Analiza velike količine podataka iz različitih izvora u području pametne poljoprivrede

master's thesis

• Title (english): Big data analysis from different sources in smart agriculture
• Author: Ana Skukan
• Mentor: Gordan Ježić (mentor)
• Committee member: Gordan Ježić (predsjednik povjerenstva)
• Committee member: Ivana Podnar Žarko (član povjerenstva)
• Committee member: Mario Kušek (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-04, Croatia

Abstract

U današnje doba brzih klimatskih promjena i napretka u pametnoj poljoprivredi, postaje vrlo bitno s kakvim meteorološkim podacima baratamo. Kroz ovaj rad fokus je stavljen na usporedbu in-situ i satelitskih podataka za nekoliko meteoroloških varijabli kroz pregled nekoliko lokacija. Varijable koje su se proučavale su vjetar, padaline, uvjeti u zraku, temperatura, temperatura tla i sunčeva radijacija, a lokacije su Budimci, Skenderovci, Osijek i Novo Selo Rok.
Statistička analiza uključuje čišćenje podataka, analizu različitih koeficijenata korelacije, statističke testove te linearnu regresiju. Osim statističke analize u pisanom obliku, rad sadrži i Web aplikaciju koja nudi detaljan pregled analize svake varijable za svaku lokaciju te prikladne vizualizacije.
Kroz analizu se zaključuje da varijable koje su povezane uz manja područja i imaju tendenciju biti više volatilne, kao što su vjetar i padaline, ne koreliraju, dok varijable koje su slične na bliskim mikro-područjima, kao što su temperatura i uvjeti u zraku, imaju visoku do vrlo visoku (snažnu) korelaciju.

Abstract (english)

In today's era of rapid climate change and progress in smart agriculture, it becomes very important what kind of meteorological data we handle. Through this paper, the focus is placed on the comparison of in-situ and satellite data for several meteorological variables through an overview of several locations. The variables studied are wind, precipitation, air conditions, temperature, soil temperature and solar radiation, and the locations are Budimci, Skenderovci, Osijek and Novo Selo Rok.
Statistical analysis includes data cleaning, analysis of various correlation coefficients, statistical tests and linear regression. In addition to statistical analysis in written form, the paper also contains a Web application that offers a detailed overview of the analysis of each variable for each location and suitable visualizations.
The analysis concludes that variables that are associated with smaller areas and tend to be more volatile, such as wind and precipitation, do not correlate, while variables that are similar in close micro-areas, such as temperature and air conditions, have strong correlation.

• Keywords: Velika količina podataka; Meteorologija; Statistička analiza; Koeficijenti korelacije; Pametna poljoprivreda; Statistički test; Linearna regresija; Znanost o podacima; Web aplikacija
• Keywords (english): Big Data; Meteorology; Statistical Analysis; Correlation Coefficients; Smart Agriculture; Statistical Test; Linear Regression; Data Science; Web application
• Language: croatian
• URN:NBN: urn:nbn:hr:168:147423
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-10-11 10:27:19