Journal of Computer Science and Technology Application (CORISINTA)

Vol. 1, No. 1, 2024, pp. 61∼69

E-ISSN: XXXX-XXXX P-ISSN: XXXX-XXXX, DOI:10.34306 ❒ 1

Machine Learning Techniques for Predicting the Success of

AI-Enabled Startups in the Digital Economy
Dariari Robert1 , Fitra Putri Oganda2 , Asep Sutarman3 , Wahyu Hidayat4 , Anandha Fitriani5
1
Dept. Information System, Eesp Incorporation, British Indian Ocean Territory 2 Dept. Information System,
University of Raharja, Indonesia 3 Dept. Economics and Business, Universitas Muhammadiyah Prof. Dr.
Hamka, Indonesia 4 Dept. Informatics Engineering, University of Raharja, Tangerang, Indonesia 5 Dept.
Digital Business, University of Raharja, Tangerang, Indonesia
1 [email protected], 2 [email protected], 3 asep [email protected], 4 [email protected] 5 [email protected]

Article Info ABSTRACT

Article history: AI-enabled startups in the digital economy have unique challenges in achiev-
ing success. Therefore, it is essential to understand the factors that influence
Received 12/07/2024
the success of these startups. This study uses data analysis methods such as
Revised 02/13/2024 Smart-PLS 4.0, which has three stages: Outer Model Analysis, Inner Model
Accepted 02/27/2024 Analysis, and Hypothesis Testing. By using five variables and ten indicators,
5x10 = 50 indicators. Of course, this cannot be separated from Machine Learn-
Keywords: ing/Machine Learning which uses the Python programming language, which
can be used to accurately predict consumer behavior, business performance, or
Smart-PLS decision-making from startup data. However, using these methods separately
Machine Learning may result in a less accurate or inadequate model. The results show that a com-
StartUp bination of Smart-PLS and Machine Learning/machine learning techniques can
Artificial Intelligence produce more accurate predictive models and can be used to predict the success
Digital rate of AI-enabled startups in the digital economy. That way, this model can as-
sist startup entrepreneurs in making strategic decisions to increase their business
success.

This is an open access article under the CC BY 4.0 license.

Corresponding Author:
Dariari Robert
Dept. Information System, Eesp Incorporation, British Indian Ocean Territory
Email: [email protected]

1. INTRODUCTION
In an increasingly sophisticated digital era, the success of startup companies that utilize artificial
intelligence (AI) as an essential element in business is attracting attention. Predicting the success of AI-
enabled startups in a rapidly evolving digital economy is a complex challenge. However, thanks to advances
in machine learning (ML) technology, various techniques can now be applied to predict the success of these
startups[1][2][3].
This research aims to explore and apply effective machine learning techniques to predict startup suc-
cess using AI in the digital economy, using a joint SmartPLS and machine learning approach. The main focus
of this research is to gain deeper insight into the factors that contribute to the success of AI startups[4].
The research method used includes collecting historical data on AI-powered startups, including vari-
ous attributes such as market size, team capabilities, funding levels, and other factors related to startup success.

Journal homepage: https://journal.corisinta.org/index.php/corisinta/

2 ❒ E-ISSN: XXXX-XXXX P-ISSN: XXXX-XXXX

In addition, data is collected regarding startup success or failure[5].

The Machine Learning technique used in this research is supervised learning[6]: The Machine Learn-
ing model will be trained using historical data, which includes attributes related to startup success and labels
that indicate whether the startup is successful or not. Algorithms such as logistic regression, decision trees, ran-
dom forests, and support vector machines will be used to build predictive models[7]. Still, this study uses the
basic principle of KNN to find the closest data set (neighbours) from new data, which will be predicted based
on Euclidean distance or other metrics. This distance is calculated between the latest data and each exercise
data in the dataset. After that, labels or values from neighbouring data are used to predict new data labels or
values[8].
This research also uses a joint approach between SmartPLS and machine learning. SmartPLS is com-
monly used to analyze structural data in business and management research. The collaboration of SmartPLS
and machine learning technologies provide deeper insights and a broader understanding of the factors influenc-
ing success[9][10].

2. RESEARCH METHOD
In this study, some questionnaires were obtained by distributing them using Google Forms to startups[11].
The questionnaire data had five variables and ten indicators, 5x10 = 50 indicators, targeting 100 samples. Then
this data is processed using machine learning analysis such as supervised learning and K-Nearest Neighbor
(KNN), which is equipped with the Python programming language, as well as using Smartpls such as outer
model analysis, inner model analysis, and hypothesis testing, which will complement this research[12]. Can be
seen in Figure 1.

Figure 1. Flowchart of the Research Process

The use of Supervised Learning research methods, such as K-Nearest Neighbor (KNN), has several
objectives that might be achieved, including:

Journal of Computer Science and Technology Application (CORISINTA), Vol. 1, No. 1, 2024: 61–69
Journal of Computer Science and Technology Application (CORISINTA) ❒ 3

• Classification: One of the main goals of Supervised Learning is to classify. Using KNN, you can type
new data into the correct class based on existing patterns in the previously labeled training data[13][14].
For example, you can classify an email as ”spam” or ”non-spam” based on the email’s features.
• Prediction: The Supervised Learning method is also used to predict continuous or discrete values[15]. In
this case, KNN can predict values based on training data closest to the new data. For example, you can
expect the price of a house based on the attributes that the house has[16].
• Identify Patterns: Supervised Learning can help identify patterns present in training data[17]. KNN, as
a non-parametric method, can help identify complex patterns in data by looking at a group of nearest
neighbours. This can help you understand the relationships between variables and find valuable insights
from the data.
• Pattern Recognition: In some cases, the goal of Supervised Learning is to recognize specific patterns in
new data based on previously known patterns[18]. For example, in facial recognition, KNN can identify
a person’s face based on previously recognized facial patterns.
Dimensionality reduction: Supervised learning methods can also reduce dimensionality, i.e. reducing
the number of features in the data without losing important information[19]. KNN can help identify the most
relevant features or combine similar features in the analysis[20].
It is essential to adjust the research objectives using the Supervised Learning method, including KNN.
Each method has different strengths and weaknesses and needs to be considered according to the characteristics
of the data and your research objectives[21].
The use of the SmartPLS research method, which includes outer model analysis, inner model analysis,
and hypothesis testing, has several objectives that might be achieved, including:
Analisis Outer Model:
Indicator Validity: The objective of the outer model analysis is to evaluate the validity of the construct
(other variables) by examining the validity of the associated indicators. This involves assessing the quality of
indicators, such as convergent and discriminant construct validity[22].
Indicator Reliability: The outer model analysis also aims to examine the reliability of the indicators,
namely the extent to which the hands used in measuring the construct are reliable.
Selection of Indicators: In the outer model analysis, you can also select indicators to ensure that the
arrows represent the construct you want to measure.
Analisis Inner Model:
Structural Model: The inner model analysis aims to build and test a structural model that reflects the
causal relationship between its latent variables. This involves determining the relationship between variables
and testing the significance of the relationship.
Causality and Influence: Analysis of the inner model helps identify direct and indirect influences
between the variables in the model. This can help explain cause-and-effect relationships and understand the
underlying mechanisms between these variables.
Model Validity: The internal model analysis also aims to evaluate the validity of the model as a whole,
namely the extent to which the built model matches the observed data.
Hypothesis test:
Statistical Test: Hypothesis testing aims to test the statistical validity of the relationships proposed in
the structural model. The bootstrapping method is often used in SmartPLS to generate interval estimates and
test the significance of the path coefficient.
Hypothesis Confirmation: Hypothesis testing helps to confirm or reject the proposed research hypoth-
esis based on statistical analysis results. This allows you to infer the relationships between the variables needed
in your research. (H1): Trust significantly impacts the success of the digital economy. (H2): Security signifi-
cantly impacts the success of the digital economy. (H3): Effectiveness and efficiency significantly impact the
success of the digital economy. (H4): The positive impact of facilitating the success of the digital economy.
(H5): Performance skills significantly impact digital economy success. Through outer model analysis, inner
model analysis, and hypothesis testing in SmartPLS, you can gain deeper insights into the causality relationship
between variables and the validity of the model you build. This will strengthen the interpretation of research
results and support the conclusions drawn.

Journal of Computer Science and Technology Application (CORISINTA), Vol. 1, No. 1, 2024: 61-
4 ❒ E-ISSN: XXXX-XXXX P-ISSN: XXXX-XXXX

2.1. Literature Review

Machine Learning
Discussing machine learning in a study certainly requires a strong reference/base. The following is a
scientific research paper that has been carried out and correlates with this research. To improve this research, it
is necessary to conduct a literature review as reference material, including:
1. Title: ”A machine learning, bias-free approach for predicting business success using Crunchbase data.”
Authors: Kamil Żbikowski & Piotr Antosiuk. Conference: Information Processing & Management.
Year: 2021. The research aims to create a predictive benchmark based on machine learning that aims to
predict success in a company.
2. Title: ”Searching for a Unicorn: A Machine Learning Approach Towards Startup Success Prediction.”
Author: Prof. Dr. Wolfgang Karl H¨ardle & Prof. Dr. Weining Wang. Conference: Humboldt-
Universität zu Berlin. Year: 2019. The research aims to obtain reproducible references to support
predictive startup success using machine learning methods.
3. Title: ” How to succeed in the market? Predicting startup success using a machine learning approach.”
Author: Jongwoo Kim, Hongil Kim & Youngjung Geum. Conference: echnological Forecasting and
Social Change . Year: 2023. The research aims to proposes a model to predict startup success, which
considers the external environment and internal conditions.
From the three research sources on machine learning above, it gives a significant impression of achiev-
ing success by utilizing machine learning as research reference material. And here, the researcher can conclude
the correlation of this research with the many research references regarding machine learning. Still, this time
focusing on the three connections above, machine learning can produce predictive models that are more accu-
rate and can be used to predict startup success rates.
Artificial Intelligence in the Digital Economy
Discussing artificial intelligence in the digital economy in a study requires a strong reference/base.
The following is a scientific research paper that has been carried out and correlates with this research. To
improve this research, it is necessary to conduct a literature review as reference material, including:
1. Title: Artificial Intelligence as a Key Driver of Business Operations Transformation in the Conditions
of the Digital Economy”. Authors: Olga I. Maslak, Mariya V. Maslak, Natalya Ye. Grishko, Olha O.
Hlazunova, Petro G. Pererva dan Yaroslava Yu Yakovenko Conference: IEEE International Conference
on Modern Electrical and Energy Systems (MEES) Year: 2021. The research aims to discuss the basis of
Artificial Intelligence and Digital Transformation of the economy and the essential part of technology in
the fast-growing global information society. It aims to study prediction algorithms and change business
systems by creating unique opportunities to use digital technology and revitalize e-commerce.
2. Title: ”Artificial Intelligence as an Enabler for entrepreneurs: a systematic literature review and an
Agenda for future research.” Authors: Guglielmo Giuggioli & Massimiliano Matteo Pellegrini. Confer-
ence: International Journal of Entrepreneurial Behavior & Research. Year: 2022. This research aims
to study the impact of Artificial Intelligence on startups as a motivation for entrepreneurs, of course, by
thinking about the most important designs of Artificial Intelligence in all Industry 4.0 technology models,
such as the Internet of things, augmented reality, blockchain, and so on.
3. Title: ”Strengthening artificial intelligence implementation of security business management in a time of
digital economy innovation.” Author: Heru Susanto & Alifya Kayla Shafa Susanto. Conference: Digi-
talization and Organization Design. Year: 2022. This study aims to inform and control the application
of Artificial Intelligence from any perspective resulting in improved business management in Brunei.
The three research sources on artificial intelligence in the digital economy above certainly give a
significant impression of achieving success by utilizing artificial intelligence in the digital economy as re-
search reference material. And here, the researcher can conclude the correlation of this research with the many
research references on artificial intelligence in the digital economy. Still, this time focusing on the three con-
nections above, AI enables better personalization in presenting content, products and services to consumers in
the digital economy. Through analysis of user data, AI can provide recommendations tailored to individual
preferences, improve user experience and strengthen customer engagement.

Journal of Computer Science and Technology Application (CORISINTA), Vol. 1, No. 1, 2024: 61–69
Journal of Computer Science and Technology Application (CORISINTA) ❒ 5

3. RESULT AND DISCUSSION

Table 1. Questionnaire target data

No Profile Information Frequency Percentage (%)
Men 35 35%
1 Gender
Woman 65 65%
15-20 Years 15 15%
21-25 30 30%
2 Age
26-30 20 20%
31 > 35 35%
Bachelor 40 50%
Equivalent High School 35 35%
3 Education
Middle School Equivalent 15 15%
No education 10 10%
Want to Know Artificial
Reasons for using intelligence (AI) 35 35%
Machine Learning Prediction System
4
and SmartPls to Facilitate Startup
Predict the Success Success Systems in 25 25%
of AI-Enabled the Digital Economy
Startups in the Digital Economy Good Security System 17 17%
Increase Effectiveness
23 23%
And Efficiency

In Table 1. It is a target that will fill out the data questionnaire in this study; four categories will be
used; in category one, there are two gender categories, both men with a frequency of 35 proportion of 35%,
and women with a frequency of 65 proportions 65%. Category 2 is an age category with four levels, level 1,
age 15-20 years, with a frequency of 15 the proportion of 15%; level 2, age 21-25 years, with a frequency of 30
proportions 30%; level 3 age 26-30 years with a frequency of 20 proportions 20%, level 4 age 31¿ Years with
frequency. The 3rd category from the education side of a startup preneur has four levels, 1st level is uneducated
with a frequency of 10 10% proportion, 2nd level is junior high school equivalent with a frequency of 15 15%
proportion, 3rd level is high school equivalent with a frequency of 35 35% proportion, 4th level undergraduate
with a frequency of 40 the proportion of 40%. The 4th category aims to discover the reasons for using Machine
Learning and SmartPls to Predict the Success of AI-Enabled Startups in the Digital Economy. This study is
divided into four levels, level 1 regarding wanting to know the prediction system of artificial intelligence (AI)
with a frequency of 35 proportions 35%; the 2nd level concerns the ease of startup success systems in the digital
economy with a frequency of 25 the proportion of 25%, the 3rd level regarding a sound security system with
a frequency of 17 the proportion of 17%, the 4th level regarding increasing effectiveness and efficiency with a
frequency of 23 the proportion of 23%.
Table 1 is a target reference for people filling out the questionnaire form data.
Table 2. This questionnaire question data will be submitted to 100 samples to support this research.
There are five variables and ten indicators in each variable. In total, 50 indicators will be included in a ques-
tionnaire.
Based on the recap of the questionnaire above using five variables and ten indicators with 100 samples,
the values are as follows: Strongly agree, total = 28, percentage = 56%. Agree, amount = 15, percentage =
30%. Neutral number = 6, percentage 12%. Disagree total = 1, percentage 2%. Strongly disagree amount = 0,
percentage 0%.
The impact of machine learning techniques to predict the success of ai-enabled startups in the digital
economy has a positive/credible outcome for successful digital economy startups.
Figure 2 is a display of the results of validating smart-pls ver 4.0 from the 3 questionnaires above in
the form of 5 variables and 10 indicators, resulting in a sample of 100, of course, the scope is by this research
to obtain an analysis of artificial intelligence, business performance, business talent, start (stop) business and
as an entrepreneur.

Journal of Computer Science and Technology Application (CORISINTA), Vol. 1, No. 1, 2024: 61-
6 ❒ E-ISSN: XXXX-XXXX P-ISSN: XXXX-XXXX

Table 2. Indicator of Questionnaire

Variabel Instrumen Indicator English
The institution has commercializedresearch findings of civitas
academia that have social, economic, and/or environmental
BP1
impacts in various forms of commercialization
Through business incubators, institutions have commercialized
startup companiesformed by students/alumni/lecturers/
researchers/institutional staff/community
Business performance
Institutions have made initial funding innovations for student/
alumni/community startups through funding from venture
capital, startup accelerators, business angels, banks,
BP3 crowdfunding, public support, grants, and others
Institutions have carried out innovations in the form of
university spin-offs, namely bringing ideas from
companies into the university for research orcommercialization
The firm is capable of sustainable development.
The quality of the firm’s products/services has improved over time.
The firm has a good reputation in the industry.
The firm’s customers appreciate its products/services quality.
BP7
The firm’s sales volume has increased over the last 3 years.
The firm offers unique products/services that are difficult to replace.
Wealthier economies have higher rates of entrepreneurial activity
Is it pandemic influenced entrepreneurship levels
SaB1
New businesses need expect to employ many people
The performance of Virtual Artist app always meets my expectations
SaB5 New business need to be innovative
Start (or Stop) a Business
SaB6 Entrepreneurs have to see the prospects growth
SaB7 Effect of age range and employee motivation
In dealing with competitors, our efforts are very often the first to
SaB8
introduce new products or services, technologies, etc.

Figure 2. Display of Outer and Inner Model Analysis

Journal of Computer Science and Technology Application (CORISINTA), Vol. 1, No. 1, 2024: 61–69
Journal of Computer Science and Technology Application (CORISINTA) ❒ 7

Table 3. R-square table

R-Square R-Square Adjusted
Business Performance 0.860 0.847
Start (or Stop) a Business 0.870 0865
Startupreneur Store 0.669 0.646
Digital Talent 0.807 0.803

The R square table in Table 3 predicts that the positive response is high enough to implement a digital
economy startup. It can be seen in the 3 points above showing point 1 business performance R square 0.860
proves a strong value, R square adjusted 0.847 proves a substantial value resulting in the highest value in the
position to 2. Point 2 Start (or stop) a business R square 0.870 proves a substantial value, R square adjusted
0.865 proves a strong value with the highest value among the others. Point 3 startupreneur store R square 0.669
proves a substantial value, R square adjusted 0.646 proves a strong value but has the lowest result. Point 4
talent digital store R square 0.807 proves a substantial value, R square adjusted 0.803 proves a strong value,
has a value result in 3rd position from the others.

Table 4. Table Cronbach’s Alpha

Cronbach alpha rho a rho c AVE
Artificial Intelligence 0.938 0.950 0.949 0.655
Business Performance 0.950 0.952 0.957 0.690
Start (or Stop) a Business 0.936 0.946 0.947 0.644
Startupreneur Store 0.932 0.939 0.943 0.628
Digital Talent 0.933 0.943 0.945 0.634

The explanation in Table 4 of the cronbach’s alpha table contains validation results regarding the
response of artificial intelligence in digital economic startups by calculating cronbach’s alpha, composite re-
liability (rho a), composite reliability (rho c), and average variance extract (AVE). The results of the variable
points above can be stated/proved to be accepted, because they are in accordance with the general provi-
sions/qualifications of the smart-pls technique.
The image above is a graphical bootstrapping which aims to obtain significance level values (p-value)
and path coefficients (t-statistics) from 5 hypotheses determined by a significant explanation in achieving dig-
ital economy startup success using artificial intelligence applying 5 variables and 10 indicators. The highest
variable point is in the business performance variable, namely 0.978; the second position is followed by the
digital talent variable, namely 0.889; the third position is in the startup store variable, namely 0.673 and the last
or lowest position is in the start (or stop) a business variable, namely 0.570.

4. CONCLUSION
Based on the research analysis above, it can be interpreted regarding the role of artificial intelligence
in digital economy startups using machine learning collaboration and SmartPLs ver 4.0. This research uses
5 variables 10 indicators with 100 response samples to assess the success of digital economy startups using
artificial intelligence has 5 core points, namely artificial intelligence, business performance, startup-preneur
store, digital talent and start (or stop) a business. The results of the smart-pls 4.0 method on 5 hypothesis testing
and machine learning supervision of iris learning with the euclidean KNN formula and Python programming
produce the following:
Hypothesis 1 (H1): Trust has a significant impact on the success of the digital economy. Hypothe-
sis 2 (H2): Security can have a significant impact on the success of the digital economy. Hypothesis 3 (H3):
Effectiveness and efficiency have a significant impact on the success of the digital economy, which can be ac-
cepted. Hypothesis 4 (H4): The positive impact of facilitating the success of the digital economy is acceptable.
Hypothesis 5 (H5): Performance skills have a significant impact on digital economy success

Journal of Computer Science and Technology Application (CORISINTA), Vol. 1, No. 1, 2024: 61-
8 ❒ E-ISSN: XXXX-XXXX P-ISSN: XXXX-XXXX

Figure 3. Bootstrapping Graphical

REFERENCES
[1] H. Susanto and A. K. S. Susanto, “Strengthening artificial intelligence implementation of security busi-
ness management in time of digital economy innovation,” in Digitalisation and Organisation Design.
Routledge, 2022, pp. 205–225.

[2] A. Kumar, S. Sharma, and M. Mahdavi, “Machine learning (ml) technologies for digital credit scoring in
rural finance: a literature review,” Risks, vol. 9, no. 11, p. 192, 2021.

[3] U. Rahardja, Q. Aini, D. Manongga, I. Sembiring, and I. D. Girinzio, “Implementation of tensor flow
in air quality monitoring based on artificial intelligence,” International Journal of Artificial Intelligence
Research, vol. 6, no. 1, 2023.

[4] A. M. Musyaffi, H. Khairunnisa, and D. K. Respati, Konsep dasar structural equation model-partial least
square (sem-pls) menggunakan smartpls. Pascal Books, 2022.

[5] I. Amsyar, E. Cristhopher, U. Rahardja, N. Lutfiani, and A. Rizky, “Application of building workers
services in facing industrial revolution 4.0,” Aptisi Transactions on Technopreneurship (ATT), vol. 3,
no. 1, pp. 32–41, 2021.

[6] K.-K. Phoon and W. Zhang, “Future of machine learning in geotechnics,” Georisk: Assessment and Man-
agement of Risk for Engineered Systems and Geohazards, vol. 17, no. 1, pp. 7–22, 2023.

[7] N. Azizah, A. N. Daswar, M. Madani, A. Rizky et al., “Exchange parameters for limiting efficiency
of back-emitting passive silicon solar cells contact,” in 2022 International Conference on Science and
Technology (ICOSTECH). IEEE, 2022, pp. 1–8.

Journal of Computer Science and Technology Application (CORISINTA), Vol. 1, No. 1, 2024: 61–69
Journal of Computer Science and Technology Application (CORISINTA) ❒ 9

[8] U. Rahardja, Q. Aini, D. Manongga, I. Sembiring, and Y. P. A. Sanjaya, “Enhancing machine learning
with low-cost p m2. 5 air quality sensor calibration using image processing,” APTISI Transactions on
Management, vol. 7, no. 3, pp. 201–209, 2023.
[9] R. Supriati, N. Lutfiani, D. Apriani, A. Rizky et al., “Utilizing the potential of blockchain technology
for leading education 4.0,” in 2022 International Conference on Science and Technology (ICOSTECH).
IEEE, 2022, pp. 01–08.
[10] M. Darwin and K. Umam, “Indirect effect analysis on structural equation modeling (comparative study
of using amos and smartpls software),” Nucleus, vol. 1, no. 2, pp. 50–57, 2020.
[11] K. Żbikowski and P. Antosiuk, “A machine learning, bias-free approach for predicting business success
using crunchbase data,” Information Processing & Management, vol. 58, no. 4, p. 102555, 2021.
[12] C. Ünal, “Searching for a unicorn: A machine learning approach towards startup success prediction,”
Master’s thesis, Humboldt-Universität zu Berlin, 2019.
[13] O. I. Maslak, M. V. Maslak, N. Y. Grishko, O. O. Hlazunova, P. G. Pererva, and Y. Y. Yakovenko, “Ar-
tificial intelligence as a key driver of business operations transformation in the conditions of the digital
economy,” in 2021 IEEE International Conference on Modern Electrical and Energy Systems (MEES).
IEEE, 2021, pp. 1–5.
[14] D. Amiruddin, G. P. Cesna, R. Supriati, N. Lutfiani, and T. Nurtino, “Startupreneur multiproduct: Vi-
sionary ilearning on digital business,” in 2022 IEEE Creative Communication and Innovative Technology
(ICCIT). IEEE, 2022, pp. 1–7.
[15] N. Lutfiani, S. Wijono, U. Rahardja, A. Iriani, and E. A. Nabila, “Artificial intelligence based on rec-
ommendation system for startup matchmaking platform,” in 2022 IEEE Creative Communication and
Innovative Technology (ICCIT). IEEE, 2022, pp. 1–5.
[16] G. Giuggioli and M. M. Pellegrini, “Artificial intelligence as an enabler for entrepreneurs: a systematic
literature review and an agenda for future research,” International Journal of Entrepreneurial Behavior &
Research, vol. 29, no. 4, pp. 816–837, 2023.
[17] D. Manongga, U. Rahardja, I. Sembiring, N. Lutfiani, and A. B. Yadila, “Dampak kecerdasan buatan bagi
pendidikan,” ADI Bisnis Digital Interdisiplin Jurnal, vol. 3, no. 2, pp. 41–55, 2022.

[18] E. Retnoningsih and R. Pramudita, “Mengenal machine learning dengan teknik supervised dan unsuper-
vised learning menggunakan python,” Bina Insani Ict Journal, vol. 7, no. 2, pp. 156–165, 2020.
[19] C. Ringle, D. Da Silva, and D. Bido, “Structural equation modeling with the smartpls,” Bido, D., da Silva,
D., & Ringle, C.(2014). Structural Equation Modeling with the Smartpls. Brazilian Journal Of Marketing,
vol. 13, no. 2, 2015.

[20] A. E. E. Sobaih and I. A. Elshaer, “Personal traits and digital entrepreneurship: A mediation model using
smartpls data analysis,” Mathematics, vol. 10, no. 21, p. 3926, 2022.
[21] E. P. Harahap, E. Sediyono, Z. A. Hasibuan, U. Rahardja, and I. N. Hikam, “Artificial intelligence in
tourism environments: A systematic literature review,” 2022 IEEE Creative Communication and Innova-
tive Technology (ICCIT), pp. 1–7, 2022.

[22] A. R. Lubis, M. Lubis et al., “Optimization of distance formula in k-nearest neighbor method,” Bulletin
of Electrical Engineering and Informatics, vol. 9, no. 1, pp. 326–338, 2020.

Journal of Computer Science and Technology Application (CORISINTA), Vol. 1, No. 1, 2024: 61-