Electronic Theses and Dissertation

Penelitian ini bertujuan untuk mengevaluasi kemampuan metode Near Infrared (NIR) yang dikombinasikan dengan analisis kemometri, dalam membedakan daun nilam (Pogostemon cablin) varietas Tapak Tuan berdasarkan tiga kondisi, yaitu kontrol (PCTC), terpapar jamur Synchytrium pogostemonis (PCTI), dan resisten
(PCTR). Selain itu, penelitian ini juga bertujuan untuk mengamati perbedaan morfologi daun antar ketiga kategori tersebut. Sampel daun nilam diambil dari Nino Park, Agricultural Research Center (ARC) Universitas Syiah Kuala (USK), Darussalam, Banda Aceh. Spektrum NIR dianalisis menggunakan metode kemometri meliputi Principal Component Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA) dan Support Vector Machine Classification (SVMC). Hasil penelitian menunjukkan bahwa kombinasi metode NIR dan kemometri dapat
digunakan untuk melihat perubahan komposisi daun nilam varietas Tapak Tuan kontrol, terpapar jamur Synchytrium pogostemonis dan resisten. Model klasifikasi PLS-DA menunjukkan performa yang baik dengan nilai R² kalibrasi sebesar 0,894 dan validasi sebesar 0,877, serta nilai RMSEC sebesar 0,21 dan RMSEP sebesar
0,23. Sementara itu, model SVMC menunjukkan akurasi klasifikasi yang lebih rendah, dengan nilai training accuracy sebesar 70,27% dan validation accuracy sebesar 71,891%, sehingga dinilai kurang cocok untuk klasifikasi kategori daun nilam pada penelitian ini. Hasil pengamatan morfologi menggunakan Scanning Electron Microscope (SEM) menunjukkan adanya perbedaan struktur permukaan daun antar ketiga kategori, terutama pada tekstur dan kerapatan permukaan daun.
Kata kunci: NIR, kemometri, PLS-DA, SVMC, daun nilam, Synchytrium pogostemonis

This study aims to activate the capability of the Near Independent Variable (NIR) method combined with chemometric analysis, in distinguishing patchouli (Pogostemon cablin) leaves of the Tapak Tuan variety based on three conditions, namely control (PCTC), exposure to the fungus Synchytrium pogostemonis (PCTI), and resistance (PCTR). In addition, this study also aims to observe the differences in leaf morphology between the three categories. Patchouli leaf samples were taken from Nino Park, Agricultural Research Center (ARC) Syiah Kuala University (USK), Darussalam, Banda Aceh. The NIR spectrum was analyzed using chemometric methods including Principal Component Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA) and Support Vector Machine Classification (SVMC). The results showed that the combination of NIR and chemometric methods can be used to observe changes in the composition of patchouli leaves of the Tapak Tuan variety control, exposed to the fungus Synchytrium pogostemonis and resistance. The PLS-DA classification model showed good performance with a calibration R² value of 0.894 and a validation value of 0.877, as well as an RMSEC value of 0.21 and an RMSEP of 0.23. Meanwhile, the SVMC model showed lower classification accuracy, with a training accuracy value of 70.27% and a validation accuracy of 71.891%, so it was considered less ideal for classifying patchouli leaf categories in this study. The results of morphological observations using a Scanning Electron Microscope (SEM) showed differences in leaf surface structure between the three categories, especially in leaf surface texture and density. Keywords: NIR, chemometry, PLS-DA, SVMC, patchouli leaves, Synchytrium pogostemonis