences of each student. For instance, students with ASD often bene????it from repetitive and structured learning activities, which can be effectively provided by adaptive systems [21]. Moreover, these technologies can reduce anxiety and frustration by ensuring that tasks are appropriately challenging without being overwhelming [22]. Adaptive learning methods utilize a variety of machine learning algorithms to analyze and interpret student data. Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Naive Bayes (NB), and Arti????icial Neural Networks (ANN) are commonly used algorithms in adaptive learning systems [23]. These algorithms help identify patterns in student performance, predict learning outcomes, and provide customized learning experiences. Moreover, data analytics plays a crucial role in interpreting the vast amounts of data generated by learners. Techniques such as predictive analytics and learning analytics are widely used to identify learning patterns and inform the adaptations made by the system [24]. Reinforcement learning, a method where algorithms learn to make a sequence of decisions by rewarding desirable outcomes, is also employed to optimize learning paths based on student interactions and feedback [25]. The application of adaptive learning methods has shown signi????icant bene????its in various educational settings. For instance, Khan Academy and Carnegie Learning have successfully implemented adaptive learning platforms that customize instructional content to individual learners’ needs, resulting in improved engagement and learning outcomes [26]. However, challenges such as data privacy, implementation costs, and resistance to change remain signi????icant hurdles to the widespread adoption of adaptive learning methods [27]. Recent Advancements in Adaptive Learning Technologies Arti????icial Intelligence and Machine Learning: Recent developments in AI and machine learning have enhanced the ability of adaptive learning systems to analyze vast amounts of data and provide highly personalized learning experiences. AIdriven platforms such as Coursera and EdX use machine learning algorithms to recommend courses and content tailored to individual learning paces and preferences [28]. Gami????ication and Engagement: Integrating gami????ication into adaptive learning systems has proven effective in increasing engagement and motivation among learners, including those with ASD. Programs like Prodigy and Class Craft incorporate game-like elements to make learning more engaging and interactive [29]. Real-Time Feedback and Assessment: Advanced adaptive learning systems now offer real-time feedback and assessment, allowing for immediate adjustments to learning paths and strategies. Platforms like Khan Academy use real-time data to provide instant feedback and suggest personalized practice exercises [6]. Applications to ASD Personalized Learning Paths: Adaptive learning technologies are particularly bene????icial for students with ASD as they provide personalized learning paths that cater to individual strengths and challenges. Programs like Teach Town and Rethink Autism offer personalized curricula designed speci????ically for students with ASD, adapting to their unique learning needs [30]. Social and Communication Skills Development: Recent adaptive learning tools are focusing on improving social and communication skills among learners with ASD through interactive and engaging activities. The Social Express is a program designed to help children with ASD develop social skills through interactive simulations and activities [31]. Behavioral and Cognitive Support: Adaptive learning systems incorporate features that provide behavioral and cognitive support tailored to the needs of students with ASD. Cognitive Behavioral Intervention for Trauma in Schools (CBITS) is a program that uses adaptive learning technologies to support cognitive and behavioral development in children with ASD [32]. Fast For Word Program The Fast ForWord program is a computer-based intervention designed to improve language and reading skills through cognitive training exercises. It is grounded in the principles of neuroplasticity, which propose that targeted cognitive activities can lead to structural and functional changes in the brain [33]. The program includes a series of adaptive exercises that aim to enhance various cognitive skills, such as memory, attention, processing speed, and sequencing [34]. Several studies have examined the effectiveness of the Fast ForWord program in improving language and reading abilities in children with learning difficulties, including those with ASD. [34] reported that children with language impairments who participated in the Fast ForWord program showed significant improvements in language skills compared to a control group. Similarly, a study by [35] found that the program led to notable gains in reading skills among school-aged children with language impairments. However, the efficacy of Fast ForWord for students with ASD has produced mixed results. Some studies have shown positive outcomes, such as improved auditory processing and language skills [36], while others have reported limited or no significant improvements [36]. These mixed ????findings highlight the need for further research to understand the specific conditions under which the program is most effective for students with ASD. Education AND ASD Education for students with ASD requires specialized approaches that address their unique cognitive, behavioral, and sensory needs. Traditional educational methods often fall short of meeting these needs, leading to challenges in academic achievement and social integration [37]. Effective educational

1. Vahia VN (2013) Diagnostic and statistical manual of mental disorders 5: A quick glance,” Indian J. Psychiatry, 55: 220-223. Link: https://bit. ly/3WEClZC

2. Jon B, Lisa W, Deborah LC, Maenner MJ, Julie D, et al. (2018) Prevalence of Autism Spectrum Disorder Among Children Aged 8 Years - Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2014. MMWR. Surveill. Summ 67: 1-23. Link: https://bit. ly/3WHxDdF

3. Penuel WR, Johnson R (2016) Review of continued progress: Promising evidence on personalized learning Boulder, CO Natl. Educ. Policy Center. Retrieved Sept 1: 12. Link: https://bit.ly/3y7QchI

4. Knight V, McKissick BR, Saunders A (2013) A review of technology-based interventions to teach academic skills to students with autism spectrum disorder. J. Autism Dev. Disord 43: 2628-2648. Link: https:// bit.ly/46FwO8F

5. Turnock A, Langley K, Jones CRG (2022) Understanding stigma in autism: A narrative review and theoretical model. Autism in Adulthood 4: 76-91. Link: https://bit.ly/3WmG5xw

6. Pane JF, Grif????in BA, McCaffrey DF, Karam R (2014) Effectiveness of cognitive tutor algebra I at scale. Educ. Eval. Policy Anal 36: 127-144. Link: https://bit.ly/4c95Y9Q

7. Liu TC (2022) A case study of the adaptive learning platform in a Taiwanese Elementary School: Precision education from teachers’ perspectives. Educ. Inf. Technol 27: 6295-6316. Link: https://bit. ly/4d26yY8

8. Walkington C, Bernacki ML (2014) Motivating students by ‘personalizing’ learning around individual interests: A consideration of theory, design, and implementation issues. in Motivational interventions. Emerald Group Publishing Limited 18: 139-176. Link: https://bit.ly/4dizomW

9. Johnson L, Becker SA, Estrada V, Freeman A (2015) NMC horizon report: 2015 museum edition. The New Media Consortium. Link: https://bit.ly/4????jl8fr

10. Ronald BG, Diane FL, Lavae MH, Thomas B, Craig AC, et al. (2008) The efficacy of Fast ForWord language intervention in school-age children with language impairment: A randomized controlled trial. J Speech Lang Hear Res 51: 97-119. Link: https://bit.ly/3SPG3gP

11. Moroschan GC (2013) Visual and auditory phonological processing during reading and listening. University of Alberta (Canada). Link: https://bit.ly/3yhwSOT

12. Slavin RE, Cheung A, Groff C, Lake C (2008) Effective reading programs for middle and high schools: A best-evidence synthesis. Read. Res. Q 43: 290-322. Link: https://bit.ly/3SsOAWN

13. Lavigne AL, Good TL (2013) Teacher and student evaluation: Moving beyond the failure of school reform. Routledge, Link: https://bit. ly/3YiNC31

14. Brusilovsky P, Millán E (2007) User models for adaptive hypermedia and adaptive educational systems,” in The adaptive web: methods and strategies of web personalization, Springer 3-53. Link: https://bit. ly/3zYB8mR

15. Chen CM, Duh LJ (2008) Personalized web-based tutoring system based on fuzzy item response theory. Expert Syst. Appl 34: 2298-2315. Link: https://bit.ly/4cQYsSd

16. Corbett AT, Anderson JR (1994) Knowledge tracing: Modeling the acquisition of procedural knowledge. User Model. User-adapt. Interact 253-278. Link: https://bit.ly/3LGrrwf

17. Wenting Ma, Olusola OA, John CN, Qing L (2014) Intelligent tutoring systems and learning outcomes: A meta-analysis. J. Educ. Psychol 106: 901-908. Link: https://bit.ly/3Wrg6Vs

18. Desmarais MC, d Baker RSJ (2012) A review of recent advances in learner and skill modeling in intelligent learning environments User Model. User-adapt. Interact 22: 9-38. Link: https://bit.ly/3YphHy0

19. Pane JF, Steiner ED, Baird MD, Hamilton LS (2015) Continued Progress: Promising Evidence on Personalized Learning. Rand Corp. Link: https://bit.ly/4cXK7mV

20. Karagiorgas DN, Niemann S (2017) Gami????ication and game-based learning. J. Educ. Technol. Syst 45: 499-519. Link: https://bit. ly/3LJihPm