Restiani, Rindi (2024) Prediksi Cuaca Berdasarkan Intensitas Curah Hujan Menggunakan Copula Bayesian Network. Diploma thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 2043201067_Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
2043201067_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2026. Download (2MB) | Request a copy |
Abstract
Cuaca dan iklim merupakan sebuah fenomena di atmosfer yang keberadaannya sangat penting dalam berbagai aktivitas kehidupan. Salah satu faktor yang berpengaruh langsung terhadap perbedaaan tipe atau variasi iklim adalah curah hujan. Curah hujan yang ekstrem dapat berpotensi mengakibatkan bencana alam. Kota Semarang sering dilanda banjir ketika musim penghujan dan daerah tertinggi dengan kejadian banjir di Provinsi Jawa Tengah. Analisis prediksi cuaca di Kota Semarang perlu dilakukan menggunakan Copula Bayesian Network. Bayesian Network dapat menunjukkan probabilitas hubungan antara kejadian-kejadian yang saling berhubungan maupun tidak berhubungan. Data yang berkaitan mengenai cuaca mayoritas numerik. Sedangkan analisis Bayesian Network memiliki beberapa batasan dari perspektif analisis multivariat, utamanya pada kurangnya kontrol terhadap distribusi marginal variabel dan ketidakmampuan untuk menangkap struktur ketergantungan non-linear. Di sisi lain, konsep fungsi Copula diperkenalkan dalam analisis korelasi. Fungsi Copula dapat secara efektif dan akurat menggambarkan keberadaan korelasi linier/non linier, dan korelasi simetris/asimetris antar variabel, memberikan hasil analisis korelasi yang sesuai. Hasil dari penelitian ini menunjukkan dengan pengujian Copula tidak terdapat hubungan antara lamanya penyinaran matahari dengan temperatur minimum dan antara kecepatan angin rata-rata dengan curah hujan. Performa model CBN yang dibangun lebih bagus daripada model BN. Akurasi model CBN yang terbaik sebesar 72,06%.
=================================================================================================================================
Weather and climate are phenomena in the atmosphere whose existence is very important in various life activities. One factor that has a direct influence on differences in climate types or variations is rainfall. Extreme rainfall can potentially result in natural disasters. The city of
Semarang is often hit by floods during the rainy season and the area with the highest incidence of flooding is in Central Java Province. Weather prediction analysis in Semarang City needs to be carried out using the Copula Bayesian Network. Bayesian Network can show the probability
of relationships between related and unrelated events. The majority of weather-related data is numerical. Meanwhile, Bayesian Network analysis has several limitations from a multivariate analysis perspective, mainly the lack of control over the marginal distribution of variables and
the inability to capture non-linear dependency structures. On the other hand, the concept of Copula function is introduced in correlation analysis. The Copula function can effectively and accurately describe the existence of linear/non-linear correlation, and symmetric/asymmetric
correlation between variables, providing appropriate correlation analysis results. The results of this research show that with the Copula test there is no relationship between the length of sunlight and minimum temperature and between average wind speed and rainfall. The performance of the CBN model built is better than the BN model. The best CBN model accuracy was 72.06%.
| Item Type: | Thesis (Diploma) |
|---|---|
| Uncontrolled Keywords: | Copula Bayesian Network, Rainfall Intensity, Weather, Cuaca, Intensitas Hujan |
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA31.3 Regression. Correlation H Social Sciences > HA Statistics > HA31.7 Estimation Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA246.8 Gaussian Q Science > QA Mathematics > QA279.5 Bayesian statistical decision theory. Q Science > QC Physics > QC925 Rain and rainfall S Agriculture > S Agriculture (General) > S600.7.R35 Rain and rainfall |
| Divisions: | Faculty of Vocational > 49501-Business Statistics |
| Depositing User: | Rindi Restiani |
| Date Deposited: | 25 Mar 2024 04:04 |
| Last Modified: | 25 Mar 2024 04:04 |
| URI: | http://repository.its.ac.id/id/eprint/107846 |
Actions (login required)
 |
View Item |