The use of synthetic feed additives in Peking ducks has adverse effects on their growth and productivity. Spirulina sp., a blue-green algae, has the potential to be used as a safe feed additive that can enhance the growth and productivity of ducks. This study aims to analyze the effects of Spirulina (Spirulina sp.) powder as a feed additive on the histomorphometry of the duodenum in Peking ducks (Anas platyrhynchos), focusing on the duodenal lumen diameter, villus height, villus width, mucosal tunic thickness, submucosal tunic thickness, and muscular tunic thickness. This study used a Completely Randomized Design (CRD) with 35 ducks divided into 5 treatments and 5 repetitions. The treatments included a control group and the addition of spirulina powder in feed at concentrations of 2.5%, 5%, 7.5%, and 10%. The data were analyzed by ANOVA with a significance level of 5%, followed by Duncan's test. The results of data analysis showed that spirulina powder had a significant effect (P<0.05) on villus height and duodenal mucosa tunica thickness. The conclusion of this study is that spirulina powder has the potential to be used as a feed additive to improve digestive performance in male Peking ducks, as evidenced by the increase in villus height and thick tunica mucosa in the small intestine. The novelty of this study lies in the utilization of spirulina powder as a feed additive to increase the size of the duodenal tissue structure in Peking ducks. The results of this study are expected to be a new reference on the potential of spirulina as a feed additive to improve poultry feed quality and digestive performance, especially in peking ducks.

Perubahan iklim global dan peningkatan variabilitas cuaca membuat prediksi temperatur udara menjadi salah satu kebutuhan penting di berbagai sektor. Temperatur udara merupakan parameter penting dalam meteorologi yang mempengaruhi berbagai aspek kehidupan manusia. Predisi temperatur udara saat ini banyak memanfaatkan algoritma machine learning, namum nilai akurasi masih belum optimal. Tujuan dari penelitian ini untuk meningkatkan akurasi prediksi temperatur udara rata-rata dengan menggunakan pendekatan berbasis machine learning. Metode dalam penelitian ini menggunakan algoritma Linear Regression dan Support Vector Regression (linier dan gaussian non linear) karena memiliki akurasi prediksi data yang cukup baik di berbagai bidang termasuk bidang hidrologi. Penelitian ini menggunakan data dari Badan Meteorologi Klimatologi dan Geofisika (BMKG) lokasi Stasiun Klimatologi Jawa Timur periode data tahun 2019-2023 dengan parameter cuaca temperatur rata-rata (TAV), kelembaban udara (HAV), kecepatan angin (WAV), curah hujan (RR), tekanan udara (PPP), Penyinaran matahari (SUN) dan titik embun (DEW_POINT). Kinerja model dievaluasi menggunakan pengukuran metrik MSE, RMSE, MAE, MAPE dan R². Hasil pengukuran kinerja model algoritma Gaussian support vector Regression (non linier SVR) lebih baik dibanding dengan linear support vector Regression (linear SVR) dan  algoritma linear regression dengan nilai yang lebih tinggi R² sebesar  0,9891 ± 0,0011 dan nilai error yang lebih rendah pada semua metrik pengukuran.
Kata kunci: Prediksi temperatur udara, machine learning, Linear Regression,  Suport Vektor Regression