|
Eduvest � Journal of Universal Studies Volume 4 Number 08, August,
2024 p- ISSN 2775-3735- e-ISSN 2775-3727 |
|
|
|
|
|
Robby
Ilhamkusuma, Irwan Robi Prastomo, Mardi Hardjianto 1,2 Fakultas Teknologi Informasi, Universitas Budi
Luhur, Indonesia Email: [email protected] |
|
|
ABSTRACT |
|
|
Efficiency in
the selection of prospective employees can have a significant impact on
increasing costs and extending the time required in the recruitment process
of HR Information Systems in boarding schools.� However, the hiring process is often still
done manually and traditionally, resulting in high costs and time and lack of
efficiency. Social media has become an important part of modern society,
including in the employee recruitment and selection process. Social media can
be used to obtain more comprehensive information about prospective employees,
such as educational background, skills, work experience, and personality.
This research investigates using Linear Regression Algorithm on social media
classification and Technique for Order Preference by Similarity to Ideal Solution
(TOPSIS) method as ranking criteria in classifying prospective employees. One
of the proposed solutions is to utilize information from social media to
evaluate the potential and personality of prospective employees. This
research specifically reviews the application of Linear Regression Algorithm
and TOPSIS method in categorizing prospective employees based on information
obtained from social media, focusing on HR Information System in a particular
boarding school. The results showed that the application of social media
classification system with linear regression algorithm with Mean absolute
error 10.50, Residual sum of squares (MSE): 147.16, R-squared: 1.0 and TOPSIS
method with an accuracy rate of 0.51238 for the first rank can improve the
efficiency of the recruitment process. |
|
|
KEYWORDS |
Linear
Regression, TOPSIS Method, Social Media Classification, Recruitment, Boarding
School |
|
|
This
work is licensed under a Creative Commons Attribution-ShareAlike
4.0 International |
In the ever-evolving digital era, the recruitment process has become one of
the most important aspects of Human Resources (HR) management in various
organizations (Solikhah & Amelia, 2022). This process
involves the selection and evaluation of prospective employees to ensure that
they have the qualifications that match the company's needs. HR effectiveness
can contribute significantly and gain a competitive advantage in achieving
educational goals, ensuring the safety and well-being of students, and
maintaining a smooth learning process (Nawaz, 2016).
The influence of human resource information systems on employees' work
tasks is so significant that maintaining the quality of HRIS has become one of
the top priorities for organizations (Srivastava, Dev and Bajaj, 2021). By
implementing HRIS, boarding schools can increase the efficiency of resource
use, reduce administrative burden, improve data accuracy, and enable more
optimized decision-making.
The recruitment process is often still done manually and traditionally,
resulting in high costs and time and lack of efficiency. With information
systems that are traditional and not integrated, schools have difficulty
managing data and information at large (Ben Moussa and El Arbi, 2020). In
addition, the development of social media also provides new opportunities in
obtaining information about potential employee candidates.
Previous research with the use of several combinations of methods from
machine learning including this personality classification method uses the
logistic regression method with TF-IDF and AHP weighting. From the
classification with these two weights on the social behavior approach has an
average accuracy of 24.95% (Prameswari and Setiawan, 2019).
Therefore, this research aims to improve the efficiency of the recruitment
process by using social media classification method supported by linear
regression algorithm and TOPSIS method. This
research will be conducted on the HR information system in a boarding school.
In addition, the HR division also takes a long time in making work reports
because they have to search and check employee data and work contracts one by
one (Chairul Anwar, 2019).
Utilizing social media as a data source can assist the
HR department in obtaining additional information about job applicants such as
their online activities or interactions with others. By using linear regression
algorithm and TOPSIS method, the information from social media can be
classified so as to facilitate the identification of potential employees who
match the needs of boarding school management. This concept makes it possible to evaluate and select
the most optimal HR management alternative for boarding schools. (Suroso & Setyawatie, 2019).
This concept makes
it possible to evaluate and select the most optimal HR management alternative
for boarding schools. This research is expected to contribute to the
development of HR information systems in boarding schools, so as to improve the
efficiency and effectiveness of the recruitment process.
The data collection process has a central role, because the
choice of data collection methods will affect the quality and accuracy of the
data collected during the implementation of research with a variety of
different methods. Data collection methods are carried out with the aim of
obtaining essential information to achieve research objectives. Data collection
techniques in this study are surveys, observations, documentation which are
described as follows:
- Survey
The survey method can be utilized to collect data from
prospective employees who are undergoing the recruitment process. The survey
can take the form of a questionnaire that includes questions related to the
prospective employee's profile, skills, experience, and other important
information relevant to the recruitment process.
- Observation
Make direct observations of prospective employees who take
part in the recruitment stage to collect data on the behavior and
characteristics of these individuals. Observations are made when prospective
employees undergo various tests or interview sessions.
- Data from Social Media
Data collection from social media platforms such as
LinkedIn, Instagram, or other platforms used by prospective employees to search
for job opportunities. Data from
social media includes profile information, work experience, qualifications, and
relevant references.
- Data from HRIS
Collect data from the HR information system (HRIS) in boarding
schools that includes information about prospective employees who have applied
and gone through the initial recruitment stages. HRIS data will provide a
comprehensive view of prospective employees who have followed the selection
process.
- Historical Data
Collection of historical data about prospective employees
who have previously been recruited by boarding schools. This historical data
will serve as a reference for analysis using a linear regression algorithm in
classifying the most optimal social media.
Data analysis techniques in this study include linear
regression analysis, TOPSIS (Technique for Order of Preference by Similarity to
Ideal Solution) method, descriptive statistical analysis, integration of HRIS
Data and Social Media Data, and Recruitment Decision Making.
New employee
selection using TOPSIS uses 5 criteria including application management
documents obtained from GPA scores (C1), test scores during the ongoing
acceptance of prospective boarding school employees obtained from the Psychotest Score (C2), through the use of Media obtained
from the level of interaction (C3), the number of uploads (C4), and the number
of followers (C5). The data used in this study are: first, the work experience
of prospective employees includes data on the type of work, company/school, and
length of service. Based on this data, the prospective employees analyzed in
this study have various work experiences, ranging from zero to more than ten
years. The most popular types of jobs are in education, finance, and services.
The most popular companies are schools, multinational companies and national
private companies. Second, the social activities of prospective employees
include data on organizational involvement, social activities, and social media
usage. Based on the data, it can be concluded that the prospective employees
analyzed in this study have different levels of social involvement. Most
prospective employees are actively involved in organizational and social
activities. Some prospective employees also actively use social media to share
information and interact with others. And, third, the future employees' social
media data is relevant to the research objective, which is to predict employee
performance. Information about a prospective employee's education and work
experience can be used to predict the employee's abilities and skills. Social
activities of prospective employees can be used to predict the employee's
personality and behavior.
In this research,
a linear regression algorithm is used to classify social media data of
prospective employees based on predetermined criteria. The criteria used in
this research are as follows:
1.�� Academic ability (GPA)
2.�� Psychotest Score
3.�� Number of Uploads
Linear regression also helps identify the relationship
between the variables in the social media data and the candidate's abilities. This relationship can be depicted in the form of a
linear regression equation.
1.�� Plotting the data
Dataset for social
media classification
Data plotting is
done between the score_psychotest as X and the number
of followers as Y.
Plotting the data
Data plot results
2. Perform data
division for train.
The data that has
been plotted will be divided based on the following steps:
Data sharing
process
The result of the
data sharing process
Figure 6. Results
of the data train
3. Plotting to Get
Regression Results
Figure 7. Results
of the plotted linear regression model
�
4. Perform
Accuracy Calculation
Mean absolute
error (MAE) is a measure of how far the average prediction result deviates from
the true value. A smaller MAE value indicates that the prediction is more
accurate.
Figure 8. Accuracy
Calculation
Residual sum of
squares (MSE) is a measure of how far the total deviation of the predicted
result is from the true value. A smaller MSE value indicates that the
prediction is more accurate.
R2-squared is a
measure of how well the model predicts the true value. A larger R2-square value
indicates that the model is more accurate.
Figure 9. Error
calculation results
Based on these
values, it can be concluded that the model predictions have good accuracy. The
low MAE and MSE values indicate that the average prediction results do not
deviate too far from the actual values. The defined R2-scuared value indicates
that the model is more accurate.
Linear regression
results are used as one of the determining factors in the selection process of
prospective employees. In addition to linear regression results, other factors
that can also be used in the employee selection process are ability test results,
interviews, and recommendations from superiors or coworkers.
Here is an example
of how linear regression results can be used in the employee selection process:
1.�� Candidates who have high ability scores
across all criteria will rank higher in the selection process.
2.�� Candidates who have high ability scores in
criteria relevant to the job they are applying for will rank higher in the
selection process.
The Topsis method is a Multiple Attribute Decision Making
(MADM) method used to determine the best choice from a number of alternatives.
This method is based on the concept of the closest distance to the positive
ideal solution and the farthest distance to the negative ideal solution. Topsis Method Implementation Steps.
- Describe the
criteria
2.�� Create a decision matrix
3.�� Determine criteria weights
4.�� Calculating positive and negative ideal
vectors
5.�� Calculating the distance of alternatives to
positive and negative ideal vectors
6.�� Ranking prospective employees
The following is
an example of the implementation of the Topsis method
for ranking prospective employees:
1.
Criteria,
Alternatives and Weights
TOPSIS analysis
table
Table 1. Criteria
data and weights
|
GPA |
(A) |
0,25 |
|
Value_psychotest |
(B) |
0,25 |
|
Interaction
level |
(C) |
1,15 |
|
Number of
uploads |
(D) |
0,15 |
|
Number of
followers |
(E) |
0,2 |
�
- TOPSIS Method
Testing� Results
From TOPSIS
research, test results are obtained through Microsoft Excell applications and
models using PHP.
TOPSIS test
results (1)
TOPSIS test
results (2)
TOPSIS test
results (3)
The data used in
this research is secondary data obtained from the company. The data is in the
form of data about the process of recruiting prospective employees, selection
criteria, and selection results.
1.
Data
analysis was conducted using the Topsis method. This
method is used to calculate the distance of alternatives to positive ideal
vectors and negative ideal vectors. The smallest total distance to the positive
ideal vector and the largest total distance to the negative ideal vector are
used to rank alternatives.
2.
The
results show that the Topsis method can be used to improve efficiency in the
recruitment process. This method can help schools to: Shorten the time of the
selection process, Save the cost of the selection process, Increase the
objectivity of the assessment of prospective employees.
3.
Topsis
method is an effective method to increase efficiency in the recruitment
process. This method can help companies to get the best employees at a more
efficient cost.
4.
A
conceptual framework can help you to theoretically explain how the Topsis
method can be used to improve efficiency in the hiring process.
5.
An
explanation of the selection criteria can help you to understand how the
selection criteria affect the selection outcome.
6.
The
explanation of the data analysis results can help you to understand how the
Topsis method can be used to calculate the distance of alternatives to the
positive ideal vector and the negative ideal vector.
7.
Recommendations
you can give to help companies improve efficiency in the recruitment process.
8.
Conceptual
framework
Efficiency in the
recruitment process can be defined as the company's ability to complete the
process in a timely, cost-effective, and resource-efficient manner. The
efficiency of the recruitment process can be improved by using the Topsis method. The Topsis method
is a MADM method used to determine the best choice from a number of
alternatives. This method is based on the concept of the closest distance to
the positive ideal solution and the farthest distance to the negative ideal
solution.
In the context of
the recruitment process, the Topsis method can be
used to shorten the selection process time, helping companies to shorten the
selection process time by reducing the number of prospective employees who need
to be selected, saving the cost of the selection process by reducing the number
of tests and interviews that need to be conducted, increasing the objectivity
of the assessment of prospective employees by using clear selection criteria
and appropriate selection methods.
Selection criteria
are factors used to assess prospective employees. The right selection criteria
can help companies to get the best employees. There are several selection
criteria that are commonly used, including Academic ability, Technical
skills, Work experience, Personality
1.
The
selection criteria used should be relevant to the job being applied for. For
example, if the job requires high communication skills, then the selection
criteria should include communication skills.
2.
The
results of data analysis show that the Topsis method can be used to improve
efficiency in the recruitment process. This method can help companies to:
3.
The
Topsis method can help schools to shorten the selection process time by
reducing the number of prospective employees who need to be selected, helping
schools to save the cost of the selection process by reducing the number of
tests and interviews that need to be conducted, helping companies to increase
the objectivity of assessing prospective employees by using clear selection
criteria and appropriate selection methods, The efficiency of the recruitment
process can help schools to increase employee satisfaction. This can happen
because new employees feel that the company values their time and effort.
There are several
things that schools can do to improve the efficiency of the recruitment
process, including Technology can help companies to streamline the recruitment
process. For example, schools can use an application system to manage
prospective employee data, conduct online selection, and send selection result
announcements, Clear selection criteria can help companies to focus on
candidates who meet the requirements. This can help the company to reduce the
time and cost needed for the selection process, the right selection method can
help the company to assess prospective employees objectively. This can help the
company to get the best employees.
A well-trained
recruitment team can help schools to make the selection process more efficient
and effective. The following are some examples of the application of efficiency
in the recruitment process A school uses an application system to manage
prospective employee data. This system can help schools to track the status of
candidates and send notifications to candidates about the selection process, Using online tests to assess the skills of candidates.
Online tests can help schools to save time and money for the selection process.
Using panel interview method to assess prospective employees. The panel
interview method can help schools to get a more objective assessment of
prospective employees. By implementing efficiency in the recruitment process,
companies can increase productivity, save costs, and increase employee
satisfaction, The selection process of prospective employees becomes faster and
more accurate because social media data can provide additional relevant
information.
Here are some of the advantages of HR
information systems in Boarding Schools
1.
The
implementation of linear regression algorithm and TOPSIS method in HR
information system in boarding schools provides advantages in improving
efficiency and accuracy in selecting prospective employees.
2.
The
system is able to identify potential employees quickly and efficiently, thus
reducing the time and effort required in the recruitment process.
Based on the description of the background, research
problems and problem formulation to improve efficiency in the recruitment
process of prospective employees in pesantren through
a method that combines social media analysis, linear regression, and the TOPSIS
approach. Some of the main findings of this research are.
1.
Integrating
social media data analysis into the recruitment process has the potential to
provide additional insights into candidate qualifications that may not be
detected through traditional methods.
2.
Although linear regression was used
in this study, it was found to be suboptimal for the classification task. The
linear regression results are more suitable for predicting continuous values
than classifying categorical labels.
3.
The results
showed that the application of social media classification system with linear
regression algorithm with Mean absolute error 10.50, Residual sum of squares
(MSE): 147.16, R-squared: 1.0 and TOPSIS method with an accuracy rate of
0.51238.
REFERENCES
Abidin, Z., Hanggara, B.T. and Prakoso, B.S.
(2021) �� Sistem Informasi Sekolah berbasis Web menggunakan Metode Extreme Programming (Studi Kasus: Al-Izzah International Islamic Boarding School Kota
Batu)�, � Informasi dan Ilmu
Komputer e-ISSN, 5(8), pp. 3222�3228.
Agusli, R., Dzulhaq, M.I. and Irawan, F.C. (2020) �Sistem
Pendukung Keputusan Penerimaan
Karyawan Menggunakan Metode
Ahp-Topsis�, Academic Journal of Computer Science Research, 2(2), pp.
35�40.
Andira Sembiring, A., Sani Sembiring, A.
and Ramadan Siregar, S. (2018) �Sistem Pendukung Keputusan Penentuan Prioritas Pengembangan Industri Kecil Menengah Di Kabupaten Karo Mengunakan Metode Topsis�, Majalah Ilmiah INTI, 5(3), pp. 269�274.
Arkarina, N.K., Widodo,
A.W. and Furqon, M.T. (2019) �Implementasi
Regresi Linier Berganda Untuk Prediksi Jumlah Peminat Mata Kuliah Pilihan�, Jurnal Pengembangan Teknologi Informasi Dan Ilmu Komunikasi, 3(11), pp.
10462�10467.
Billy Renaldo Potale, Viktor Lengkong, S.
(2016) �Pengaruh Proses Rekrutmen
Dan Seleksi Terhadapkinerja
Karyawan Pada Pt Bank Sulutgo�,
16(04), pp. 453�464.
Breaugh, J.A. (2013)
�Employee recruitment�, Annual Review of Psychology, 64, pp. 389�416.
Chairul Anwar (2019)
�Analisa dan Perancangan Sistem
Informasi Human Resources Development Pada PT. Semacom Integrated dengan Menggunakan Metode Waterfall�, International Journal of
Education, Science, Technology, and Engineering, 2(1), pp. 19�38.
Chen, J. et al.
(2019) �A comparison of linear regression, regularization, and machine learning
algorithms to develop Europe-wide spatial models of fine particles and nitrogen
dioxide�, Environment International, 130(May).
Claudy, Y.I., Setya Perdana, R. and Fauzi, M.A. (2018) �Klasifikasi Dokumen Twitter Untuk Mengetahui Karakter Calon Karyawan Menggunakan Algoritme K-Nearest
Neighbor (KNN)�, 2(8), pp. 2761�2765.
Dedi, Hilmi Fuad,
A.S. (2014) �Perancangan Sistem
Informasi Human Resources Pada PT .
Indonesia Koito�, STMIK Bina Sarana Global, 4(1).
Dyas Marsa Pranoto,� dan A.M.S. (2021) �Kepemimpinan
di Sekolah Boarding School�, Jurnal
Ilmiah Wahana Pendidikan
https://jurnal.unibrah.ac.id/index.php/JIWP, 7(1), pp. 391�402.
Eva, N. and Fadel,
M. (2022) �Rancang Bangun Sistem Informasi Penilaian Kinerja Karyawan Menggunakan Metode Topsis Dan 360
Derajat Pada Pt. Murni Mandiri
��, Jurnal Sains &Teknologi
Fakultas Teknik [Preprint].
Hartoko, M.S. (2016) �Rekrutmen Calon Karyawan Tetap�, JURNAL AKUNTANSI, EKONOMI dan MANAJEMEN BISNIS |
e-ISSN: 2548-9836, 4(2), pp. 122�127.
KURNIASIH, S.
(2021) �Audit Sistem Informasi
Human Resource Information System (Hris) Pada Bagian
Human Resource (Hr) Menggunakan
Framework Cobit 5 Domain Dss01�, Nuansa
Informatika, 15(2), pp. 53�63.
Kwok, E. and
Susanti, W. (2019) �Penerapan Metode Regresi Linier dalam Aplikasi Sistem Peramalan Jumlah Bahan Baku untuk Produksi Tahu�, Mahasiswa Aplikasi Teknologi Komputer dan Informasi, 1(2), pp. 1�8.
Maulana, M.A., Setyanto, A. and Kurniawan, M.P. (2018)
�Analisis Sentimen Media Sosial Universitas Amikom�, Seminar Nasional
Teknologi Informasi dan Multimedia 2018 UNIVERSITAS AMIKOM Yogyakarta, 10
Februari 2018, pp. 7�12.
Maulud, D. and
Abdulazeez, A.M. (2020) �A Review on Linear Regression Comprehensive in Machine
Learning�, Journal of Applied Science and Technology Trends, 1(4), pp.
140�147.
Ben Moussa, N. and
El Arbi, R. (2020) �The impact of Human Resources Information Systems on
individual innovation capability in Tunisian companies: The moderating role of
affective commitment�, European Research on Management and Business
Economics, 26(1), pp. 18�25.
Muslimin, S.
(2014) �Problem dan Solusi di Sekolah Berasrama Boarding School�, pp. 1�8.
Nawaz, N. (2016) �ISSN : 2249-0558�, (August).
Novianty, D., Palasara, N.D. and Qomaruddin, M. (2021)
�Algoritma Regresi Linear pada Prediksi Permohonan Paten yang Terdaftar di
Indonesia�, Jurnal Sistem dan Teknologi Informasi (Justin), 9(2), p. 81.
Prameswari, K. and Setiawan,
E.B. (2019) �Analisis Kepribadian Melalui
Twitter Menggunakan Metode Logistic Regression dengan Pembobotan TF-IDF dan
AHP�, e-Proceeding of Engineering, 6(2), pp. 9667�9682.
Putra, L.M. (2023)
�Penerapan Metode Technique for Order Preference by
Similarity to Ideal Solution (TOPSIS)�, Intisari,
1(531408058), pp. 416�425.
Rong, S. and
Bao-Wen, Z. (2018) �The research of regression model in machine learning
field�, MATEC Web of Conferences, 176, pp. 8�11.
Safitri, K. and Tinus Waruwu, F. (2017) �SISTEM PENDUKUNG KEPUTUSAN PEMILIHAN
KARYAWAN BERPRESTASI DENGAN MENGGUNAKAN METODE ANALYTICAL HIEARARCHY PROCESS
(Studi Kasus : PT.Capella Dinamik Nusantara Takengon)�,
1(1), pp. 12�16.
Srivastava, S.,
Dev, S. and Bajaj, B. (2021) �Human resource information system use,
satisfaction, and success�, International Journal of Enterprise Information
Systems, 17(1), pp. 106�124.
Suroso & Setyawatie (2019) �Seleksi Penerimaan Karyawan Baru Menggunakan Metode Topsis�, JIPTK
(Jurnal Ilmu Pengetahuan dan Teknologi Komputer), 4(2), pp. 1�8.
Yohana, N.D. and
Marisa, F. (2018) �Perancangan Proses Bisnis Sistem Human Resource
Management (HRM) Untuk Meningkatkan
Kinerja Pegawai�, J I M P - Jurnal
Informatika Merdeka Pasuruan,
3(2), pp. 23�32.
Solikhah, M., & Amelia, A. (2022). Analisis CSF dan SWOT
Studi Kasus: PT. Educational Development Consultant (EDC) Cirebon. Jurnal
Ekonomi Teknologi Dan Bisnis (JETBIS), 1(3), 124�130.