Identification Of Water Content And Fat For Green Beans Arabica North Sumatra With NIRS And Artificial Neural Networks

Feb 27, 2025 by ADMIN 119 views

**Identification of Water Content and Fat in Green Beans Arabica North Sumatra using NIRS and Artificial Neural Networks**

Introduction

Arabica coffee from the Lintong and Sidikalang regions, North Sumatra, is renowned for its distinctive taste and aroma. In an effort to understand more about the quality of the coffee, this study focuses on determining the water and fat content in the Green Beans Coffee using the Near Infrared Spectroscopy (NIRS) method and the Artificial Neural Network (ANN). The identification of water and fat content is crucial in understanding the chemical composition of coffee, which can affect its flavor, aroma, and overall quality.

Research Methods

This study aims to get a spectrum and destructive water and fat content for each type of coffee. This process involves 180 coffee samples extracted using the NIRS method. Before getting the spectrum results, the coffee data undergoes initial processing in the form of Multivariate Scatter Correction (MSC), Segmentation, and Principal Component Analysis (PCA). This data processing aims to improve the quality of the data before it is entered into the JST model. The JST model is a type of Artificial Neural Network (ANN) that is trained to predict the water and fat content in the coffee samples.

Data Processing and JST Training

After the data is ready, the JST training is conducted by varying the amount of hidden layers and iteration. This study found that Lintong Coffee water content was identified in the spectrum of 1209.48 Nm, 1458 Nm, and 1933.49 Nm. Meanwhile, fat content is detected in the spectrum of 1728.91 Nm and 1761.08 Nm. For Sidikalang coffee, the water content was identified at 1208.31 Nm, 1459.43 Nm, and 1939.49 Nm, with identified fat at 1731.3 Nm and 1760.56 Nm. The results showed that the best JST architecture to suspect the water content in Lintong coffee was 15-7-1 with 5,000 iterations, while for fat content using architecture 15-12-1 with 1,000 iterations. For Sidikalang coffee, the best architecture used for water content is 15-7-1 with 1,000 iterations, and for fat content is 15-12-1 with 15,000 iterations.

Results and Architecture JST

The use of the NIRS method in this study is an innovative and efficient step in analyzing the chemical components of coffee. NIRS allows non-destructive and fast analysis, but in this study, the analysis is carried out destructively for data validation. In addition, the selection of the right architecture of the JST play play an important role in increasing the accuracy of the prediction of water and fat content. The JST method applied has advantages in terms of the ability to learn complex patterns in data. The use of variations in the amount of hidden layers and iteration can provide more optimal results and improve the ability of the model in making predictions.

Additional Analysis and Explanation

Accurate data availability is very important in this kind of research. Therefore, the results of this study are not only relevant for academics, but can also be used by coffee farmers, processing, and coffee industries to improve the quality of the products produced and understand more about the characteristics of Arabica coffee from North Sumatra. The use of NIRS and JST in this study provides a new approach in analyzing the chemical composition of coffee, which can be applied in various industries related to coffee production.

Conclusion

By utilizing Nirs and JST, this study succeeded in identifying water and fat content in the Arabica coffee beans from Lintong and Sidikalang regions. Through a systematic process, this research opens new opportunities in the development of analytical methods to improve the quality of coffee, as well as providing benefits for various stakeholders in the coffee industry. Through this scientific approach, it is expected to encourage innovation and strengthen the position of North Sumatra coffee in the global market.

Recommendations

Based on the results of this study, it is recommended that the use of NIRS and JST be applied in various industries related to coffee production. This can include coffee farmers, processing, and coffee industries to improve the quality of the products produced and understand more about the characteristics of Arabica coffee from North Sumatra. Additionally, further research is needed to explore the application of NIRS and JST in other types of coffee and to improve the accuracy of the prediction of water and fat content.

Limitations

This study has several limitations. Firstly, the study only focused on Arabica coffee from the Lintong and Sidikalang regions, North Sumatra. Therefore, the results of this study may not be applicable to other types of coffee. Secondly, the study only used 180 coffee samples, which may not be sufficient to represent the entire population of Arabica coffee from North Sumatra. Finally, the study only used NIRS and JST methods, which may not be the most accurate methods for analyzing the chemical composition of coffee.

Future Research Directions

Based on the results of this study, several future research directions can be identified. Firstly, further research is needed to explore the application of NIRS and JST in other types of coffee. Secondly, the study can be replicated using a larger sample size to improve the accuracy of the prediction of water and fat content. Finally, the study can be extended to include other chemical components of coffee, such as protein and sugar content.

Conclusion

In conclusion, this study has successfully identified water and fat content in the Arabica coffee beans from Lintong and Sidikalang regions using NIRS and JST. The results of this study provide a new approach in analyzing the chemical composition of coffee, which can be applied in various industries related to coffee production. Through this scientific approach, it is expected to encourage innovation and strengthen the position of North Sumatra coffee in the global market.

Introduction

In our previous article, we discussed the identification of water content and fat in Green Beans Arabica North Sumatra using Near Infrared Spectroscopy (NIRS) and Artificial Neural Networks (ANN). In this article, we will answer some of the frequently asked questions related to this study.

Q: What is Near Infrared Spectroscopy (NIRS)?

A: NIRS is a non-destructive analytical technique that uses near-infrared radiation to measure the chemical composition of a sample. It is widely used in various industries, including food, agriculture, and pharmaceuticals.

Q: How does NIRS work?

A: NIRS works by shining near-infrared radiation on a sample, which absorbs or reflects the radiation at specific wavelengths. The absorbed or reflected radiation is then measured, and the data is used to calculate the chemical composition of the sample.

Q: What is Artificial Neural Network (ANN)?

A: ANN is a type of machine learning algorithm that is inspired by the structure and function of the human brain. It is composed of interconnected nodes or "neurons" that process and transmit information.

Q: How does ANN work in this study?

A: In this study, ANN is used to predict the water and fat content in the Green Beans Arabica North Sumatra. The ANN algorithm is trained on a dataset of NIRS spectra and corresponding water and fat content values. Once trained, the ANN can predict the water and fat content in new samples based on their NIRS spectra.

Q: What are the advantages of using NIRS and ANN in this study?

A: The use of NIRS and ANN in this study provides several advantages, including:

Non-destructive analysis: NIRS allows for non-destructive analysis of the sample, which means that the sample is not damaged during the analysis.
Fast analysis: NIRS is a fast analytical technique that can provide results in a matter of seconds.
High accuracy: ANN can provide high accuracy in predicting the water and fat content in the Green Beans Arabica North Sumatra.

Q: What are the limitations of this study?

A: The limitations of this study include:

Limited sample size: The study only used 180 coffee samples, which may not be sufficient to represent the entire population of Arabica coffee from North Sumatra.
Limited chemical components: The study only focused on water and fat content, and did not include other chemical components such as protein and sugar content.

Q: What are the future research directions?

A: The future research directions include:

Exploring the application of NIRS and ANN in other types of coffee.
Replicating the study using a larger sample size to improve the accuracy of the prediction of water and fat content.
Extending the study to include other chemical components such as protein and sugar content.

Q: What are the practical applications of this study?

A: The practical applications of this study include:

Improving the quality of coffee production by identifying the optimal water and fat content.
Developing new coffee products with specific water and fat content.
Providing a new approach in analyzing the chemical composition of coffee.

Conclusion

In conclusion, this Q&A article provides answers to some of the frequently asked questions related to the identification of water content and fat in Green Beans Arabica North Sumatra using NIRS and ANN. The study provides a new approach in analyzing the chemical composition of coffee, which can be applied in various industries related to coffee production.