Artificial Intelligence in Education
Artificial intelligence (AI) has garnered a lot of attention in recent years, including a vigorous discussion how it might best be used to serve human kind. Whether AI is providing a solution to high frequency analytical tasks or generating complex solutions, there is potential for AI to resolve many of the problems facing human kind today. Education is one of the important areas where the application of AI might be able to address existing challenges and could transform how we provide an education to future generations.
Teachers face a variety of personal and professional challenges including low pay, poorly maintained physical space, and budget cuts resulting in a lack of classroom supplies. As a result of these and many other challenges, some teachers face something called “burnout syndrome”, a condition which has lead the International Labor Organization to classify teaching as a risky profession worldwide (Campos & Lucena, 2017).
This risky profession, combined with the pressure to constantly keep up with the introduction of new information and communication technology to the classroom leads me to believe that we must innovate in order to continue improving the ways in which we provide for our teachers and students. AI can address challenges such as these by providing numerous types of assistance as well as introducing blended learning styles while improving the teachers quality of life. Markovic and Jovanovic (2011) have shown that if educators can modify their teaching styles to match the learning styles of their students, educators can enhance the learning that is taking place in their classrooms.
One of the advantages that universities have over institutions of compulsory education is the varying cultures represented by the population. According to Kearney & Lincoln (2017) there is an increased demand to study abroad due to emerging economies and the global nature of advanced knowledge. AI will allow educators to bring this global nature into the classrooms of institutions of compulsory education.
When young minds gain exposure to other cultures around the world the result is an increase in global knowledge and the likelihood that students will use the skills they have developed abroad (Miglietti, 2015). AI has been touted as one of the leading methods of reducing repeat tasks, but recent developments in quantum neural networks will provide us with the ability to translate curriculum into multiple languages that allows educators to teach in multi-cultural classrooms without the need for language fluency. The same AI that can improve upon the global nature of our classrooms can help bring students from around the world together allowing institutions of compulsory education to benefit from the same varying cultures that universities strive to recruit.
Goals of this paper
The purpose of this paper is to explore the potential for AI to provide solutions to the challenges facing teachers and students. My aim is not to thoroughly explore what AI is or how it is being developed but instead to briefly explain the details of what it might accomplish in the field of education and how that accomplishment can be a positive change for compulsory education. For general purposes, the majority of what I will be discussing will be related to the ability of AI to sift through vast amounts of programmed information in a significantly reduced amount of time, develop into expert systems, interact and interpret natural languages, improve upon student outcomes.
Tutoring has often been used as an opportunity for students to increase their ability to learn a specific topic. However, tutoring typically requires many man hours to develop and alter a syllabus to cater to the students’ needs. Using expert systems and natural language, AI could reduce man hours and provide guidance on a level that is equal to a human tutor (Boulay, 2016).
The ability for AI to consume and analyze data in record time is important (to? In what way?). According to Campos and Lucena (2017) burnout syndrome is one of the common mental disorders experienced by educators. educators can use AI to analyze the flow of how an educator interacts with a system, the system being their daily responsibilities, and to determine if there are activities that AI could assist with or automate entirely. The specific learning styles of students has been shown, specifically in the e-learning platform, to influence the learning outcomes of students (Markovic & Jovanovic, 2011).
According to Boulay (2016) individual adaptation is paramount in the educational process experienced by students. Considering that the human brain evolves based on every interaction an individual faces. However, when an educator is responsible for thirty or more students in every class, adapting a learning plan to each individual student is hardly practical. AI provides the potential to form individualized learning plans based on the performance of students.
International experience is considered an advantage among graduated students and according to Miglietti (2015) those benefits are largely produced through “...exposure to diverse cultures, greater sensitivity and openness to cultural differences, as well as an understanding of domestic and global environments of organizations” (p. 46). Researchers are currently looking into automatic dictionary translation which would be implemented in human-in-the-middle systems (Krajewski, Rybinski, & Kozlowski, 2015). The research into semantic translation could produce the foundation for live translation inside schools providing educators with the opportunity to develop cross-cultural classrooms providing study-abroad benefits without requiring the expense of travel.
Quality of Life and Cognitive Assistants
In a study conducted in Brazil, 174 educators showed varying levels of emotional exhaustion and depersonalization, with six educators qualifying with burnout syndrome (Campos & Lucena, 2017). All of the individuals in this study taught at the collegiate level and a majority had earned a master’s degree or higher level of education. An environment leading to stressors such as these leads to a faculty with a compromised quality of life and a significant decrease in intellectual production (Campos & Lucena, 2017).
Burnout Syndrome has been identified as a serious occupational hazard through research concentrating on secondary school teachers. (Loonstra, Brouwers, & Tomic, 2009). According to Campos and Lucena (2017) Burnout Syndrome in teachers is often the result of a high level of emotional involvement combined with an increased workload. AI could prove to be an advantage for instructors and students. Educators face challenges on multiple fronts and are not only responsible for their students but the necessary economics and politics involved with out educational system (Campos & Lucena, 2017)
In AI the expert system acts as an intelligent being with access to a significant source of knowledge and experience by design (Cuiye, 2016). An expert system is used to simulate a human expert in a specific field or set of fields. Cognitive tutors are developed to act as a source of expertise that a student has direct and uninterrupted access to in lieu of the assigned educator.
Pairing this kind of system with natural language understanding or the ability to communicate with a computer using a natural language provides a unique opportunity to create a classroom assistant in the forms such as a chat-bot that the students can interact with. According to Cuiye (2016) two middle school classrooms were tested and showed that the classroom that interacted with the chat-bot scored higher than the controlled class on all four conducted exams over the course of a semester.
Tutoring is a discipline as old as education and many students have prospered under the specific guide of a tutor. However, tutors are often a luxury that many cannot afford for their children and within our educational system are only available for a period of time per student. Intelligent Tutoring Systems (ITSs) have been developed “...to use techniques from AI and cognitive science to attempt to understand the nature of learning and teaching and to build systems to assist learners to master new skills or understand new concepts in way that mimic the actions of a skilled human tutor working one-on-one with the learner” (Boulay, 2016, p. 76). Multiple variations on this concept have been researched and implemented with the two most common being step by step and student authored testing. In step by step systems, the tutor provides guidance to the student based on errors made during problem solving. Student authored testing provides the student with an opportunity to design the testing materials used by the software in a pre-programmed problem. Once the student has completed their design, the system then tests that design based on the data specific to that problem.
Testing of these kinds of systems has seen comprehensive success. A cognitive algebra tutoring system was used in a study consisting of both high school and middle school classrooms and over two years the system scored on par with human to human based tutoring (Boulay, 2016). Whether the systems were used in the best manner possible is up for debate due to the learning curve associated with the introduction of technology such as this, but the effectiveness of these systems has been proven under certain conditions. If these systems can assist classroom instructors through consistent interactions with students, the instructors can act as an escalation point for students requiring additional specialized assistance.
Cognitive digital assistants could significantly decrease the workload experienced by educators due to the overpopulation of classrooms allowing educators to address other responsibilities associated with their positions. The improved quality of life of our educators is paramount to the future of education and AI could assist in the reduction of stress associated with teaching positions.
Learning and Teaching Styles
The adoption of technology in the classroom has fostered a change in how we teach. Technology focused classrooms, online courses, study abroad programs, and blended classrooms have all created a teaching space where ensuring that each student is supported with their ideal plan of instruction is difficult (Almohammadi et al., 2017). Considering the speed at which students move throughout school systems, the ability to gain a formal understanding of the needs of each individual student is outpacing the ability of the educator.
“Adaptability in the context of e-learning represents the creation of students’ experiences under different circumstances (personal characteristics, pedagogical knowledge, his/her interaction, the result of the previous learning process) in a certain period with a tendency of increasing the predefined criteria of success (efficiency of learning: result, time, price, user’s satisfaction, etc.)” (Markovic & Jovanovic, 2011, p. 305). Fuzzy logic provides an opportunity to enhance the educators’ ability to determine which method of instruction provides a suitable form of adaptability for the student. In normal logic we have two states of being: True and False. Fuzzy logic acknowledges that there is room to negotiate in between these two states enabling the system to understand stages in the learning process. This is an extremely important factor in contextualizing the state of a students learning and predicting the best method of instruction for the student.
In a system such as this, not only can instructor methods provide data to the system but there are certain methods in which the state of mind of the student can be determined: sensors that can track the gaze of users to determine how data is being interpreted, skin sensors that can help determine a students emotional state, and chair sensors that can read into body movements to determine their level of engagement with the material (Almohammadi et al., 2017). Taking these inputs into account, the system is able to provide suggestions on how to improve the instructional methodologies implemented by the educator. As a result, educators are no longer forced to sacrifice the learning outcomes of individual students in an effort to streamline their instruction. Reaching each individual student is achieved through the process of guided instruction provided by an AI system that provides the best possible suggestions from student to student. Studies at Essex University have concluded that these systems have are effective in increasing student outcomes such as engagement and satisfaction level (Almohammadi et al., 2017). Consistent with the previous section, this system will not only improve student learning outcomes but provide adaptable methods of catering their instruction to a developing pool of students from year to year and doing so in less time.
Creating an International Learning Environment
As the world continues to internationalize with the development of the internet and instant communication, we have seen an increased need for professionals with an understanding of the international context. Miglietti (2015) states that “...business schools recognize that they must internationalize their curriculum to prepare students for work in the global economy” (p. 46). Additionally, international education promotes “...the mobility of scholars and their knowledge; dialogue between cultures; and academic freedom through intellectual inquiry and research-based teaching.” (Kearney & Lincoln, 2017, p. 823).
The idea of an international classroom is exciting as it removes the boundaries often faced in a traditional teaching model. Providing diverse viewpoints, especially cross-cultural viewpoints, can broaden the scope of a lesson in a way that allows students to better relate to an idea, theme, or context. In recent history, the closest we have been able to come to internationalizing the classroom is through the use of online collegiate education but offerings are often limited to local students and instruction is based on current classroom teaching models (Shahbazova, 2012).
Recent research in dictionary translation makes a case for the future of an integrated international classroom at the primary level. Current research is focused on the translation of only two languages through the use of mining repositories but the results are encouraging (Krajewski et al., 2015). While the technology is not developed to the point of live translation, the ability to automatically translate dictionaries could provide students with the opportunity to study abroad electronically. Imagine a classroom consisting of two or more cultures running on a unique syllabus with an educator representing either end of the world.
Semantic meaning is currently the most difficult hurdle for making this dream a reality. Current research has shown that the use of extensive repositories and seeding current translations improves upon current translations through iterative progression (Krajewski et al., 2015). While research is far from finished, creating an opportunity to educate on an international level will not only enrich the education of students but bring new opportunities for educators to develop their craft.
The success of our education systems depends on our teachers and our students. However, there has been a significant decline in educator satisfaction due to the challenges of an overwhelming workload and insufficient support for students in the classroom. AI can be a solution for those challenges and can provide new opportunities that promote the success of our compulsory institutions.
Currently, the most relevant source of AI support in the classroom is in the form of cognitive assistants. Through the development and implementation of tutors using advanced AI, we now have the ability to meet and exceed the results of student-educator instruction without placing the burden on the educator.
Learning styles and the international context in which education currently takes place are important factors in our pursuit of a better education for all. AI provides opportunities to improve learning style recognition, teaching style, and integrate many cultures into a single learning environment.
Future research will need to work on bridging the gaps between the problems facing educators today. From burnout syndrome to language differences among students at all levels of education, the increasing amount of time we spend researching how we can resolve issues in classrooms, the better the education and quality of life we can provide for students and educators alike. AI can be a force multiplier in the classroom if we can properly integrate it into classrooms.