Developing Socialmobile Learningapplication

Developing SocialMobile LearningApplication

on Android Platform

Seno Adi Putra1, Amelia Kurniawati2, Adityas Widjajarto3, KartikoEdhi Widodo4

1,3,4Information System Study Program, Telkom University

2Industrial Engineering Study Program, Telkom University

Jl. Telekomunikasi No.1, TerusanBuahBatu, Bandung 40257

Email :, , ,

Abstract

The evolution of smart phone and social media encourages mobile learning application transformation towards mobile social learning. Learning concept which is currently teacher centric has been shifted to the new paradigm in which learners can participate in developing learning process and content. Nowadays learning definition is changed to be a learning content generation and learners communication. Such learning concept is applied on social mobile learning.

Social mobile learning is an additional feature of social e-Learning, which is known as e-Learning 2.0. Social mobile learning can not be separated from social e-Learning. Collaboration, information and content sharing, and integration with other learning applications are the key features of social e-Learning. Now teachers no longer play a role as the center of learning sources but also as facilitator for learner in learning process. Thus, the research objective described on this paper is developing social mobile learning to enhance current e-Learning capabilities.

This research uses ADDIE methodology, leverages Java Android Technology as a platform of mobile device and Java EE as a platform on server side, and implements multitier architecture and RESTFUL web service. The overall result is that the system performs good functionality and processbetter capability in term of latency and throughput when it implementsmultitier architecture.

Key words: mobile social learning, e-Learning 2.0, ADDIE, Java Android, Java EE, multitier architecture, RESTFUL web service latency, throughput.

I. INTRODUCTION

One of strategies that can be considered for improving learning process quality in educational institution is enhancement of learning interactivity. A good learning method is the one in which students are engaged to participate in learning process, which is known as students-centered. One of technologies that can support this interactive learning process is social e-Learning.

The term of emerging technology such as open technology, sharing and collaboration paradigm, distributed computing, scalability, mobility, transparency, concurrency, intelligence, and simplicity has inspired the innovation of e-learning and m-learning application development.

Mobile technology both hardware and software are crucial components in the implementation of social mobile learning concept. The potency of this technology utilization is quite high. According to [5], mobile technology user has increased three times in 2008 and it is projected to reach 89 million in 2008, and 100 million in 2012. In 2015, China, India, and Indonesia will have the most mobile learning user after USA.

Today, smartphone has various platforms, such as Java Android, Symbian, Bada, iOS, and Windows Phone. According to [6], Android has dominant market share (61%). Thus, Android becomes the leader of smartphone in the world today. According to [5], In Indonesia Android has 52% market share in second quarter. For this reason, the research focused on mobile learning development with Android as a platform.

According to the background mentioned in the previous paragraph, the research objective is defined as learning management system enhancement featuring mobile social learning based on Android technology and Java Technology.

II. E-LEARNING 2.0 : NEW ONLINE LEARNING TECHNOLOGY

In many years, learning model which is used is face to face in class, with one way conversation from the teacher to student. Most e-Learning today just automates that process. The difference is only that in e-Learning, the teacher and student do not need to stay in the same place at the same time.But, even though much fund has been spent to develop interactive multimedia system, an important element has been disappeared, that is the existence of other people. This problem is handled by using blended learning. But often the face to face interaction is executed traditionally [2].

Cross [4]stated that learning process often happen in informal situation, such as observation on someone else, asking to peer, calling help desk, trial and error, or cooperate with someone else in understanding the material. Learning process academically also uses various informal channels, such as game, simulation, experiment, story, and discovery.

In the last view years,the role of web has been shifted from as a media to send information into platform as a place where the content is generated, shared, combined, and delivered (Downes, 2005, in [2]). This new generation has these characteristics: user-centric, open, dynamic web, with peer production, sharing, collaboration, collective intelligence, distributed content, and decentralized authority in the foreground.This web generation is calledWeb 2.0 (O’ Reilly, 2005 in [4]). Social software appears as the main component of Web 2.0movement (Alexander, 2006, in [4]). Social softwarecan be defined as tool to increase human social and collaborative ability, as a media facilitating social relationship and information change, and also as a ecology enabling a system which consists of people, practice, value, and technology in a certain local environment (Coates, 2003, in [4]).

Social software approach gives several advantages for online learning in compared with traditional approach. This approach reflects natural characteristics of learning, whichare social, personal, distributed, flexible, dynamic, and complex. Besides, this approach represents the shift of learning management system model to be more social, personal, open, and dynamic [4].

The interesting trend nowadays is the fast development in social network and mobility solution. Social network collaboration creates online communities and mobility to be favorable delivery channel, not only to improve return on investment, but also to expand the global coverage and improve operational efficiency of workers in company or institution [7].This trend brings out mobile social network enablement,which plants on-demand social networking and collaboration capacity into mobile application to improve its utility [7].

Mobile services and social collaboration applications not only provide social experiences to users but also create collective intelligence and develop new ways to gain opinions, perceptions, interests, and discussions with all communities. Hence, social network application is also expected to be integrated with all enterprise applications.

Social network and collaborative solutions could be applied in mobile learning method. Adoption of social network and collaborative solution will improve interaction between teachers and students or among students, expand learner network, and create information sharing culture.

According to [11], Social software implementation inspires development of new e-Learning generation, called e-Learning 2.0, featuring:

1. Social or collaborative learning Environment
2. Learning is about generating course contents and communicating with people.
3. Aggregating (RSS) and tagging
4. Knowledge sharing
5. Personalization
6. Collective intelligence (wisdom of the crowd)
7. Network usage with various technology.

III. RESEARCH METHODOLOGY

This research uses ADDIE instructional design model. As mentioned from its name, ADDIE defines five steps, those are analysis, design, development, implementation, and evaluation. According to Driscoll, 2002 in [12], ADDIE model is described as following Figure1 :

Figure 1 ADDIE Model

ADDIE ensures that learners will achieve learning goal, learning requirements evaluation is defines in each phase, therefore it will reduce cost and time by capturing and solving problem earlier. It also ensures that design and development are complied with learning material and learner’s culture by evaluating effectiveness of developed application more specific and measurable.

IV. M-LEARNING 2.0

Mobile learning design refers to National E-Learning eXchange Technology (NEXT), according to [1], which was sounded in APTIKOM congress in Bali. The concept is about a mechanism to learn from various sources, not just focused on reference given by teachers. This mechanism is called multisourcing concept.

The NEXT is in line with e-University concept. According [10], e-University concept is a paradigm in which the universities build the competitive advantage in global environment by leveraging information technology.

According to P.Mc.Questen in [3], E-University is defined as translation of key services, such as academic services and business process, into internet technology usage. E-University Framework, according to [9], consists of people, e-Business in the context of universities, information technology infrastructure, business process redesign, and e-Learning or mobile Learning.

Functional requirements of mobile learning 2.0 application built on this research are:

1. Users are joined in a study program group and university group.
2. Group Administrator must approve user who wants to join in a group
3. Lecturers who are responsible for a certain course subject could distribute content and generate course exam for students.
4. Learners may upload their own learning content, then authorized lecturers verify and validate it.
5. All users can write their comments and questions related to uploaded contents.
6. System displays interactive multimedia contents (video, presentation slide, and flash).
7. System provides evaluation feature to examine students understanding level of each material.

Point 2, 3 and 4 above are implemented on web-based learning management system meanwhile other points are implemented on mobile learning application.

There are two applications that are built in this research, learning management system and mobile learning. Learning management system application has four actors such as e-university administrator, university administrator, study program administrator, lecturers, and students. On the other hand, mobile learning application has students and lecturers as the only actor. Description of respective actors could be shown at figure 2 and 3.

Figure 2. Use Case Diagramon Learning Management System

Figure 2 describes use case diagram on learning management system. E-University administrator is the highest role in system which is designed to manage user list in the system such as changing role users, making a new user, and determining university administrator. University administrator is responsible for giving approval or rejection over other user request who wants to join in the university group. Meanwhile, university administrator could study program administrator. Study program administrator has the same responsibility as university administrator such as giving approval or rejection over the user who wants to join in study program group. The actor assigned for teaching is the lecturer who is responsible for verifying and validating learning content, exam and quiz, and evaluating the learning process.

Figure 3.Use Case DiagramonMobile Learning System

Figure 3describes use case diagram on mobile learning system. To initiate learning process, the students should join in university group then join in study program. The students could view the multimedia content, discuss in the multimedia contents uploaded by themselves, other students, or lecturers. Students also can do on-line quiz and exam.

V. MOBILE LEARNING TECHNOLOGY AND ARCHITECTURE

Android-based mobile application has adequate media format to be interactive media device. Interactive media run in Android could be saved in the local storage or network. Android has supported interactive media running via network for example by using HTTP/HTTPS protocol. This research leverages multimedia contents such as image (JPG), video (AVI), flash (SWF) which are placed in file server.

Architecture of mobile learning system implements multitier architecture by separating system into four main tiers that are service tier, presentation tier, business logic tier, and data tier. Service tier plays role as middleware that connects e-Learning server and mobile device. This tier is developed for interoperability purpose using RESTFUL web service. Presentation tier plays role as if a single system which connected directly to the browser client. It is generally a form of virtual host name that distributes client request to various e-Learning business logic tiers. This tier leverages Struts Framework that implements Model-View-Controller (MVC) architecture. Business logic tier performs main business process of e-Learning. Technology playinga role in this tier is Enterprise Java Bean (EJB). EJB is the objects whose ability to be called remotely. For such reason, EJB is the main component to build multitier and distributed application. Data tier is primary data storage tier such as database server. Database server used in this research is Microsoft SQL Server 2008 Enterprise Edition. Figure 4 describes the whole system architecture.

Figure 4. The Whole System Architecture of Mobile Learning System

Figure 5 describes management system design which shows model class or domain class. User Class has four sub classes that are AdminEUniversity, AdminUniversity, AdminEUniversity and Lecturer. Aggreration between University and User class and University and Department shows one-to-many multiplicity.

Figure 5. Model Classes of Mobile Learning System

Figure 6 is an example of sequence diagram that describes class interaction occurred in the join student’s approval process. The following is general description of the approval process:

1. Students submit a request to join in the group via mobile learning application.
2. University and study program admin will verify and validate students request and take a decision and/or confirmation whether the students are accepted or not to join in the group.
3. Before joining in study program groups, students must ask to join a university group of targeted study program.

Figure 6.Sequencediagram for Students Approval Process in Mobile Learning System

VI. SYSTEM DEPLOYMENT

In this research, mobile learning application is deplyed to smartphone with specification: Android 4.0 Ice Cream Sandwich, 1GHz Processor (ARM Cotex-A8), 1GB RAM, 7 inch TFT LED (1024 × 600) Capacitive Multitouch Screen, Internal Storage 4GB Nand Flash, Wifi 802.11b/g/n, Bluetooth 2.1, and Adobe Flash Player ver 10.3.

Presentation tier, business logic tier, and service tier are respectively deployed in instance of Glassfish 3.0. This research is configured by server virtualization. Server specification used in the deployment is : Processor Intel Xeon CPU E3-1230 V2 3.30 GHz (*CPUs) ~3.3GHz, Harddisk Drive 1TB, OS : Windows Server 2008 Enterprise R1 x64 Edition, and RAM 16 GB. The environment setup is performed by installing 3 virtual machines respectively for presentation tier, business logic tier, and service tier. The specification for respective tiers is OS Windows Server 2008 Enterprise Rq SP2 X64 Edition and RAM 2 GB. Apache Tomcat used for multimedia contents server and Microsoft SQL Server 2008 used for database tier are placed in another remote machine (Host Machine).

In general, the system functionality works well as shown in Figure 7a for Android device and Figure 7b for web browser.

(a)

(b)

Figure 7. Android based Mobile Learning (a) and Web based E-Learning (b)

VII. SYSTEM TESTING

The discussion below is defined on cost-benefit approach and nature of application. Cost-benefit approach is modeled in computational resources (RAM memory) and test-bed implementation.

Implementation Discussion

The implementation of both multitier and single-tier services is designed on campus intranet. For this time, campus intranet is built on both copper cable and fiber optic. On an intranet segment, measurement of experiment resulted average of intranet speed is 25 ms, in peak time. It uses 25 ms as average network time.

User acces is defined on local user. Local user accesses application service on LAN media without proxy server. Local user accessed both single-tier and multitier application model. The number of user ismodeled 1000 users (most number). It simultaneously accessed application services.

Hardware implementation model is defined on virtualization model. Limited computational resources problem is solved with best-practices strategy. In this approach, expansion of system capacity on limited resources is considered as a cost factor.

Nature of the Application

From testing phase, nature of application service is defined as follow. Nature of application service is based on system response within computational resources, i.e. server RAM.

It is resulted two graphics of system response. Throughput graphic, shown in Figure 8a, varies up to 1000 users accessed application services. Then response time (in ms), shown in Figure 8b, varies up to 1000 users. Both graphic reflects how the application services reacted to variably users requests. The application service is defined both in multitier and single tier model.

Single tier model had transitional and saturation phase. Transitional phase, up to 400 users, reached response time about 50 ms. It is under time limit of 1 second, and system is reacting instantaneously [8] . The system still had available resources, before saturation phase.

(a) (b)

Figure 8. Throughput (a) and Response Time (b) of m-Learning application

Saturation phase interval varied between 350-400 ms, with number of users is about 450 users. Response time is about 50 ms.It assumed that 50 ms is network time, then in 0.1 ms, system is reacting instantaneously, in human perceptual ability.

In multitier model, simulation conducted at variable rate of user request. At about 450 users, response time exceeded 50 ms. With limited resources, 2 GB RAM, response time reached up to 350 ms, if total user is about 600 users. In multi-tier system, it added 4 GB RAM, the response time became less, 300 ms.If the response time, by design, is up to 350 ms, the user could be limited about 600 users. The objected number, 1000 users, could not be reached in this simulation result.

Nature of application is based on use-case diagram. Classification of application services is modeledin informational and transactional type. In this e-learning application, informational type referred to services such as: news browsing, examination result browsing, problem solution and explanation browsing. Transactional type, in this e-learning application, referred to updating process of student’s examination.

If the present measurement and testing focused on informational web page, then for discussion forum, chat room for lecturer and student menu or other interactivity services should be considered as whole application services. For whole application services, it should consider response time under 10 s, at 600 users who access services simultaneously.

Expansion of system capacity is defined on system memory. It should allocate about four times of RAM capacity, based on single-tier model. In this e-learning application mobile service, single-tier model is a feasible implementation strategy based on benefit for about 600 students.

VIII. CONCLUSION AND SUGGESTION

There are two applications that are built in this research, learning management system and mobile learning. Java Enterprise Edition technology is implemented on Learning management system, whereas Java Android on mobile device. A sociallearning is implemented by featuring university grouping, study program grouping, and discussion or collaboration among users in developing learning contents.