Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 39 for “"HDBSCAN"”.
-
Scenario Classification via Imagery and Video Data for ADAS Trigger Analysis
… 128 principal components, and cluster them using HDBSCAN or K-Means. The evaluation comprises 18 controlled experiments on the same 8,017 RAB trigger images. Cluster count, noise ratio, Silhouette score, and Davies–Bouldin index are considered together with PCA and UMAP visualizations. DINOv2 …
-
Epidemiologia da leptospirose humana, a partir de modelos de análise espacial, Porto Alegre - RS
… análise espacial Análise de Agrupamentos (Índice HDBSCAN e Local Moran’s I) e Análise Hierárquica de Processo (AHP). Foram utilizados dados do SINAN, entre 2007 e 2019. Para a álgebra de mapas foi utilizado o software ArcGIS Pro 2.9.1; e as ferramentas Density-based Clustering (para o Índice …
-
Shluková analýza šífrovaného provozu
… exportovaných přes IPFIX. Algoritmy K-Means a HDBSCAN jsou testovány na datasetu CESNET-QUIC22 a hodnoceny pomocí metrik homogenity a normalizované vzájemné informace. Bhattacharyyova vzdálenost slouží k posouzení oddělitelnosti tříd. Výsledky ukazují, že shlukování umožňuje identifikaci vzorců …
-
The long haul: reimagining grain supply chains in the shrinking infrastructure landscape of Manitoba
… Spatial Clustering of Applications with Noise (HDBSCAN) algorithm, and statistical modeling to examine the spatial structure and determinants of route distance. The findings show that elevator access is not uniformly poor across Manitoba. Instead, severe transportation burdens are concentrated …
-
Analýza a fingerprinting síťových relací z paketových záznamů
… pomocí algoritmu UMAP a shlukována algoritmem HDBSCAN. Z vytvořených shluků jsou generována pravidla v podobě profilů služeb (tzv. otisků/fingerprintů) připravených pro další strojové zpracování. Validace prokázala, že pokud jsou vhodná trénovací data, dokáže nástroj spolehlivě izolovat hlavní …
-
Třídění dotazníkových odpovědí s využitím strojového učení
… porovnáva výsledky algoritmov K-Means a HDBSCAN nad sémantickými embeddingmi, optimalizuje parametre a hodnotí kandidátne riešenia pomocou viacerých metrík. Experimenty ukázali, že navrhnutá metóda dosiahla priemernú zhodu s referenčnými štítkami 86,55%, čím prekonala najsilnejšie …
-
A data-driven road pavement segmentation framework
… MinMax, followed by two methodologies that use HDBSCAN clustering, one for management, covering 469km of highway, and one for project design, being tested in 10 sections of 20km each. This step was approached to counter the current CDM, performing visual and statistical comparisons, such as …
-
Anglų kalbos naujienų straipsnių klasterizavimas taikant šiuolaikinius teksto įterpinių modelius /
… lemmatization, UMAP dimensionality reduction and HDBSCAN clustering algorithm, a coherence cv = 0.8215 was achieved. LLM based embedding methods outperformed classical statistical methods. Testing dimensionality reduction algorithms it was noticed, that nonlinear methods (UMAP, PaCMAP) produced …
-
A Spatial Analysis of Ceramics in Northwestern Alaska: Studying Pre-Contact Gendered Use of Space
… specific use of space. I tested this using the HDBSCAN (Hierarchical Density Based Spatial Clustering of Applications with Noise) algorithm in Python, a programming language. HDBSCAN identifies discrete clusters of artifacts, as well as the persistence, or stability, of the cluster. Birnirk and …
-
Modelo exploratorio no supervisado para el análisis semántico del comportamiento económico a partir de fuentes oficiales de difusión audiovisual, mediante técnicas de NLP y Clustering
… y algoritmos de clustering (K-Means, DBSCAN y HDBSCAN). Así hemos obtenido como resultado que el modelo Word2Vec combinado con K-Means y 2 PCA presentó las métricas de validación más optimas y aplicables al negocio, superando a modelos basados en SBERT en separación y estabilidad en los grupos …
-
Vectorising Tabular Data for RAG and Vector Database Validation - A study of tabular data retrieval and anomaly detection methods against poisoning attacks in RAG
… considered. Unsupervised anomaly detection using HDBSCAN and Isolation Forest is combined with pre-trained classifiers for prompt injection and hate speech detection as a potential validation layer. Using only classifiers achieves the highest F1-score (0.592 for prompt injection on Regeringen), …
-
Video Object Counting in Unconstrained Environments Using Density-Based Clustering
… We discuss the implementation of the HDBSCAN algorithm which we use in our novel approach for matching multiple local feature descriptors. We show that the detected clusters provide very good matches for the features and demonstrate our approach to cluster analysis and validation. We …
-
Strukturelle Differenzierung und Plastizität präsynaptischer Aktiver Zonen
… neue Analyse-Algorithmen auf der Basis von HDBSCAN entwickelt, um eine objektive, in weiten Teilen automatisierte Quantifizierung bis auf Ebene der Substruktur der AZ zu ermöglichen. Die Differenzierung wurde am Beispiel phasischer und tonischer Synapsen, die an dieser NMJ durch Is- und …
-
Mapping nearby young stellar structures and their star formation histories
… distance of 333 pc using Gaia DR2. Applying the HDBSCAN clustering algorithm to this sample, we identify 27 top-level groups, nearly half of which have little to no presence in previous literature. Ten of these groups have visible substructure, including notable young associations such as Orion, …
-
Adapting Regenerative Braking Strength to Driver Preference
… hardware. By leveraging advanced techniques like HDBSCAN clustering, fuzzy logic inference, and online Q-learning, the research achieves accurate driving condition identification and adaptation to individual driver preferences in a control scheme that can be practically deployed in-vehicle. …
-
Comparison of applying advanced spatiotemporal clustering algorithms with machine learning methods for forecasting of urban mobility demand /
… were successfully segmented using K-means and HDBSCAN clustering. Comparative evaluation indicates that machine learning approaches significantly outperform the baseline SARIMA model. Specifically, the XGBoost model achieved the highest accuracy, reducing the Root Mean Square Error (RMSE) for …
-
Microanalysis of sulphide rims: desulphurisation, latestage oxidation, and possible effects on PGE formation in chromite seams of the Rum layered intrusion, Scotland
… hierarchical density-based clustering algorithm (HDBSCAN). Results indicated that the optimum number of average counts per pixel lies ~ 1200. The mineralogical data set addressed elemental changes across the BMS and explored their origins, aiding understanding of magmatic formation and …
-
MACHINE LEARNING-GUIDED PREDICTION OF FUNCTIONAL OUTCOMES OF EGFR EXTRACELLULAR DOMAIN MUTATIONS IN COLORECTAL CANCER: IMPLICATIONS FOR CETUXIMAB RESISTANCE
… analysis using UMAP dimensionality reduction and HDBSCAN clustering revealed that resistance mutations cluster into two distinct mechanistic clusters. XGBoost, Random Forest, and Logistic Regression classifiers trained on the resulting 830 (mutation, run) examples achieved a perfect AUC-PRC of …
-
Detecção de corpos hídricos utilizando imagens SAR Sentinel-1 com índice SEWI e técnicas de classificação
… CLARA, K-means, SOM, Autoencoder, HDBSCAN e Expectativa- Maximização. Os resultados demonstraram que os classificadores Autoencoder e CLARA alcançaram os melhores desempenhos, com coeficientes de Dice médios de 0,81 e altos valores de recall (0,90 e 0,89, respectivamente). De forma …
Page 1 of 2