PRIKUPLJANJE, OBRADA I MODELIRANJE PODATAKA ZA 24-SATNU VREMENSKU PROGNOZU ; Collection, Processing and Modeling of Data for 24-hour Weather Forecast
Gespeichert in:
| Titel: | PRIKUPLJANJE, OBRADA I MODELIRANJE PODATAKA ZA 24-SATNU VREMENSKU PROGNOZU ; Collection, Processing and Modeling of Data for 24-hour Weather Forecast |
|---|---|
| Autoren: | Kurtić, Dora |
| Weitere Verfasser: | Čelar, Stipo |
| Verlagsinformationen: | Sveučilište u Splitu. Fakultet elektrotehnike, strojarstva i brodogradnje. University of Split. Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture. |
| Publikationsjahr: | 2024 |
| Bestand: | Croatian Digital Theses Repository (National and University Library in Zagreb) |
| Schlagwörter: | API, Docker, ML, Vremenska prognoza, Mikroservisna arhitektura, Linearna regresija, Polinomna regresija, Python, PostgreSQL, RabbitMQ, Weather forecast, Microservis arhitecture, Linear regression, Polinomial regression, TEHNIČKE ZNANOSTI. Računarstvo, TECHNICAL SCIENCES. Computing |
| Beschreibung: | U ovome radu prikazan je proces razvoja sustava za prognozu vremena u naredna 24 sata. Kreirana su dva dijela sustava. Jedan je za dohvat, obradu i pohranu podataka. Ovaj sustav realiziran je pomoću mikroservisne strukture, koristi više izvora podataka i vanjske servise poput RabbitMQ i PostgreSQL baze podataka. Drugi dio sustava služi za proces modeliranja podataka, unutar kojega čistimo i nadopunjujemo podatke, te kreiramo ML modele za predikciju. ML modeli koji se koriste su linearna predikcija i polinomna predikcija. Korisnih predikcije o prognozi vremena dobiva u obliku grafova koje Python skripta pomoću spremljenih ML modela generira. ; This thesis presents the process of developing a system for forecasting the weather in the next 24 hours. Two parts of the system were created. One is for retrieving, processing and storing data. This system is implemented using a microservice structure who uses multiple data sources and external services such as RabbitMQ and PostgreSQL database. The second part of the system is used for the data modeling process, in which we clean and supplement the data, and create ML models for prediction. The used ML models are linear prediction and polynomial prediction. Useful predictions about the weather forecast are obtained in the form of graphs generated by a Python script using saved ML models. |
| Publikationsart: | master thesis |
| Dateibeschreibung: | application/pdf |
| Sprache: | Croatian |
| Relation: | https://zir.nsk.hr/islandora/object/fesb:2256; https://urn.nsk.hr/urn:nbn:hr:179:035646; https://repozitorij.svkst.unist.hr/islandora/object/fesb:2256; https://repozitorij.svkst.unist.hr/islandora/object/fesb:2256/datastream/PDF |
| Verfügbarkeit: | https://zir.nsk.hr/islandora/object/fesb:2256 https://urn.nsk.hr/urn:nbn:hr:179:035646 https://repozitorij.svkst.unist.hr/islandora/object/fesb:2256 https://repozitorij.svkst.unist.hr/islandora/object/fesb:2256/datastream/PDF |
| Rights: | http://rightsstatements.org/vocab/InC/1.0/ ; info:eu-repo/semantics/restrictedAccess |
| Dokumentencode: | edsbas.FD624B4A |
| Datenbank: | BASE |
Nájsť tento článok vo Web of Science | Abstract: | U ovome radu prikazan je proces razvoja sustava za prognozu vremena u naredna 24 sata. Kreirana su dva dijela sustava. Jedan je za dohvat, obradu i pohranu podataka. Ovaj sustav realiziran je pomoću mikroservisne strukture, koristi više izvora podataka i vanjske servise poput RabbitMQ i PostgreSQL baze podataka. Drugi dio sustava služi za proces modeliranja podataka, unutar kojega čistimo i nadopunjujemo podatke, te kreiramo ML modele za predikciju. ML modeli koji se koriste su linearna predikcija i polinomna predikcija. Korisnih predikcije o prognozi vremena dobiva u obliku grafova koje Python skripta pomoću spremljenih ML modela generira. ; This thesis presents the process of developing a system for forecasting the weather in the next 24 hours. Two parts of the system were created. One is for retrieving, processing and storing data. This system is implemented using a microservice structure who uses multiple data sources and external services such as RabbitMQ and PostgreSQL database. The second part of the system is used for the data modeling process, in which we clean and supplement the data, and create ML models for prediction. The used ML models are linear prediction and polynomial prediction. Useful predictions about the weather forecast are obtained in the form of graphs generated by a Python script using saved ML models. |
|---|