Št. zadetkov: 9
Diplomsko delo
Oznake:
knowledge graphs;word embedding;knowledge graph embedding;natural language processing;computer and information science;diploma thesis;
Structured and unstructured textual data requires efficient representation for computation and manipulation. Many different methods have been developed to represent text in numerical form. Some of these methods are based only on statistical metrics, and some introduce the concept of word context. St ...
Leto:
2020
Vir:
Fakulteta za računalništvo in informatiko (UL FRI)
Izvirni znanstveni članek
Oznake:
razvrščanje dokumentov;optimizacija;grafi znanja;document classification;Bayesian optimisation;representation learning;knowledge graphs;
Building on the success of large language models (LLMs), LLM-based representations have dominated the document representation landscape, achieving strong performance on document embedding benchmarks. However, high-dimensional, computationally expensive LLM embeddings can be too generic or inefficien ...
Leto:
2026
Vir:
Institut Jožef Stefan (IJS)
Izvirni znanstveni članek
Oznake:
Over the past few years, large language models have become technologically ubiquitous and now offer a powerful route to accelerate discoveries in chemistry. In this article, we highlight current impactful applications of large language models in inorganic chemistry, from smart text mining of the ino ...
Leto:
2026
Vir:
Institut Jožef Stefan (IJS)
Izvirni znanstveni članek
Oznake:
medjezikovno učenje;medjezikovni prenos;cross-lingual transfer;cross-lingual learning;catastrophic-forgetting;document classification;
Cross-lingual transfer leverages knowledge from a resource-rich source language, commonly English, to enhance performance in less-resourced target languages. Two widely used strategies are: Cross-Lingual Validation (CLV), which involves training on the source language and validating on the target la ...
Leto:
2025
Vir:
Fakulteta za računalništvo in informatiko (UL FRI)
Diplomsko delo
Oznake:
knowledge graph construction;relation extraction;large language models;biomedical natural language processing;computer science;computer and information science;diploma thesis;
The proliferation of biomedical literature on stroke creates a severe information-overload problem that hinders systematic research and discovery. Knowledge graphs (KGs) can address this by structuring unstructured text into machine-readable form, yet building high-quality, domain-specific KGs remai ...
Leto:
2026
Vir:
Fakulteta za računalništvo in informatiko (UL FRI)
Izvirni znanstveni članek
Oznake:
aspect-based sentiment analysis;benchmark;large language models;learning-to-defer;media monitoring;South Slavic languages;targeted sentiment analysis;
In the contemporary digital media landscape, the ability to automatically distill public opinion from a vast and continuous stream of information is highly important. Aspect-Based Sentiment Analysis (ABSA) offers this granular capability. In this work, we address a specific, industrially relevant fo ...
Leto:
2026
Vir:
Institut Jožef Stefan (IJS)
Izvirni znanstveni članek
Oznake:
retrieval augmented generation;robot action prediction;robot learning in low-data regimes;robotic recycling of electronic waste;vision-language models;
Reliable action prediction is essential for robotic electronic-waste disassembly, where device diversity and damage make preplanned sequences impractical. Vision-language models (VLMs) offer broad visual and semantic knowledge, but zero-shot predictions can be unreliable, while fine-tuning requires ...
Leto:
2026
Vir:
Institut Jožef Stefan (IJS)
Izvirni znanstveni članek
Oznake:
spektralno gručenje;inkrementalna gradnja grafov;grafi k-najbližjih sosedov;besedilne vdelave;Laplaceove lastne preslikave;spectral clustering;incremental graph construction;k-nearest neighbor;(k-nn) graphs;text embeddings;Laplacian eigen maps;
Neighborhood graphs over document embeddings are a common component of text-mining pipelines, supporting topic discovery, deduplication, semi-supervised label propagation, and retrieval-index construction, and they are a critical but fragile step in spectral clustering of text embeddings. On realist ...
Leto:
2026
Vir:
Institut Jožef Stefan (IJS)
Izvirni znanstveni članek
Oznake:
cat;tumour;age;occurrence;retrospective study;Slovenia;
We performed a retrospective study of feline tumours diagnosed in Slovenia over a 21-year period (2000–2020), aiming to analyse their occurrence and related epidemiological features. We systematically reviewed histopathological reports from biopsies and necropsy specimens of cats submitted to the In ...
Leto:
2026
Vir:
Institut Jožef Stefan (IJS)