Academy of Strategic Management Journal (Print ISSN: 1544-1458; Online ISSN: 1939-6104)

Research Article: 2021 Vol: 20 Issue: 4

The Recommendation System Development of Student Registration with Collaborative Filtering

Piyawat Tratsaranawathin, King Mongkut’s University of Technology North Bangkok

Tanawat Jariyapoom, King Mongkut’s University of Technology North Bangkok

Abstract

Background: Recently, most students register for classes without any information and options to support their decision because the systems used do not recommend the registration instructions of each class and the number of students is increasing. Moreover, most teachers have a lot of and limited tasks, and most office staff are responsible for paper documents which can be easily lost and the right time of making an appointment between teachers and students is also limited.

Aims: The purposes of this research were 1) to design the recommendation system of student registration with collaborative filtering, 2) to develop the recommendation system of student registration with collaborative filtering, and 3) to assess the competency of recommendation system of student registration with collaborative filtering.

Findings: The results of the research revealed that the research instrument used was the model of assessment system. The sample group of subjects was the students of King Mongkut’s University of Technology North Bangkok with simple random sampling. The data was analyzed by simple statistics: percentage and standard deviation (S.D.).

Conclusion: This research could be concluded that the success found that 1) the participants of the recommendation system of student registration with collaborative filtering were divided into three groups: students, teachers and office staff, 2) the recommendation system of student registration with collaborative filtering consisted of four development phases: planning, analysis, design, implementation, and 3) the overall efficiency of the recommendation system of student registration with collaborative filtering was at a high level (xËÂ? = 4.51 ) and the standard deviation (S.D.) was of 0.61. This showed that the performance of the recommendation system was efficient.

Practical Implications: The students had the guideline for course enrolment through the smart registration system and the system could enhance the student’s learning outcomes. The teachers could reduce the procedures and time of consultation or finding some more information to enhance the students’ learning efficiency.

Originality/Value: This research studied the recommendation system of student registration with the data of the students’ learning outcomes by implementing the theory of data recommendation and the collaborative filtering to recommend and predict the students’ learning outcomes.

Keywords

Recommendation System, Collaborative Filtering.

Introduction

Today education is essential for students at all levels. The information and communication technology (ICT) or digital technology is implemented to enhance the learning competency. This is in accordance with the Strategy of Ministry of Education (2020-2022) stating in Part 5 that digital technology of education must be developed and promoted and the learners must access the digital platforms to meet the most updated learning (Bureau of Policy and Strategy, Office of Permanent Secretary, Ministry of Education, 2020).

Therefore, the division of registration and evaluation, which supports and promotes the teaching and learning tasks of the institutions, needs to apply the technological knowledge for registration system of added subjects to meet the office staff and all involved personnel’s needs. Additionally, it is necessary to develop and improve the service quality all the time (Musiri et al., 2019). Moreover, the recommendation system of student registration is implemented to access the smart education information effectively.

At present, most students register for classes without any information and options to support their decision because the systems used do not recommend the registration instructions of each class and the number of students is increasing. Moreover, most teachers have a lot of and limited tasks, and most office staff are responsible for paper documents which can be easily lost and the right time of making an appointment between teachers and students is also limited. Therefore, the system with a calculating formula was developed to support the operations of each individual, to help decision making, and provide some more channels of consulting and recommendations, especially collaborative filtering based on the former data to analyze and recommend the information, expectations or possible proximity to provide the information or information recommendations to the users for decision making (Bobadilla et al., 2013).

In this research, the researcher implemented the collaborative filtering to develop the recommendation system with two functions: similarity calculation and learning outcome prediction to help the students find the recommendations of contents with the students’ learning outcomes and grades predicted to enhance the accuracy of the system.

Due to the problems mentioned above, the researcher had the idea of creating and developing the recommendation system of student registration with the collaborative filtering to solve the problems of registration to analyze the educational data, calculate the proximity, predict, and create the instructions of registration to enhance the learning proficiency.

Objectives

The objectives of this research were 1) to design the recommendation system of student registration with collaborative filtering, 2) to develop the recommendation system of student registration with collaborative filtering, and 3) to assess the competency of recommendation system of student registration with collaborative filtering.

Scope of the Study

The population was the students of King Mongkut’s University of Technology North Bangkok who used the recommendation system of student registration.

The sample group of subjects was 30 students of King Mongkut’s University of Technology North Bangkok with simple random sampling.

Literature Review

As mentioned above, there are three groups of people involved in the implementation of the recommendation system of student registration with collaborative filtering: student, teacher, and office staff. The system with theory of data recommendation consists of two formulas: similarity calculation and learning outcome prediction. The system selects the objects based on the students’ satisfaction or the similar objects and operates as follows: 1) to make the students’ profile based on the characteristics or qualifications of the objects calculated by the movement of the students’ previous activities, 2) to calculate the relationship between the student profile and the objects, 3) to adjust the weighting scale of each qualification of the objects, and 4) to select the objects most related to the students by focusing on the necessity of the users in each qualification and weighting scale of the objects on the system. This is in accordance with the study on the title “Challenges of Entrepreneurship E-education: Evidences from a Developing Country” of Jafari Moghaddam et al. (2012) stating that the e-learning system is implemented on the recognized universities in Iran to meet the entrepreneur needs in the educational business. There are three existing groups of driver and barrier factors of e-education at the Faculty of Entrepreneurship, based on a system approach, as follows: 1) input factors (students, professors, curriculum, ICT infrastructures, staffs and budget), 2) process factors (teaching methods, cultural and structural factors at university), and 3) environmental factors. And this is also in accordance with the study on the title “Entrepreneurial Characteristics: Insights from Undergraduate Students in Iran” of Salamzadeh et al. (2014) stating that there are eight entrepreneurial characteristics namely: 1) open mindedness; 2) need for achievement; 3) pragmatism; 4) tolerance of ambiguity; 5) visionary; 6) challenge taking; 7) risk taking; 8) internal locus of control.

Methodologies

The research methodology was the system development life cycle (SDLC) according to the four development phases: planning, analysis, design, and implementation (Jariyapoom, 2018).

Planning Phase

1. Planning was the first phase to develop the recommendation system of student registration with collaborative filtering. The developer must have a sound understanding on the system to meet the necessity of creation and development in use. The researcher, therefore, divided the people involved in the system into three groups according to the following steps: 1) studied the data of the recommendation system of registration and the theories related to collaborative filtering, 2) analyzed and synthesized the data, 3) created the interview model with the system components and needs, 4) collected the system needs from the system specialists, 5) summarized the data of the system, 6) analyzed the system, 7) designed the system, (8) developed the system (9) tried the system, and 10) implemented the system. The subjects involved were divided into three groups as follows:

2. Students: The students can access the system and register the assigned courses and then the system will process the GPA through the prediction formula. After that the students can make an appointment with the counselor through the calendar on the system and consult with their counselor through other channels and then the students can register and confirm their registration.

3. Teachers: The teachers can check the schedule of student registration on the system after that they can make an appointment with their students through the calendar on the system. After the appointment is confirmed, the teachers will provide the recommendations to the students through various channels both online and offline and then the stunts can register and confirm their registration.

4. Office staff: The office staff can add, delete, and correct all the data on the system and also check the schedules of students and the results of student registration.

Analysis Phase

The data analysis was done after the planning phase. The researcher analyzed the data with the previous and current recommendation systems as shown in Table 1 below.

Table 1 The Analysis of Original and Current Operational Systems
Subjects involved Original operational system Current operational system
Students - find the information of each course
- register on the system
- make an appointment with the counselor for recommendation of registration at the right time and meet the counselor at the university to ask for recommendation.
- discuss about the registration with their counselor
- add or withdraw the courses to meet the student’s need and get the counselor’s signature
- access the ready system to register
- The system analyzes and processes the data with collaborative filtering in the database.
- make an appointment with the counselor for recommendation of registration online or face to face
- Both the counselor and the student discuss on the registration for at least 5 courses.
- can add and change the courses and confirm the registration immediately
Teachers - make an appointment with the students
- provide some recommendation when the students ask for
- study the information of each course
- help the students to select the courses to register
- the calendar available in the system to make an appointment at the right time
- There are a lot of channels of recommendations both online and face to face.
- The data is analyzed and processed with collaborative filtering in the database.
- At least 5 courses are analyzed to select for registration.
Office staff - follow the paper work from the first step until the final step - follow the computer system, website and internet network from the first step until the final step

Table 1 shows the data analysis of the original and current operational systems and three people involved: students, teachers and office staff. The system operates through the implementation of the students’ relationship management, database system, data cloud, internet and communication network, and also the tools to comfort the learning of the students and teachers.

Design Phase

1. The design phase was the step that the developer of recommendation system of student registration with collaborative filtering must design the new system after the data analysis phase to show the relationship of the structure on the system as follows.

2. The office staff can manage and check the students’ data, course details, contents of the lessons, appointment calendar, face to face recommendations, online recommendations, recommendation system of student registration, and electronic bulletin board.

3. The teachers can check the students’ data, course details, contents of the lessons, appointment calendar, face to face recommendations, online recommendations, recommendation system of student registration, and electronic bulletin board.

4. The students can check their personal information, course details, contents of the lessons, appointment calendar, face to face recommendations, online recommendations, recommendation system of student registration, and electronic bulletin board.

Implementation Phase

1. The implementation phase was to create and enhance the system to meet the requiement and consistency of the system design. At this step, all parts of the program were developed for implementation and it could be explained as follows.

2. After the codes were put in the system, the three groups of people involved including students, teachers, and office staff were authorized to access the system. The development of the recommendation system of student registration with collaborative filtering consisted of two formulas of programming, i.e., similarity calculation and learning outcome prediction using the language and data of the Code Editor program. After that, the new system was designed and developed by using the programming languages of HTML, MySQL and jQuery to enhance the total operations of the recommendation system of student registration with collaborative filtering. The researcher implemented two formula of the system development: similarity calculation and learning outcome prediction to help the students see the results of their learning outcome prediction. The system could show the results at least five subjects as shown in Figure 1 below.

Figure 1 Learing Outcome Prediction

Results

After the recommendation system of student registration with collaborative filtering was developed through the formulas of similarity calculation and learning outcome prediction, the data, outputs and the system performance were displayed in accordance with the following objectives.

Objective 1, the recommendation system of student registration with collaborative filtering based on the architectural design with the system performance and all the other parts related to the system as shown in Figure 2 below.

Figure 2 Architectural Design

Figure 2 shows that the recommendation system of student registration with collaborative filtering was developed for the operations of the three groups of the authorized people: students, teachers and office staff.

The students can access the recommendation system of registration and register the required courses. And then, the system processes the registration through the collaborative filtering with the two formulas of similarity calculation and learning outcome prediction. When the system finishes the registration calculation, the results of the recommendation system of student registration are displayed. After that the students can make an appointment through the appointment calendar with their counselors and ask for some more information from their teachers through channels of both online conference and face-to-face consultation. They can register and confirm the registration in the system.

The teachers can access the system by using their own code provided and check the students’ registration calendar of the recommendation system of student registration with collaborative filtering. And then, the system processes the registration through the collaborative filtering with the two formulas of similarity calculation and learning outcome prediction. When the system finishes the registration calculation, the results of the recommendation system of student registration are displayed. After that the teachers can confirm the appointment with the students through the appointment calendar. Then the teachers can provide recommendations to the students through both online conference and face-to-face consultation. They can register and confirm the registration in the system.

The office staff can add or delete or edit the recommendation system to facilitate the users including the students.

Objective 2, after the recommendation system of student registration with collaborative filtering was developed with the two formulas of similarity calculation and learning outcome prediction. The students can register as usual through the operational system. Before the recommendation system of student registration processes, the learning outcome prediction database and course database are available on the system. When the students register, the system processes the students’ registration and displays the recommendation of at least five courses.

Objective 3, the researcher used the purposive sampling method to divide the users into three groups: students, teachers, and office staff. The questionnaire with five-level Likert scale was implemented and the results of the system quality were evaluated by the seven experts as shown in Table 2 below.

Table 2 Results of the System Quality Evaluated by the Experts
Evaluation items Quality level
  S.D.
Overall 4.51 0.61
1. Performance Test 4.43 0.79
2. Functional Test 4.57 0.53
3. Reliability Test 4.71 0.49
4. Usability Test 4.57 0.53
5. Security Test 4.29 0.76

Table 2 shows the results of evaluating the quality of the recommendation system of student registration with collaborative filtering. It revealed that the overall mean of the system quality was of 4.51 and the standard deviation (S.D.) of 0.61. Therefore, the quality of the recommendation system of student registration with collaborative filtering technique was at the highest level.

Discussion

In this research, the researcher developed the recommendation system of student registration with collaborative filtering and the results of this research were discussed as follows.

The Development of Registration System and Data Processing

The current situations cause the social distancing and the government manipulates the policy to enhance the quality of the communication and internet projects for the better online access and benefits of both students and teachers. This is in accordance with the study of Jareonsettasin (2020) stating that the communication system management during the crisis of situations is necessary for the development of the future educational management and the design of educational administration after the critical situations. Therefore, effective communication and fast operations are important for all the sectors to get the highest benefits (Thongkaew, 2020) and this is also in accordance with the study of Ausawawiwatkul & Kijmee (2013) on “The Development of Registration and Data Processing of Chiang Mai Commercial Technology College” stating that problems of registration and data processing of Chiang Mai Commercial Technology College are the redundant works and multiple operational steps which can delay the documentation and increase the workload, and in accordance with the study of Weerapalin et al. (2018) on “The Development of Registration Information System and Data Processing of Phitsanulok University” stating that the data management through the application platform and the linkage of different databases are important for the effective functions of the data processing of the collaborative database of students’ admissions and records, and also in accordance with the researches of Sitti & Sopeerak (2015) and Jugo et al. (2016) which studied the guidelines for network learning as the process of analyzing the knowledge management affecting the efficiency of the learning process.

The System Development Life Cycle

The system of development life cycle consists of four phases. The formulas were calculated through the collaborative filtering and the system development must be conducted from the beginning to the end. This is in accordance with the studies of Ricci et al. (2010); Jannach et al. (2011); Ghauth & Abdullah (2010) stating that the collaborative filtering consists of four steps, i.e., 1) similarity calculation of the data by implementing the tabulated data of the two users , (2) selection process of the similar members by looking at a brief data of the users on the system for further prediction, (3) prediction process as the anticipation of users’ authorizations by considering the endorsements and similarities of the other items and ranking the similar items selected, and (4) creation of recommended items as the last step by prioritizing the recommended items from the highest prediction values to the lowest prediction values, in accordance with the study of Capron & Johnson (2004) stating that the system development consists of five steps, i.e., (1) system planning, (2) system analysis, (3) system design, (4) system creation, and (5) system implementation and support. This is also in accordance with the studies of Daengdech (2007) and Aiemsiriwong (2004) stating that the development of content management system using the network diagram or NBCLMS with collaborative filtering consists of five steps, i.e., system planning, system analysis, system design, system development and system implementation. The results of the system development were in accordance with the assigned objectives and the experts evaluated the system quality at a high level.

The Evaluation of System Quality

The evaluation of system quality consists of five tests, i.e., performance test, functional test, reliability test, usability test and security test for the rapid and comfortable access of the system users in various dimensions. This is in accordance with the study of Barron & Ivers (1996) stating that the advantages of implementing the educational internet network are: the reduction of paper use, cost saving, effective data recording, simplicity of data preparation and document delivery and also the development of updated data. The internet network is the effective data link between the teachers and the students through various communication channels.

Recommendations

At the macro level, the recommendations are as follows:

1. The national government should have the national strategic plan for education to promote the implementation of the recommendation system of student registration with collaborative filtering.

2. The academic institutions should implement the recommendation system of student registration with collaborative filtering for the social distancing situation and the control of the COVID-19 pandemic.

At the micro level, the recommendation is that the communication and network systems are essential for the operational process of the recommendation system of student registration with collaborative filtering to support the simultaneous communications of all the users.

References

  1. Aiemsiriwong, O. (2014). Analysis and design of information system. Bangkok: SE-Education Public Company Limited.
  2. Ausawawiwatkul, N., & Kijmee, P. (2013). Development of registration and data processing performance of Chiangmai Commercial Technological College. FEU Graduate Journal, 2(1), 1-14.
  3. Barron, A., & Ivers. K. (1996). An Internet Research Model. Presented to the 17th National Educational Computing Conference, Minnesota, USA.
  4. Bobadilla, J., Ortega, F., Hernando, A., & Gutiérrez, A. (2013). Recommender systems survey. Knowledge-Based Systems, 46, 109-132.
  5. Bureau of Policy and Strategy, Office of Permanent Secretary, Ministry of Education. (2020). The strategic plan of the Ministry of Education (2020-2022). Retrieved February 15, 2020, from https://www.moe.go.th/
  6. Capron, H.L., & Johnson, J.A. (2004). Computer: Tools for an Information Age. 8th ed. New Jersey: Pearson Education International.
  7. Daengdech, J. (2007). Problems of development of information system.Presented to the Seminar on the overall outcome of system design and the model development and action plan of information technology, Bangkok,Thailand..
  8. Ghauth, K.I., & Abdullah, N.A. (2010). Measuring learner’s performance in e-learning recommender system. Australasian Journal of Educational Technology. 26(6). 764-774.
  9. Jafari Moghaddam, S., Zaefarian, R., & Salamzadeh, A. (2012). Challenges of entrepreneurship e-education: Evidences from a developing country. International Conference-Entrepreneurship education- a priority for the higher education institutions, 8-9, October.
  10. Jannach, D., Zanker, M., Felferning, A., & Friedrich, G. (2011). Recommender system an introduction.New York: Cambridge University.
  11. Jareonsettasin, T. (2020). Panel discussion education for future world: The best education money cannot buy. Bangkok: Ministry of Education.
  12. Jariyapoom, T. (2018). System development life cycle: SDL. Bangkok: Textbook Publication Center of King Mongkut's University of Technology North Bangkok
  13. Jugo, I., Kova?i?, B., & Slavuj, V. (2014). Using data mining for learning path recommendation and visualization in an intelligent tutoring system, Presented to the 37th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO 2014), Opatija, Croatia.
  14. Musiri R., Saemua, I., & Pradapthong, W. (2019). Online elective subjects’ registration system. Case study: Samkhok school, Provincial Administrative Organization Pathum Thani. Journal of Industrial Technology Phranakhon Rajabhat, 2(1), 39-46.
  15. Ricci, F, Rokach, L., & Shapira, B. (2010). Recommender system handbook. New York: Springer.
  16. Salamzadeh, A., Farjadian, A.A., Amirabadi, M., & Modarresi, M. (2014). Entrepreneurialcharacteristics: Insights from undergraduate students in Iran. International Journal of Entrepreneurship and Small Business, 21(2), 165-182.
  17. Sitti, S., & Sopeerak, S. (2015). The web-based instruction model based on connectivism learning theory to enhance problem-solving skill in information and communication technology of higher education students. FEU Academic Review, 8(2). 102-112.
  18. Thongkaew, T. (2020). New normal based design in education: Impact of COVID-19. Journal of Teacher Professional Development, 1(2). 1-10.
  19. Weerapalin, W., Chatrangsan, P., Kongmuenruk, S., & Tonggla, K. (2018). The development of registration and education processing information system Phitsanulok University. Presented to the 1st National Conference on Humanity and Society, Songkhla, Thailand.
Get the App