Sirvi Autor "Kmoch, Alexander, juhendaja" järgi

Nüüd näidatakse 1 - 5 5

Analyzing the relationships between crime and socio-economic and spatial factors using random forest: a case study of Tallinn
(Tartu Ülikool, 2024) Yu, Cheng-Wei; Uuemaa, Evelyn, juhendaja; Kalm, Kadi, juhendaja; Kmoch, Alexander, juhendaja; Zalite, Janis, juhendaja; Tartu Ülikool. Geograafia osakond; Tartu Ülikool. Loodus- ja täppisteaduste valdkond
The spatial factors of crime and its socioeconomic background are important topics in crime research. This study uses a grid framework to represent various spatial, environmental, and socioeconomic factors across Tallinn in 500-meter grids. The study aims to predict the number of crimes in each grid cell through a random forest machine learning model and identify the main contributing factors. Machine learning models do not explain causal relationships between variables but highlight possible correlations, so crime factors need to be discussed within Tallinn's context. Among various types of crime, the factor of commercial locations shows the strongest relationship with the number of crimes. These reflect the concentration of economic activities, assets, and the gathering of people, which are important conditions for crime motivations. Secondly, factors such as the number of renters and the population with low socioeconomic status are associated with the number of crimes against public order.
Comparing raster visualization techniques for environmental indices: univariate and bivariate mapping in Estonia
(Tartu Ülikool, 2025) Fang, Wenyi; Kmoch, Alexander, juhendaja; Tartu Ülikool. Geograafia osakond; Tartu Ülikool. Loodus- ja täppisteaduste valdkond
Visualizing environmental data effectively is crucial for understanding ecological patterns and communicating spatial information. This study compares three cartographic visualization techniques for raster data: univariate mapping, bivariate mapping, and value-by-alpha mapping. To achieve this, this research applied Principal Component Analysis (PCA) to five remote sensing indices (NDVI, NDWI, BSI, NDMI, and LST), and visualized the first two principal components (PC1 and PC2) across four diverse landscapes in Estonia. The objective is to assess the strengths and limitations of these methods in representing complex ecological conditions. Univariate maps were created using a GIS-based weighted overlay method, applying PC1 loadings to generate single-variable sustainability maps. Bivariate mapping included standard bivariate maps, corner models emphasizing extreme combinations, and diagonal models highlighting variable interactions. Value-by-alpha maps employed color to encode PC1 and transparency for PC2, enhancing interpretability. At the same time, quantile and equal interval classifications were applied and compared to illustrate their influence on visual contrast and interpretation. Results indicate that univariate maps offer clear, easily interpretable spatial distributions which are suitable for public communication. Bivariate maps effectively display complex ecological states, with the corner model highlighting extremes and the diagonal model emphasizing variable divergence. Value-by-alpha maps selectively highlight critical zones, though their performance depends on transparency settings. Quantile classification enhances contrast but can overemphasize rare values, while equal interval classification maintains a direct representation of data ranges. Ultimately, the choice of classification and visualization method must align with map objectives whether for exploration analysis, public communication, or expert interpretation. The proposed framework advances multivariate environmental visualization by integrating PCA-based dimensionality reduction with tailored cartographic strategies, offering practical guidance for effective ecological communication and decision-making.
Geospatial data harmonization and machine learning for large-scale water quality modelling
(2022-10-11) Virro, Holger; Uuemaa, Evelyn, juhendaja; Kmoch, Alexander, juhendaja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond
Põllumajanduslik reostus põhjustab jätkuvalt magevee kvaliteedi üleilmset halvenemist. Tõhusate veemajandamise meetmete väljatöötamisel on oluline osa veekvaliteedi modelleerimisel. Veekvaliteedi laialdaseks modelleerimiseks on aga vajalik hea ruumilise katvusega lähteandmete olemasolu. Töö eesmärk oli parandada ja harmoniseerida veekvaliteedi modelleerimiseks vajalikke andmestikke ning arendada välja masinõppe raamistik, mida saaks kasutada riigiüleseks veekvaliteedi modelleerimiseks. Töö üheks väljundiks on Eesti mullastikuandmebaas EstSoil-EH. EstSoil-EH atribuudid olid sisendiks masinõppe mudelile, mida kasutasin mulla orgaanilise süsiniku sisalduse prognoosimiseks. Selgus, et proovivõtukohtade keskkonnatingimused mõjutasid mudeli prognoosi täpsust. Globaalse veekvaliteedi andmete parandamiseks loodi viie andmestiku põhjal andmebaas Global River Water Quality Archive (GRQA). Mullasüsiniku mudeli loomise käigus õpitu põhjal arendati välja raamistik üle-eestiliseks veekvaliteedi modelleerimiseks. Mudel prognoosis toitainete kontsentratsioone 242 Eesti jõe valglas. Saadud mudelite täpsus on võrreldav Baltimaades varem rakendatud mudelitega. Mudelite täpsust mõjutas valglate suurus, kuna prognoosid olid üldjuhul ebatäpsemad väiksemates valglates. Seejuures piisas rahuldava täpsuse saavutamiseks vähem kui pooltest tunnustest, mis näitab, et tunnuste arvust olulisem on nende kirjeldusvõime. Seega on loodud masinõppe mudelid rakendatavad piirkondades, kus tunnuste tuletamiseks vajalike lähteandmete katvus on piiratud.
Measuring Uncertainty Related to Ingesting Data to DGGS
(Tartu Ülikool, 2025) Rammul, Aleksandra; Kmoch, Alexander, juhendaja; Uuemaa, Evelyn, juhendaja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Arvutiteaduse instituut
Discrete Global Grid Systems (DGGS) offer a modern alternative to traditional coordinate-based spatial frameworks by dividing the Earth’s surface into equal-area cells, allowing for standardized, multiresolution geospatial analysis. However, while DGGS helps reduce geometric distortion and improves data handling, it does not eliminate the spatial uncertainty caused by the Modifiable Areal Unit Problem (MAUP). This thesis investigates how aggregation to DGGS grids influences the reliability of landscape metrics and whether uncertainty can be meaningfully quantified across different resolution levels. Using high-resolution land cover data from the Estonian Topographic Database (ETAK), landscape metrics such as patch density, percentage of like adjacencies, Shannon Diversity Index, and class proportion were calculated for hexagonal cells at multiple DGGS resolutions. The land cover class for each cell was assigned using nearest-neighbor matching with raster pixels. The analysis reveals that landscape metrics react differently to resolution changes depending on the spatial structure of the landscape. The results confirm that MAUP cannot be universally measured or avoided. Instead, spatial uncertainty must be approached through context-aware experimental design. While DGGS supports scalable and consistent analysis, its use does not remove the need for careful methodological planning to ensure valid interpretations of spatial patterns.
Veebipõhise kaardirakenduse kui ruumilisi otsustusi toetava süsteemi loomine päästevõrgustiku planeerimiseks Eestis
(Tartu Ülikool, 2024) Remmelg, Ats; Uuemaa, Evelyn, juhendaja; Kmoch, Alexander, juhendaja; Tartu Ülikool. Geograafia osakond; Tartu Ülikool. Loodus- ja täppisteaduste valdkond
Uurimistöö tulemusena valmis veebipõhine kaardirakendus, mille loomiseks kasutati vabavaralist tehnoloogiat: kasutajaliidese puhul Leaflet, Vue.js ja Quasar ning andmehalduseks GeoServer ja PostgreSQL PostGIS laiendiga. Rakendusse lisati Päästeameti poolt soovitud funktsionaalsused, töö käigus kaasati Päästeameti esindajad kogumaks tagasisidet, et luua nende vajadustele vastav süsteem. Loodud veebipõhine GIS leiab kasutust Eesti päästevõrgustiku planeerimisel ning jääb avalikult kättesaadavaks.