NoMark 13 | - Freelance Writer
Dec 15, 2015 | #1
ABSTRACT
CHAPTER 1
INTRODUCTION
1.1.1 Learning Management Systems
Learning management systems (LMS) consist of software, which can be used for reporting, tracking, documentation, and administration of educational material (Graf). There are a number of elements, which are key to these types of systems. The software must be able to reuse knowledge. It must also deliver the information in different formats such as webinars or more personalized interactive learning. The learning system must include both self-guided services and content which can be delivered at a rapid pace. Effective learning systems will automate nearly all of the administration involved with online education. There now exist standards regarding the portability of the learning management systems. An important element of these systems which will be discussed further in a separate section is scalability (Kritikou et al).
An entire industry has developed around LMS technology. There is a wide variety of education applications, which are provided by commercial vendors. Some of these vendors have been around since the 1990s when the Internet first developed for public use, while others are relatively new additions. One of the leaders regarding the LMS technology is Blackboard. This company has more than 50% of the market. Other important providers are Desire2Learn and Moodle.
There are several online universities and colleges, which use the eCollege system. The system is a "software as a service" provider who caters to both post-secondary and secondary institutions of learning. Prior to being purchased by Pearson during 2007, eCollege was traded on the New York Stock Exchange beginning in 2002. Other large providers in this realm are Sumtotal Systems and SuccessFactors.
Nearly all the learning management systems are web-based in order to allow access of content and the administration in remote locations (Kritikou et al). These systems can be used by industries such as pharmaceutical research and financial services in order to ensure that professionals can keep apprised of any changes, which occur in regulations. These systems are widely used by educational institutions in order to support and enhance the more traditional courses offered in a classroom. Frequently, students in remote locations would otherwise be unable to access education offered by institutions in other countries. The systems make the learning available anywhere there is an Internet connection.
The majority of institutions which make use of learning management systems also take advantage of an authoring tool which allows educators to prepare e-learning content, which can be hosted on the system (Graf). This is generally in the form of authoring tools, which allow for content manipulation. If there is a need for more advanced levels of content creation, authoring software must be used, which integrates well with the LMS. Both the Shareable Content Object Reference Model (SCORM), as well as the Aviation Industry Computer-Based Training Committee (AICC) has standards related to authoring tools for LMSs (Graf).
It is important that learning management systems not be confused with learning content management systems (LCMS) (Kritikou et al). The LMS is used to deliver courses online and allow students to track their progress as well as performance. In these instances, the LMS also usually reports the progress of students to the instructors, carrying out administrative tasks, and performing a wide variety of content delivery services. The LCMS software is used to organize the digital information, so they can be used by the LMS (Kritikou et al.).
1.1.2 Learning Styles
Individuals learn in different ways (Kritikou et al.). These are due to the variance in how people process information and habitually acquire necessary knowledge. Learning styles became a popular topic of educational research beginning in the 1970s. Scholars who support the idea of learning styles explained that is necessary for teachers to determine the learning style of students in order to maximize their learning ability. Once the learning style has been determined, the information and manner of its presentation can be adjusted in order to help the student learn more rapidly and in depth.
A wide variety of models has been developed to represent learning styles. Nearly all people have a mix of styles which they use to learn. However, most people have certain types of learning, which are dominant in their approach. Some people use the different learning styles according to the circumstances or task.
One common model for understanding learning styles is to split them into the types of information, which are most efficiently processed by an individual. Using this scheme of understanding learning generally leads to seven styles. The styles are logical, solitary, social, physical, verbal, visual, and auditory.
Logical learning is sometimes referred to as a mathematical style. These people will generally be skilled at establishing connections and recognizing patterns (Moenikia). There are also efficient at grouping and classifying information in order to make it more easily understood. Individuals using a logical learning style are often able to do relatively complex calculations without the use of a calculator, computer, or even paper.
The solitary learning style is also known as an interpersonal style. These people learn best on their own. They tend to be independent and introspective. Most can concentrate well and are adept at analyzing their own feelings and thoughts regarding a topic. They are nearly always good at self-analysis. Many will take considerable time in order to assess situations and develop an opinion or solution.
A social learning style is often referred to as an interpersonal approach. These people learn best when they are able to communicate with others. This includes both nonverbal and verbal interactions. These individuals are generally good at evaluating the feelings and motivations of others. A common learning approach among these people is forming a study group.
The physical learning style involves making use of kinesthetic information. These learners will acquire information most efficiently when it is provided in the form of bodily senses. Touch is frequently an important part of how these people learn. A popular method of solving problems for these individuals is the taking of a long walk to ponder an issue.
Verbal learning makes use of both the spoken and written word. These people can express themselves well in both spoken and written forms (Moenikia). Learners who prefer this approach will have a large vocabulary and always be attempting to improve. Most authors and poets make extensive use of this linguistic approach to learning. Individuals who are fond of verbal learning will generally acquire new words or language is rapidly.
People who prefer a visual learning style will make use of spatial information. This type of information can include maps, colors, pictures, or images. They can visualize objects, as well as plans and possible outcomes mentally. They are usually good at maps and will not generally become lost.
The auditory learning style makes use of rhythm and music. These people will generally have a considerable skill at both rhythm and pitch. They are often quite good at playing a number of musical instruments. They may be subject to strong emotions when particular music is played.
When the seven learning styles discussed are used, they are usually understood as consisting of two dimensions. The first dimension is social or solitary. The second dimension includes verbal, physical, auditory, visual, and logical. This means that there are 10 dimensions of learning including combinations such as solitary verbal or social physical.
It should be noted that no single style of learning is superior to another (Graf, 2008). Skilled learners might use any of these styles or a combination of them.
1.1.3 Cognitive Traits
As part of a PhD in Information Systems, a study. was done to establish a cognitive trait model (CTM) which is both persistent and domain transcending. This means a student can use their individual cognitive traits as explained by the model for learning in a wide variety of situations. The CTM allows for learning systems to be finely tuned with regard to their system-wide adaptivity in the support of a student's cognitive traits.
The CTM is not a replacement for the performance-based models which are presently being used in LMSs. Instead, it is meant to complement performance-based systems, which are already being used in many LMSs. The performance-based models are used to record domain specific and student specific data. They also maintain a record of the adaptivity of the system. The CTM makes use of data regarding cognitive traits of the students and helps the system adapt to these individual cognitive traits. In other words, the performance-based models and the CTM can be used independently for a given LMS. However, it is likely that the best approach will be to combine the performance-based models with CTM.
The CTM conceptualizes cognitive traits as falling into the three general categories of working memory capacity, inductive reasoning ability, and divergent associative learning. The working memory capacity of a learner is also known as the short-term memory. This is the ability of the student to store information in a transient fashion. The working memory is different from memory, which stores information in a more permanent manner. This type of learning is referred to as long-term memory. An analogy for the human memory with regard to computers would reveal that the random-access memory (RAM) corresponds to the short-term memory. To continue the analogy the long-term memory would be the storage capacity of the computer on the hard drive.
Inductive reasoning ability is the process of using inductive logic to process information. With this approach, general propositions are evaluated, which have been formulated from specific examples. Induction allows the discovery of new aspects of knowledge based on those which already exist. This type of reasoning has led to numerous scientific discoveries.
Divergent associative learning occurs when a student establishes links between existing and new concepts. This approach to learning involves both divergent thinking and learning in an associative fashion. Associative learning involves making an Association between various stimuli or stimuli and behavior. This includes both operant and classical conditioning. A stimulus which was previously neutral is presented with another stimulus, and the two are later associated with each other. Divergent thinking occurs when an individual explores multiple solutions and compares these in order to arrive at a solution.
1.1.4 Scalability
Scalability refers to the ability of the process, network, for system to deal with an increasing amount of use (Gilmore & Tribastone). With regard to learning management systems, this would mean a growing number of teachers, students, or most likely both. It could also include adding additional courses and learning material. The system is understood to be scalable when it is relatively easy for it to adjust to higher levels of information processing and storage.
Scalability of a learning system involves a number of dimensions (Rapuano & Zoino). These dimensions include load, geographic, functional, and administrative. A learning management system which has load scalability will be able to contract or expand to handle the appropriate load. When there are more teachers and students on the site, this might involve additional processing capabilities. The learning system is said to have geographic scalability when it is useful and performs well in a variety of locations. The Internet is excellent for this type of scalability. The learning system has functional scalability when new courses, and the learning material can be added without excessive effort. The system is scalable regarding administrative functions when an additional number of organizations can make use of it.
The learning management system must be capable of scaling both vertically and horizontally. Vertical scaling will add resources to the system generally in the form of increased processing power. This is done at a single node within the system. This often will involve adding processing units to an individual computer. In the case of a learning management system, vertical scaling would be necessary if the system has varying demands regarding processing.
Scaling horizontally is sometimes known as scale out. This is the process of adding nodes within a system. If the learning system is using a distributed computing application, additional computers would be added. This we usually involve using a multitude of small computers rather than a single supercomputer. It can now be less expensive to use 20 PCs working together rather than a lone Cray supercomputer. Horizontal scaling involves adding another computer to the distributed system).
There are advantages and disadvantages with regard to using a single supercomputer to handle the learning management system versus multiple PCs (Rapuano & Zoino). Using several small computers increases the complexity of managing them is a solitary unit. There must also be complex programming models, which ensure that the latency and throughput between the nodes is appropriate. Historically, it was cheaper to purchase the single supercomputer rather than a multitude of small computers. However, the second decade of the 21st century has seen substantial decreases in the price of personal computers. This means it is frequently less expensive to purchase a multitude of small computers rather than a supercomputer.
The goal of this dissertation is to propose a model for a scalable learning management system which makes use of the learner's cognitive traits and learning styles. The system will be scalable so that it can handle both an increased number of users and increased processing demands. There will be a comparison of the advantages and disadvantages of using a single supercomputer versus distributed computing among personal computers. There will also be a description of how the system can be evaluated when it is complete.
This dissertation will use a systematic review of the literature (Pawson, Greenhalgh, Harvey, & Walshe) to achieve the goal of developing a system for evaluating the success of a scalable learning system in a large network. This type of review involves an exhaustive summary of the recent literature addressing the topic. Due to the progressive nature of the topic, only articles and books which have been published since 2009 will be used. The articles will primarily come from peer-reviewed journals. However, white papers and other technical publications will be used when appropriate. Information from websites will be used if there is evidence of proper references and due diligence regarding the accuracy of the information provided (Pawson et al).
In order to ensure that all the necessary information is used, information provided in gray literature will be used as well. The term "gray literature" is used by information science to refer to written material which cannot be obtained through more traditional channels such as monographs and published journals. This is usually because it is not easily accessed by the general public. However, this type of literature often contains valuable information. Examples of documents which would be considered as gray literature include white papers, working papers of research groups, government agency reports, technical reports, and patents. This information is sometimes not included in reports due to the difficulty of obtaining the material. At the same time, the Internet has made the gray literature substantially easier to obtain.
The systematic literature review will make use of Master's theses and doctoral dissertations These are sometimes included as gray literature, especially when they are unpublished works. These are also more readily available due to the Internet.
Multiple Internet search engines will be used to locate the data, including Google, Yahoo!, and Google scholar. Meta-search engines will also be used such as Dog pile and Ixquick.
The remainder of this dissertation is divided into chapters. Chapter 2 is the longest section and will consist of the systematic review of the literature. Chapter 3 presents the findings and a discussion of them with regard to how the literature relates to the goals and objectives of this dissertation. Chapter 4 is the final chapter and presents the conclusion which can be drawn from the findings and discussion of the literature review. This last chapter includes suggestions of additional research which may prove beneficial.
CHAPTER 2
LITERATURE REVIEW
CHAPTER 3
FINDINGS AND DISCUSSION
CHAPTER 4
CONCLUSION
References
Lin, T. Cognitive trait model for adaptive learning environments (Doctoral dissertation, PhD thesis, Massey University, Palmerston North, New Zealand).
Graf, S. Adaptivity in learning management systems focusing on learning styles (Doctoral dissertation, Vienna University of Technology).
Graf, S., & Liu, T. C. Identifying Learning Styles in Learning Management Systems by Using Indications from Students' Behaviour. In Advanced Learning Technologies. ICALT'08. Eighth IEEE International Conference on (pp. 482-486). IEEE.
Kritikou, Y., Demestichas, P., Adamopoulou, E., Demestichas, K., Theologou, M., & Paradia, M. User Profile Modeling in the context of web-based learning management systems. Journal of Network and Computer Applications,31(4), 603-627.
Moenikia, M. The role of learning styles in second language learning among distance education students. Procedia-Social and Behavioral Sciences,2(2), 1169-1173.
Gilmore, S., & Tribastone, M. Evaluating the scalability of a web service-based distributed e-learning and course management system. Web Services and Formal Methods, 214-226.
Rapuano, S., & Zoino, F. A learning management system including laboratory experiments on measurement instrumentation. Instrumentation and Measurement, IEEE Transactions on, 55(5), 1757-1766.
Pawson, R., Greenhalgh, T., Harvey, G., & Walshe, K. Realist review-a new method of systematic review designed for complex policy interventions. Journal of health services research & policy, 10(suppl 1), 21-34.
CHAPTER 1
INTRODUCTION
1.1 Learning Styles and System Background
1.1.1 Learning Management Systems
Learning management systems (LMS) consist of software, which can be used for reporting, tracking, documentation, and administration of educational material (Graf). There are a number of elements, which are key to these types of systems. The software must be able to reuse knowledge. It must also deliver the information in different formats such as webinars or more personalized interactive learning. The learning system must include both self-guided services and content which can be delivered at a rapid pace. Effective learning systems will automate nearly all of the administration involved with online education. There now exist standards regarding the portability of the learning management systems. An important element of these systems which will be discussed further in a separate section is scalability (Kritikou et al).
An entire industry has developed around LMS technology. There is a wide variety of education applications, which are provided by commercial vendors. Some of these vendors have been around since the 1990s when the Internet first developed for public use, while others are relatively new additions. One of the leaders regarding the LMS technology is Blackboard. This company has more than 50% of the market. Other important providers are Desire2Learn and Moodle.There are several online universities and colleges, which use the eCollege system. The system is a "software as a service" provider who caters to both post-secondary and secondary institutions of learning. Prior to being purchased by Pearson during 2007, eCollege was traded on the New York Stock Exchange beginning in 2002. Other large providers in this realm are Sumtotal Systems and SuccessFactors.
Nearly all the learning management systems are web-based in order to allow access of content and the administration in remote locations (Kritikou et al). These systems can be used by industries such as pharmaceutical research and financial services in order to ensure that professionals can keep apprised of any changes, which occur in regulations. These systems are widely used by educational institutions in order to support and enhance the more traditional courses offered in a classroom. Frequently, students in remote locations would otherwise be unable to access education offered by institutions in other countries. The systems make the learning available anywhere there is an Internet connection.
The majority of institutions which make use of learning management systems also take advantage of an authoring tool which allows educators to prepare e-learning content, which can be hosted on the system (Graf). This is generally in the form of authoring tools, which allow for content manipulation. If there is a need for more advanced levels of content creation, authoring software must be used, which integrates well with the LMS. Both the Shareable Content Object Reference Model (SCORM), as well as the Aviation Industry Computer-Based Training Committee (AICC) has standards related to authoring tools for LMSs (Graf).
It is important that learning management systems not be confused with learning content management systems (LCMS) (Kritikou et al). The LMS is used to deliver courses online and allow students to track their progress as well as performance. In these instances, the LMS also usually reports the progress of students to the instructors, carrying out administrative tasks, and performing a wide variety of content delivery services. The LCMS software is used to organize the digital information, so they can be used by the LMS (Kritikou et al.).
1.1.2 Learning Styles
Individuals learn in different ways (Kritikou et al.). These are due to the variance in how people process information and habitually acquire necessary knowledge. Learning styles became a popular topic of educational research beginning in the 1970s. Scholars who support the idea of learning styles explained that is necessary for teachers to determine the learning style of students in order to maximize their learning ability. Once the learning style has been determined, the information and manner of its presentation can be adjusted in order to help the student learn more rapidly and in depth.
A wide variety of models has been developed to represent learning styles. Nearly all people have a mix of styles which they use to learn. However, most people have certain types of learning, which are dominant in their approach. Some people use the different learning styles according to the circumstances or task.
One common model for understanding learning styles is to split them into the types of information, which are most efficiently processed by an individual. Using this scheme of understanding learning generally leads to seven styles. The styles are logical, solitary, social, physical, verbal, visual, and auditory.
Logical learning is sometimes referred to as a mathematical style. These people will generally be skilled at establishing connections and recognizing patterns (Moenikia). There are also efficient at grouping and classifying information in order to make it more easily understood. Individuals using a logical learning style are often able to do relatively complex calculations without the use of a calculator, computer, or even paper.
The solitary learning style is also known as an interpersonal style. These people learn best on their own. They tend to be independent and introspective. Most can concentrate well and are adept at analyzing their own feelings and thoughts regarding a topic. They are nearly always good at self-analysis. Many will take considerable time in order to assess situations and develop an opinion or solution.
A social learning style is often referred to as an interpersonal approach. These people learn best when they are able to communicate with others. This includes both nonverbal and verbal interactions. These individuals are generally good at evaluating the feelings and motivations of others. A common learning approach among these people is forming a study group.
The physical learning style involves making use of kinesthetic information. These learners will acquire information most efficiently when it is provided in the form of bodily senses. Touch is frequently an important part of how these people learn. A popular method of solving problems for these individuals is the taking of a long walk to ponder an issue.
Verbal learning makes use of both the spoken and written word. These people can express themselves well in both spoken and written forms (Moenikia). Learners who prefer this approach will have a large vocabulary and always be attempting to improve. Most authors and poets make extensive use of this linguistic approach to learning. Individuals who are fond of verbal learning will generally acquire new words or language is rapidly.
People who prefer a visual learning style will make use of spatial information. This type of information can include maps, colors, pictures, or images. They can visualize objects, as well as plans and possible outcomes mentally. They are usually good at maps and will not generally become lost.
The auditory learning style makes use of rhythm and music. These people will generally have a considerable skill at both rhythm and pitch. They are often quite good at playing a number of musical instruments. They may be subject to strong emotions when particular music is played.
When the seven learning styles discussed are used, they are usually understood as consisting of two dimensions. The first dimension is social or solitary. The second dimension includes verbal, physical, auditory, visual, and logical. This means that there are 10 dimensions of learning including combinations such as solitary verbal or social physical.
It should be noted that no single style of learning is superior to another (Graf, 2008). Skilled learners might use any of these styles or a combination of them.
1.1.3 Cognitive Traits
As part of a PhD in Information Systems, a study. was done to establish a cognitive trait model (CTM) which is both persistent and domain transcending. This means a student can use their individual cognitive traits as explained by the model for learning in a wide variety of situations. The CTM allows for learning systems to be finely tuned with regard to their system-wide adaptivity in the support of a student's cognitive traits.
The CTM is not a replacement for the performance-based models which are presently being used in LMSs. Instead, it is meant to complement performance-based systems, which are already being used in many LMSs. The performance-based models are used to record domain specific and student specific data. They also maintain a record of the adaptivity of the system. The CTM makes use of data regarding cognitive traits of the students and helps the system adapt to these individual cognitive traits. In other words, the performance-based models and the CTM can be used independently for a given LMS. However, it is likely that the best approach will be to combine the performance-based models with CTM.
The CTM conceptualizes cognitive traits as falling into the three general categories of working memory capacity, inductive reasoning ability, and divergent associative learning. The working memory capacity of a learner is also known as the short-term memory. This is the ability of the student to store information in a transient fashion. The working memory is different from memory, which stores information in a more permanent manner. This type of learning is referred to as long-term memory. An analogy for the human memory with regard to computers would reveal that the random-access memory (RAM) corresponds to the short-term memory. To continue the analogy the long-term memory would be the storage capacity of the computer on the hard drive.
Inductive reasoning ability is the process of using inductive logic to process information. With this approach, general propositions are evaluated, which have been formulated from specific examples. Induction allows the discovery of new aspects of knowledge based on those which already exist. This type of reasoning has led to numerous scientific discoveries.
Divergent associative learning occurs when a student establishes links between existing and new concepts. This approach to learning involves both divergent thinking and learning in an associative fashion. Associative learning involves making an Association between various stimuli or stimuli and behavior. This includes both operant and classical conditioning. A stimulus which was previously neutral is presented with another stimulus, and the two are later associated with each other. Divergent thinking occurs when an individual explores multiple solutions and compares these in order to arrive at a solution.
1.1.4 Scalability
Scalability refers to the ability of the process, network, for system to deal with an increasing amount of use (Gilmore & Tribastone). With regard to learning management systems, this would mean a growing number of teachers, students, or most likely both. It could also include adding additional courses and learning material. The system is understood to be scalable when it is relatively easy for it to adjust to higher levels of information processing and storage.
Scalability of a learning system involves a number of dimensions (Rapuano & Zoino). These dimensions include load, geographic, functional, and administrative. A learning management system which has load scalability will be able to contract or expand to handle the appropriate load. When there are more teachers and students on the site, this might involve additional processing capabilities. The learning system is said to have geographic scalability when it is useful and performs well in a variety of locations. The Internet is excellent for this type of scalability. The learning system has functional scalability when new courses, and the learning material can be added without excessive effort. The system is scalable regarding administrative functions when an additional number of organizations can make use of it.
The learning management system must be capable of scaling both vertically and horizontally. Vertical scaling will add resources to the system generally in the form of increased processing power. This is done at a single node within the system. This often will involve adding processing units to an individual computer. In the case of a learning management system, vertical scaling would be necessary if the system has varying demands regarding processing.
Scaling horizontally is sometimes known as scale out. This is the process of adding nodes within a system. If the learning system is using a distributed computing application, additional computers would be added. This we usually involve using a multitude of small computers rather than a single supercomputer. It can now be less expensive to use 20 PCs working together rather than a lone Cray supercomputer. Horizontal scaling involves adding another computer to the distributed system).
There are advantages and disadvantages with regard to using a single supercomputer to handle the learning management system versus multiple PCs (Rapuano & Zoino). Using several small computers increases the complexity of managing them is a solitary unit. There must also be complex programming models, which ensure that the latency and throughput between the nodes is appropriate. Historically, it was cheaper to purchase the single supercomputer rather than a multitude of small computers. However, the second decade of the 21st century has seen substantial decreases in the price of personal computers. This means it is frequently less expensive to purchase a multitude of small computers rather than a supercomputer.
1.2 Goals and Objectives
The goal of this dissertation is to propose a model for a scalable learning management system which makes use of the learner's cognitive traits and learning styles. The system will be scalable so that it can handle both an increased number of users and increased processing demands. There will be a comparison of the advantages and disadvantages of using a single supercomputer versus distributed computing among personal computers. There will also be a description of how the system can be evaluated when it is complete.
1.3 Methodology
This dissertation will use a systematic review of the literature (Pawson, Greenhalgh, Harvey, & Walshe) to achieve the goal of developing a system for evaluating the success of a scalable learning system in a large network. This type of review involves an exhaustive summary of the recent literature addressing the topic. Due to the progressive nature of the topic, only articles and books which have been published since 2009 will be used. The articles will primarily come from peer-reviewed journals. However, white papers and other technical publications will be used when appropriate. Information from websites will be used if there is evidence of proper references and due diligence regarding the accuracy of the information provided (Pawson et al).
In order to ensure that all the necessary information is used, information provided in gray literature will be used as well. The term "gray literature" is used by information science to refer to written material which cannot be obtained through more traditional channels such as monographs and published journals. This is usually because it is not easily accessed by the general public. However, this type of literature often contains valuable information. Examples of documents which would be considered as gray literature include white papers, working papers of research groups, government agency reports, technical reports, and patents. This information is sometimes not included in reports due to the difficulty of obtaining the material. At the same time, the Internet has made the gray literature substantially easier to obtain.
The systematic literature review will make use of Master's theses and doctoral dissertations These are sometimes included as gray literature, especially when they are unpublished works. These are also more readily available due to the Internet.
Multiple Internet search engines will be used to locate the data, including Google, Yahoo!, and Google scholar. Meta-search engines will also be used such as Dog pile and Ixquick.
1.4 Dissertation Organization
The remainder of this dissertation is divided into chapters. Chapter 2 is the longest section and will consist of the systematic review of the literature. Chapter 3 presents the findings and a discussion of them with regard to how the literature relates to the goals and objectives of this dissertation. Chapter 4 is the final chapter and presents the conclusion which can be drawn from the findings and discussion of the literature review. This last chapter includes suggestions of additional research which may prove beneficial.
CHAPTER 2
LITERATURE REVIEW
CHAPTER 3
FINDINGS AND DISCUSSION
CHAPTER 4
CONCLUSION
References
Lin, T. Cognitive trait model for adaptive learning environments (Doctoral dissertation, PhD thesis, Massey University, Palmerston North, New Zealand).
Graf, S. Adaptivity in learning management systems focusing on learning styles (Doctoral dissertation, Vienna University of Technology).
Graf, S., & Liu, T. C. Identifying Learning Styles in Learning Management Systems by Using Indications from Students' Behaviour. In Advanced Learning Technologies. ICALT'08. Eighth IEEE International Conference on (pp. 482-486). IEEE.
Kritikou, Y., Demestichas, P., Adamopoulou, E., Demestichas, K., Theologou, M., & Paradia, M. User Profile Modeling in the context of web-based learning management systems. Journal of Network and Computer Applications,31(4), 603-627.
Moenikia, M. The role of learning styles in second language learning among distance education students. Procedia-Social and Behavioral Sciences,2(2), 1169-1173.
Gilmore, S., & Tribastone, M. Evaluating the scalability of a web service-based distributed e-learning and course management system. Web Services and Formal Methods, 214-226.
Rapuano, S., & Zoino, F. A learning management system including laboratory experiments on measurement instrumentation. Instrumentation and Measurement, IEEE Transactions on, 55(5), 1757-1766.
Pawson, R., Greenhalgh, T., Harvey, G., & Walshe, K. Realist review-a new method of systematic review designed for complex policy interventions. Journal of health services research & policy, 10(suppl 1), 21-34.