Use of ANN in service quality analysis

Oct 13, 2025 by recoveryshake.com 94 views

Analysis of Service Quality Measurement and Satisfaction with the Servqual and Importance-Performance Analysis Methods Using Artificial Neural Networks

Introduction

In today's competitive education landscape, understanding and improving the quality of service is crucial to maintaining the attractiveness of a tertiary institution. One of the key factors that contribute to student satisfaction is the quality of service provided by the institution. Therefore, it is essential to measure and analyze service quality to identify areas of improvement. This research focuses on analyzing the level of student satisfaction in a tertiary institution using the Servqual method and the Importance-Performance Analysis (IPA) method. The study utilizes Artificial Neural Networks (ANN) to analyze the data and identify patterns in student satisfaction.

Background

The Servqual method is a widely used instrument to measure service quality based on five dimensions: Tangibles (X1), Reliability (X2), Responsiveness (X3), Assurance (X4), and Empathy (X5). These dimensions are essential in understanding the perception of students regarding the services provided by the institution. However, traditional methods of analyzing service quality may not provide accurate and reliable results. Therefore, this study aims to utilize ANN to analyze the data and identify patterns in student satisfaction.

Methodology

The study utilizes a series of multi-time questions to obtain data regarding student satisfaction. The questions are specifically designed to get a clear picture of the students' perception of the services provided. The data is then analyzed using ANN with a feedforward backpropagation algorithm. The best network structure is identified with a 5 neuron configuration in the input layer, one hidden layer with 7 neurons, and one neuron in the output layer. The chosen activation function is a binary sigmoid function that has a range of values between 0 to 1.

Results

The results of the analysis show that the architecture is able to recognize patterns quite well, recorded in the 14th iteration with a Mean Squared Error (MSE) of 0.000192. In addition, the test results from 56 data showed an accuracy rate of 91.07% or equivalent to 51 data that was successfully identified correctly. Only 5 data or 8.93% experienced errors in patterns. This confirms that artificial neural networks with feedforward backpropagation algorithms can be used effectively to analyze the level of student satisfaction with higher education services.

Additional Analysis and Explanation

The Servqual method is an essential tool in understanding the perception of students regarding the services provided by the institution. By using ANN, the data obtained can be analyzed more deeply, so that the manager of the tertiary institution can make decisions based on valid and accurate results. High accuracy in the introduction of patterns can provide a better picture of the perception of students and which areas need to be improved.

The importance of applying performance analysis also cannot be ignored. Through this approach, universities can identify the dimensions of which services are considered lacking by students, as well as priority improvement based on impact on satisfaction. By using data that has been analyzed, educational institutions can design strategies to improve service quality so as to produce higher satisfaction among students.

Conclusion

In conclusion, this study demonstrates the effectiveness of using ANN to analyze the level of student satisfaction with higher education services. The results show that the architecture is able to recognize patterns quite well, and the accuracy rate is high. This confirms that artificial neural networks with feedforward backpropagation algorithms can be used effectively to analyze the level of student satisfaction with higher education services.

Recommendations

Based on the findings of this study, the following recommendations are made:

Use of ANN in service quality analysis: The study demonstrates the effectiveness of using ANN to analyze the level of student satisfaction with higher education services. Therefore, it is recommended that educational institutions use ANN in their service quality analysis.
Importance of performance analysis: The study highlights the importance of applying performance analysis in understanding the perception of students regarding the services provided by the institution. Therefore, it is recommended that educational institutions apply performance analysis in their service quality analysis.
Designing strategies to improve service quality: The study demonstrates the importance of designing strategies to improve service quality based on the analysis of data. Therefore, it is recommended that educational institutions design strategies to improve service quality based on the analysis of data.

Limitations

The study has several limitations, including:

Limited sample size: The study uses a limited sample size of 56 data. Therefore, the results may not be generalizable to the entire population.
Limited dimensions: The study uses only five dimensions of the Servqual method. Therefore, the results may not be comprehensive.
Limited activation function: The study uses only one activation function, which may not be sufficient to capture the complexity of the data.

Future Research Directions

The study has several future research directions, including:

Use of other activation functions: The study uses only one activation function. Therefore, future research can explore the use of other activation functions to capture the complexity of the data.
Use of other algorithms: The study uses only one algorithm, which may not be sufficient to capture the complexity of the data. Therefore, future research can explore the use of other algorithms to analyze the data.
Use of other dimensions: The study uses only five dimensions of the Servqual method. Therefore, future research can explore the use of other dimensions to capture the complexity of the data.

Conclusion

Q: What is the Servqual method and how is it used in service quality analysis?

A: The Servqual method is a widely used instrument to measure service quality based on five dimensions: Tangibles (X1), Reliability (X2), Responsiveness (X3), Assurance (X4), and Empathy (X5). These dimensions are essential in understanding the perception of customers or students regarding the services provided by an institution.

Q: What is the Importance-Performance Analysis (IPA) method and how is it used in service quality analysis?

A: The IPA method is a tool used to identify the importance and performance of different service quality dimensions. It helps institutions to prioritize areas that need improvement and allocate resources accordingly.

Q: What is Artificial Neural Networks (ANN) and how is it used in service quality analysis?

A: ANN is a type of machine learning algorithm that is inspired by the structure and function of the human brain. It is used to analyze complex data and identify patterns, making it an effective tool for service quality analysis.

Q: What are the benefits of using ANN in service quality analysis?

A: The benefits of using ANN in service quality analysis include:

Improved accuracy and reliability of results
Ability to analyze complex data and identify patterns
Ability to identify areas that need improvement and prioritize resources accordingly
Ability to design strategies to improve service quality based on data analysis

Q: What are the limitations of using ANN in service quality analysis?

A: The limitations of using ANN in service quality analysis include:

Limited sample size and data quality
Limited dimensions and variables used in analysis
Limited activation functions and algorithms used
Limited understanding of the underlying mechanisms and processes

Q: How can institutions use the results of ANN analysis to improve service quality?

A: Institutions can use the results of ANN analysis to:

Identify areas that need improvement and prioritize resources accordingly
Design strategies to improve service quality based on data analysis
Monitor and evaluate the effectiveness of service quality improvement initiatives
Continuously improve and refine service quality based on customer feedback and data analysis

Q: What are the future research directions in service quality analysis using ANN?

A: Future research directions in service quality analysis using ANN include:

Exploring the use of other activation functions and algorithms
Investigating the use of other dimensions and variables
Developing more sophisticated and accurate models of service quality
Exploring the use of ANN in other areas of service quality analysis, such as customer satisfaction and loyalty.

Q: How can institutions ensure the validity and reliability of ANN analysis results?

A: Institutions can ensure the validity and reliability of ANN analysis results by:

Using high-quality and relevant data
Ensuring that the ANN model is properly trained and validated
Continuously monitoring and evaluating the performance of the ANN model
Using multiple models and techniques to validate results
Ensuring that the results are interpreted and communicated effectively to stakeholders.

Q: What are the implications of using ANN in service quality analysis for institutions and customers?

A: The implications of using ANN in service quality analysis for institutions and customers include:

Improved service quality and customer satisfaction
Increased efficiency and effectiveness of service quality improvement initiatives
Better understanding of customer needs and preferences
Improved decision-making and resource allocation
Increased competitiveness and reputation of institutions.