Providing Personalized E-Learning Content Services by Adapting to

Multiple Feedback using Mobile Agents

Ashish Jain and Jean George

Department of Computer Applications, BSSS College, Bhopal, India

Keywords: Personalized E-Learning Systems, Dynamic Service Oriented Architecture, User Profiling, Mobile Agents.

Abstract: Personalized e-learning services are becoming increasingly crucial for delivering quality content to

subscribers. In open and dynamic environments, personalized e-learning services are essential, and emerging

technologies such as web services and agents are playing a significant role in their development. User profiling

is a promising approach to personalizing e-learning services, as it allows for the assessment of user profiles

during the learning process. By collecting feedback from users and storing it in their profiles, the proposed

system aims to provide efficient e-learning content services that can adapt to multiple feedbacks and cater to

both individual and group users. Mobile agents are used for effective retrieval and distribution of learning

content. The proposed system aims at providing personalized e-learning content services which can adapt

multiple feedbacks and provides efficient e-learning content services to individual users and to group of users

using mobile agents for effective retrieval and distribution of learning content.

1 INTRODUCTION

Personalization is important in today's service-

oriented society, and it has proven to be critical for

Internet and mobile telecommunications network

service acceptance. Because of the widespread

adoption of e-learning in all environments, scientific

research in the field of adaptive and intelligent

systems has been stimulated in order to provide high-

quality services to e-learning end users. Adaptive and

intelligent systems are those that meet the high

demands for personalization of e-learning users.

Personalization can determine a user's characteristics

based on previous purchases, products, or pages

viewed, or both. Adaptive and dynamic technologies

are required. The traditional, static "One size fits all"

approach provides all learners with a single set of

learning resources. Because the conditions that

determine which part of the educational material is

appropriate for different learner characteristics were

not described, learning contents delivered using this

approach are not feasible. As a result, providing a

personalized e-learning system that can automatically

adapt to learners' interests and levels is critical.

User profiling is a promising approach to

personalized e-learning systems that assesses a user's

interests, levels, and learning patterns while they are

learning. Based on the profile, a personalized learning

resource could be generated to match the individual

preferences and levels. Additionally, learners with

similar interests and levels can be grouped, and one

person's feedback can be used to guide information

delivery to other members of the same group [13]. To

ensure interoperability and scalability, the proposed

system uses a service-oriented approach that

encapsulates learning content within a web service,

and all system components are implemented as web

services.

The system is based on IEEE LTSA (Learning

Technology Systems Architecture) and includes a

chat room, a customized web browser, and a white

board for information delivery. These comments can

be used to determine the skill level and expertise of a

user. These various feedback measures, such as

reading time, the number of scrolls and prints, and the

relational index on chatting history, will be combined

by a feedback service. The collaborative filtering

algorithm is used by mobile agents to deliver

personalized information to learners by collecting

user profiles that store user preferences and levels of

expertise. Feedback can be tailored to individual

users based on their characteristics, as well as to

groups of users based on their shared characteristics.

620

Jain, A. and George, J.

Providing Personalized E-Learning Content Services by Adapting to Multiple Feedback Using Mobile Agents.

DOI: 10.5220/0012610200003739

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Artiﬁcial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics (AI4IoT 2023), pages 620-623

ISBN: 978-989-758-661-3

Figure 1: Architecture of the proposed system.

2 RELATED RESEARCH

As part of the related research, personalized service

selection and personalization of an e-learning system

are discussed. Also briefly discussed are e-learning

services with agents, adaptive e-learning systems, and

feedback adaptation.

Personalized Selection of Services: Users cannot be

expected to browse all services until they find an

adequate match, given the growing diversity of

services. As a result, the task of service selection and

composition must be supported by recommending a

set of possible services or combinations from which

users can then choose. Expanding the query for a

service with preferences can essentially improve

selection quality by excluding insufficient services.

Personalized E-Learning System: A personalized

e-learning system is one that can adapt to the interests

and levels of learners automatically. In order to infer

user preferences, a feedback extractor with fusion

capability is proposed. The user profiler collects user

profiles, which store user preferences and levels of

expertise, in order to deliver personalized information

using the collaborative filtering algorithm. A

personalized learning resource could be generated

based on the profile to match the individual

preferences and levels. When the learner interacts

with the system, the feedback extractor extracts the

learner's preferences and level of expertise and stores

them in the user profile. The user profiler collects user

preferences from the feedback extractor and provides

learning content to the learner based on those

preferences.

E-Learning Services with Agents: Victor

Pankratius proposed a distributed, service oriented

architecture for Web-based e-learning systems, as

well as extensions to support software agents. The

architecture's benefits and the use of intelligent

software agents for distributed retrieval of

educational content are also discussed. Nasir Hussain

proposed an e-learning service-oriented architecture.

The architecture is built with web service-based

intelligent agents that allow users to consume each

other's educational services in e-learning networks,

enhancing the learning experience of individual

learners, teachers, and authors through user agent

interactions.

Adaptive E-Learning: The adaptation of e-learning

content to provide personalized content to the user is

known as adaptive E-learning. In the business world,

a conceptual model is used to provide adaptive e-

learning to employees' educational needs. It can be

provided by adapting feedback and providing the user

with personalized content. Feedback can be tailored

at both the individual and group levels.

Providing Personalized E-Learning Content Services by Adapting to Multiple Feedback Using Mobile Agents

621

Conceptual Model for Adaptive E-Learning: A

conceptual model is required for developing a

personalized adaptive e-learning system in a business

environment for identifying employees' educational

needs from formal and semi-structured data and

selecting an appropriate learning strategy based on

that data. This model has been proposed as a general

model for employees' continuous learning in a

dynamic business environment (Trelepis and

Stephanides, 2021).

Feedback Adaptation: A system is described that

fulfils learning objectives by automatically selecting

and integrating appropriate learning materials for a

learner using web services based on the learner's

initial knowledge, goals, preferences, and so on.

Services appropriate to achieving a specific learning

goal can be dynamically selected, composed, and

invoked based on the learning goals as well as web

services. Feedback can be tailored and

personalization can be provided at both the individual

and group levels. The concept of group adaptation

implies that the system adapts feedback to common

characteristics of a group of user

3 PROPOSED APPROACH

In this paper it is proposed that e-learning services are

personalized by using multiple feedback measures

and tailoring feedback to individual and group users

via content retrieving mobile agents. Knowledge

about a user derived from interactions with e-learning

systems is stored in user profiles, which are then used

to adapt and guide a learner through offered e-

learning resources. The presented contribution

introduces a scalable implementation architecture

based on mobile agents. This platform is used to

manage the execution of the various mobile agents

that are used to support legacy e-learning systems. In

addition, web services technology is used to enable

communication and interoperability between mobile

agents and e-learning systems. The proposed e-

learning system architecture, depicted in Figure-1,

consists of following three layers:

The Presentation Layer is made up of the learner

service. The application interface is the learner

service. In the service layer, the feedback service

interacts with the learner service to collect multiple

feedbacks from the learner.

The Service Layer consists of feedback service,

profiler service, adaptation service, and content

retrieving mobile agents comprise the service layer.

The feedback service collects and stores multiple

feedbacks from the learner service in the profiler

service. The adaptation service will retrieve feedback

from the user profiler service and adapt it. Based on

multiple feedbacks stored in the user profile, the

content retrieving mobile agents will deliver

personalized information to individual users and

groups of users.

The Database Layer is made up of the profile

database and the e-content ontology. The profile

database stores the user's preferences and level of

expertise. The e-learning courses and contents will be

saved in the knowledge base of e-content ontology.

The e-content ontology database stores the courses

that can be delivered to the user, such as software

engineering, operating systems, C++, and computer

architecture, among others. Some data sources are

also saved, such as textbooks, notes, and tutorials.

The proposed system's operation is as follows: First,

the feedback service will determine the learner's

preferences, behaviour, and levels of expertise from

the learner service. The feedback service will save

current learner information in the user profile, such as

preferences and expertise. The profiler service will

extract user preferences and multiple feedbacks from

the user profile and provide the collected information

to the adaptation service. The adaptation service will

process the various feedbacks and forward them to

the content retrieving mobile agents.

By combining multiple feedbacks, these agents will

retrieve content from the e-content ontology.

Personalized information is stored in the profile

database and delivered to individual users as well as

groups of users who share common characteristics or

a stereotype. Means tailored to each student and his

or her unique (combination of) characteristics. Active

users could receive immediate feedback in a brief

format, while reflective learners could receive

detailed elaborated feedback. The feedback could

include references to the previous course (if the

student has passed it) or the detailed explanation.

4 USER PROFILE DATA

The information gathered about the user can be saved

in the user profile. These data are divided into two

categories: static data and dynamic data. Static data

are those that do not change during the interaction

between the student and the system. The dynamic

data are those that change as a result of the student's

learning progress and system interaction.

AI4IoT 2023 - First International Conference on Artiﬁcial Intelligence for Internet of things (AI4IOT): Accelerating Innovation in Industry

and Consumer Electronics

622

The static data is divided into five sections:

Personal, Personality, Cognitive, Pedagogical, and

Preference data are included. The dynamic data

consists of two data sets. They are Student Knowledge

data and Performance data. Data is constantly

gathered in order to keep an up-to-date record, and

this data can be gathered from student-system

interactions.

Multiple Feedbacks which are stored in user

profile can be adapted to provide personalized e-

learning service to the users. Individual Feedback

Adaptation means that feedback is tailored to each

student and his or her unique set of characteristics.

The timing and manner of feedback delivery could be

tailored to these individual characteristics.

Traditionally, group feedback adaptation is

performed on the basis of a group for stereotype

model.

The primary goal of stereotype modelling is to

model a group of users so that they can be adapted to

as a group of users. Group adaptation is carried out in

accordance with the characteristics of those groups.

The parameters that characterize the user's personal

preferences, interests, goals, habits, and mood are

included in user preferences. Users may prefer some

links or parts of the pages over others and this can

influence the adaptation of the feedback in e-learning.

For example, the user can prefer to receive feedback

in a pop-up window.

5 CONCLUSIONS

The assistance Personalization is the ability to tailor a

service to a specific user's needs. Personalization of

services refers to tailoring services to a user's or a

group of users' needs and preferences. Learners

require personalization of e-learning services, which

is especially important in open and dynamic

environments. Because personalization is carried out

with multiple feedbacks, the proposed e-learning

services personalization by using multiple feedback

measures will result in better performance. These

multiple user feedbacks are stored in the user profile,

which is a promising approach to personalizing e-

learning services.

Previous research on personalization of e-learning

services using implicit and explicit feedback has

many drawbacks, such as user profile input,

personalization using fewer measures, and

personalization based on only pages browsed. These

drawbacks are overcome by our approach, which

employs multiple feedbacks for personalization and

allows these feedbacks to be tailored to the individual

user and group of users by delivering personalized e-

learning content via content retrieving mobile agents.

The agents and web services mobility makes the

architecture robust, scalable, and efficient, and

services can be provided to any legacy e-learning

environment.

REFERENCES

Ekaterian Vasilyeva et al (2020), Adaptation of Feedback

in E-learning system at Individual and Group level.

Govindan Ravindran, Muhammad Jascemudin, Abdallah

Rayahan (2022), AManagement for Service

Personalization.

Jiaqian Zheng, Jing Yao, Junyu Niu (2021), Web User De-

Identification in Personalization, Proceedings of the

International Conference on World Wide Web,

Beijing.

K. K. Thyagharajan, Ratnmmanjari Nayak (2019),

Adaptive Content Creation for Personalized E- learning

Using Web Services, Journal of Applied sciences

Research, Vol 3, Issue 1

Lilia Cheniti-Belcadhi et al (2019), Implementation of a

Personalized Assessment Web Service, Proceedings of

the 6

IEEE International Conference on Advanced

Computing Technologies.

Nicola Henze, Daniel Krause (2021), Personalized Access

to Web Services in the Semantic Web.

Okkyung Choi, SangYong Han (2022), Personalization of

Rule-based Web Services.

Owen Conlan, Vincent Wade (2020), Evaluation of

APELS- An Adaptive E-learning Service based on the

Multi-Model, Metadata-driven approach, White Paper.

S. Trelepis, G. Stephanides, (2021), A Conceptual Model

for Developing a Personalized Adaptive E-Learning

System in a Business Environment.

Won Kim, (2019), Starting Directions for Personalized E-

Learning, Springer-Verlag Berlin.

Xin Li, Shi-Kuo Chang (2021), A Personalized E-learning

System Based on User profile Constructed Using

Information Fusion.

Xinyou Zhao (2020), Personalized Adaptive Content

System for Context-Aware Mobile Learning,

International Journal of Computer Science and

Network Security, Vol 12, Issue 3.

Yacine Sam, Omar Boucelma (2020), Dynamic Web

Services Personalization, Proceedings of the 10

IEEE

International Conference on E-commerce Technology.

Zakaria Maamar, Emad Bataineh (2020), Context Meets

Web Services Personalization, 4

International

Conference on Innovations in Information Technology.

Providing Personalized E-Learning Content Services by Adapting to Multiple Feedback Using Mobile Agents

623