Sa`adah, Layla Mazidatus (2025) Pengembangan Model Rekomendasi Berbasis Machine Learning pada Aplikasi Herbamate untuk Optimalisasi Pemanfaatan Tanaman Herbal di Indonesia. Project Report (Praktek Kerja Lapang dan Magang). UPN Veteran Jawa Timur. (Unpublished)
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
22082010105-cover.pdf Download (839kB) |
|
Text (Bab 1)
22082010105-bab1.pdf Download (143kB) |
|
Text (Bab 2)
22082010105-bab2.pdf Restricted to Repository staff only until 18 July 2029. Download (125kB) |
|
Text (Bab 3)
22082010105-bab3.pdf Restricted to Repository staff only until 18 July 2029. Download (247kB) |
|
Text (Bab 4)
22082010105-bab4.pdf Restricted to Repository staff only until 18 July 2029. Download (859kB) |
|
Text (Bab 5)
22082010105-bab5.pdf Download (128kB) |
|
Text (Daftar Pustaka)
22082010105-daftarpustaka.pdf Download (249kB) |
|
Text (Lampiran)
22082010105-lampiran.pdf Restricted to Repository staff only until 18 July 2029. Download (1MB) |
Abstract
Indonesia is one of the world's most biodiverse countries, with more than 30,000 plant species growing in its tropical forests. However, the utilization of herbal plants for healthcare purposes remains far from optimal despite the country's abundant natural resources. Of the approximately 9,600 plant and animal species known to possess medicinal properties, only around 200 plant species have been utilized by the traditional medicine industry to produce herbal products. The lack of well-organized information and the difficulty of matching health conditions with appropriate herbal remedies have become major obstacles. This issue reflects a significant information gap that limits the effective utilization of herbal plants by the public. To address these challenges, the Herbamate application was developed as an innovative technology-based solution to optimize the utilization of herbal plants in Indonesia. Herbamate employs machine learning technology to develop a herbal plant recommendation system based on users' health symptoms. The system processes user inputs, such as reported symptoms, and matches them with relevant herbal plant data. In addition, the application provides features for searching herbal plants, accessing information about their medicinal benefits, usage instructions, and dosage guidelines to promote the safe and effective use of herbal remedies. The application development process began with a user needs survey involving 31 respondents, most of whom were between 17 and 25 years old, an age group considered highly adaptable to technology. The survey revealed that 100% of respondents did not know which herbal plants were suitable for their health complaints, 77.4% experienced difficulty obtaining reliable information, and 58.1% lacked knowledge regarding proper usage and dosage. The information considered most important by respondents included the medicinal benefits of herbal plants (100%), usage instructions (96.8%), and composition (45.2%). Furthermore, 92.2% of respondents expressed interest in using a herbal plant recommendation application. These findings served as the foundation for designing the application's features, ensuring that Herbamate effectively addresses users' needs. Herbamate was designed with a simple and intuitive user interface featuring herbal plant recommendations based on health symptoms, herbal plant search, and a favorites list. On the home page, users can easily access various features according to their needs. The recommendation feature enables users to input their health symptoms, which are then processed by the machine learning model to generate relevant herbal plant recommendations. The system workflow is illustrated through flowcharts and use case diagrams that describe user interactions with the application in a structured manner. The system development process consisted of data collection, preprocessing, feature selection, dataset splitting, and machine learning model development. The dataset contained 1,001 herbal plant records, including associated health symptoms, descriptions of medicinal benefits, compositions, and usage methods. The data were preprocessed through numerical format conversion, removal of missing values, and symptom binarization to facilitate data analysis. Feature selection ensured that only relevant information was used during model training. The Herbamate recommendation model was developed using an Artificial Neural Network (ANN) consisting of multiple hidden layers. The model was trained using the Adam optimization algorithm, with Mean Squared Error (MSE) as the loss function and Mean Absolute Error (MAE) as the evaluation metric. The training results demonstrated that the model successfully learned data patterns, as indicated by low MSE and MAE values and stable learning curves without signs of overfitting. Model evaluation showed excellent performance, achieving an R-squared (R²) value of 0.9989, indicating that the model explained 99.89% of the variance in the dataset. In addition, the Root Mean Squared Error (RMSE) value of 0.0038 further confirmed the model's prediction accuracy. The trained model was integrated into the application's recommendation feature through the recommend_herbs function, which processes users' symptom inputs, predicts the relevance of herbal plants, and displays personalized recommendation results. With a carefully designed algorithm, the application is capable of providing accurate, relevant, and user-oriented herbal plant recommendations. In conclusion, Herbamate offers an innovative solution to overcome the challenges of herbal plant utilization in Indonesia. By providing features designed according to users' needs and supported by advanced machine learning technology, Herbamate not only improves public access to herbal plant information but also increases public awareness of their benefits. The application has significant potential to support the development of Indonesia's herbal industry, improve public health, and strengthen Indonesia's position as a leading contributor to the global herbal medicine industry.
| Item Type: | Monograph (Project Report (Praktek Kerja Lapang dan Magang)) | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||
| Subjects: | Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software Q Science > QA Mathematics > QA76.6 Computer Programming Q Science > QA Mathematics > QA76.87 Neural computers |
||||||||
| Divisions: | Faculty of Computer Science > Departemen of Information Systems | ||||||||
| Depositing User: | Layla Mazidatus Sa`adah | ||||||||
| Date Deposited: | 20 Jul 2026 04:40 | ||||||||
| Last Modified: | 20 Jul 2026 04:40 | ||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/56123 |
Actions (login required)
 |
View Item |