Analysis of the Accuracy Level of the Fuzzy Sugeno and Tsukamoto Inference Model in Predicting the Inflation Rate in North Sumatra

Oct 13, 2025 by orovajewels.com 141 views

Analysis of the Accuracy Level of the Fuzzy Sugeno and Tsukamoto Inference Model in Predicting the Inflation Rate in North Sumatra

Introduction

Fuzzy logic is a powerful tool in handling uncertain data, making it an ideal choice for various applications, including predicting inflation rates. In this study, we aim to analyze the accuracy level of two fuzzy inference models, namely the Fuzzy Sugeno and Tsukamoto models, in predicting the inflation rate in North Sumatra. The prediction results of the two models are then compared with actual data to determine which model is more accurate.

Background of Fuzzy Logic and Fuzzy Inference Models

Fuzzy logic is an extension of traditional reasoning that allows an element to become a member of a set with a certain degree of membership (μ). This ability to map uncertain values makes it an ideal tool in the fuzzy inference model. The fuzzy inference model is often used in various cases to produce clear and certain outputs from uncertain data. The two fuzzy inference models used in this study are the Fuzzy Sugeno and Tsukamoto models.

Fuzzy Sugeno Model

The Fuzzy Sugeno model is a type of fuzzy inference model that uses a linear combination of the input variables to produce the output. This model is widely used in various applications, including control systems and decision-making systems. The Fuzzy Sugeno model is known for its simplicity and ease of use, making it a popular choice among researchers and practitioners.

Fuzzy Tsukamoto Model

The Fuzzy Tsukamoto model is another type of fuzzy inference model that uses a non-linear combination of the input variables to produce the output. This model is known for its ability to handle complex relationships between the input variables and the output. The Fuzzy Tsukamoto model is widely used in various applications, including control systems and decision-making systems.

Methodology

This study aims to analyze the accuracy level of the Fuzzy Sugeno and Tsukamoto models in predicting the inflation rate in North Sumatra. The prediction results of the two models are then compared with actual data to determine which model is more accurate. The analysis shows that the level of accuracy of the Fuzzy Sugeno model is influenced by multiple linear regression values. That is, the Fuzzy Sugeno model is more accurate in predicting the inflation rate if the data used to train the model has a strong correlation with variables that affect inflation.

On the other hand, the level of accuracy of the Fuzzy Tsukamoto model is influenced by the accuracy of the input value. The Fuzzy Tsukamoto model is more accurate in predicting the inflation rate if the input data given has a more appropriate value.

Results

From the test results of the inflation rate case in North Sumatra, the Fuzzy Tsukamoto model shows a higher accuracy compared to the Fuzzy Sugeno model. This indicates that the Fuzzy Tsukamoto model is more effective in predicting the inflation rate in North Sumatra, especially if the input data given has a high level of accuracy.

Conclusion

This study shows that the Fuzzy Tsukamoto inference model is more accurate in predicting the inflation rate in North Sumatra compared to the Fuzzy Sugeno model. These results are important in determining a more effective model to predict the rate of inflation in the future.

Suggestion

Researchers are advised to examine more in the influence of significant economic variables on the rate of inflation in North Sumatra to increase the accuracy of the Fuzzy Tsukamoto model.
It is important to ensure that the input data used to train the fuzzy model has high accuracy so that the model can produce accurate predictions.
Subsequent research can be done by comparing the Fuzzy Tsukamoto model with other predictive models, such as multiple linear regression models or Arima models, to get a more comprehensive understanding of inflation rate predictions in North Sumatra.

Implications

By understanding the factors that influence the accuracy of the fuzzy inference model, as well as by utilizing the right data, the fuzzy model can be an effective tool in predicting the rate of inflation and helping decision makers in developing more effective economic strategies.

Future Research Directions

Future research can be done by:

Examining the influence of other economic variables on the rate of inflation in North Sumatra
Developing a hybrid model that combines the Fuzzy Tsukamoto model with other predictive models
Applying the Fuzzy Tsukamoto model to other regions or countries to see if the results are generalizable

Limitations

This study has several limitations, including:

The study only analyzed the accuracy of the Fuzzy Sugeno and Tsukamoto models in predicting the inflation rate in North Sumatra
The study did not examine the influence of other economic variables on the rate of inflation in North Sumatra
The study only used a limited dataset to train the fuzzy models

Conclusion

In conclusion, this study shows that the Fuzzy Tsukamoto inference model is more accurate in predicting the inflation rate in North Sumatra compared to the Fuzzy Sugeno model. These results are important in determining a more effective model to predict the rate of inflation in the future. Future research can be done by examining the influence of other economic variables on the rate of inflation in North Sumatra and developing a hybrid model that combines the Fuzzy Tsukamoto model with other predictive models.
Frequently Asked Questions (FAQs) about the Fuzzy Sugeno and Tsukamoto Inference Models in Predicting the Inflation Rate in North Sumatra

Q: What is the main difference between the Fuzzy Sugeno and Tsukamoto models?

A: The main difference between the Fuzzy Sugeno and Tsukamoto models is the way they handle the input variables. The Fuzzy Sugeno model uses a linear combination of the input variables, while the Fuzzy Tsukamoto model uses a non-linear combination of the input variables.

Q: Which model is more accurate in predicting the inflation rate in North Sumatra?

A: The Fuzzy Tsukamoto model is more accurate in predicting the inflation rate in North Sumatra compared to the Fuzzy Sugeno model.

Q: What are the factors that influence the accuracy of the Fuzzy Sugeno model?

A: The accuracy of the Fuzzy Sugeno model is influenced by multiple linear regression values. That is, the Fuzzy Sugeno model is more accurate in predicting the inflation rate if the data used to train the model has a strong correlation with variables that affect inflation.

Q: What are the factors that influence the accuracy of the Fuzzy Tsukamoto model?

A: The accuracy of the Fuzzy Tsukamoto model is influenced by the accuracy of the input value. The Fuzzy Tsukamoto model is more accurate in predicting the inflation rate if the input data given has a more appropriate value.

Q: Can the Fuzzy Tsukamoto model be used in other regions or countries?

A: Yes, the Fuzzy Tsukamoto model can be used in other regions or countries to predict the inflation rate. However, the accuracy of the model may vary depending on the specific economic conditions of the region or country.

Q: What are the limitations of this study?

A: The study has several limitations, including:

The study only analyzed the accuracy of the Fuzzy Sugeno and Tsukamoto models in predicting the inflation rate in North Sumatra
The study did not examine the influence of other economic variables on the rate of inflation in North Sumatra
The study only used a limited dataset to train the fuzzy models

Q: What are the future research directions?

A: Future research can be done by:

Examining the influence of other economic variables on the rate of inflation in North Sumatra
Developing a hybrid model that combines the Fuzzy Tsukamoto model with other predictive models
Applying the Fuzzy Tsukamoto model to other regions or countries to see if the results are generalizable

Q: What are the implications of this study?

A: The study shows that the Fuzzy Tsukamoto inference model is more accurate in predicting the inflation rate in North Sumatra compared to the Fuzzy Sugeno model. These results are important in determining a more effective model to predict the rate of inflation in the future.

Q: Can the fuzzy model be used in other fields?

A: Yes, the fuzzy model can be used in other fields, such as:

Control systems
Decision-making systems
Forecasting systems
Optimization systems

Q: What are the benefits of using the fuzzy model?

A: The benefits of using the fuzzy model include:

Ability to handle uncertain data
Ability to produce clear and certain outputs from uncertain data
Ability to handle complex relationships between input variables and output
Ability to improve the accuracy of predictions

Q: What are the challenges of using the fuzzy model?

A: The challenges of using the fuzzy model include:

Difficulty in selecting the right fuzzy parameters
Difficulty in handling large datasets
Difficulty in interpreting the results of the fuzzy model

Q: Can the fuzzy model be used in real-time applications?

A: Yes, the fuzzy model can be used in real-time applications, such as:

Predicting the inflation rate in real-time
Controlling systems in real-time
Making decisions in real-time

Q: What are the future developments of the fuzzy model?

A: The future developments of the fuzzy model include:

Developing new fuzzy algorithms
Improving the accuracy of the fuzzy model
Applying the fuzzy model to new fields
Developing hybrid models that combine the fuzzy model with other predictive models.