Attributing uncertainties to the input space elevates the trustworthiness and explainability of machine learning applications. This paper proposes a novel method called Smoothness Constrained Attribution (SCA), which uses the uncertainty propagation mechanism to propagate the output uncertainty back to the input space. This input uncertainty attribution relies solely on test-time data, the trained uncertainty-aware Machine Learning (ML) model, and assumes a smooth input space, resulting in an efficient and simple system. SCA is compared to existing input Uncertainty Attribution Mechanisms (iUCAMs) based on eXplainable Artificial Intelligence (XAI) and an oracle reference using heteroscedastic noise in different synthetic datasets. These evaluations demonstrate the robustness and improvements of SCA compared to existing methods.

Retrieval over knowledge graphs is typically performed using specialized, complex query languages such as SPARQL.We propose a novel system, Ontology and Semantic Exploration Toolkit (OnSET), that allows novice users to quickly build queries with visual user guidance provided by topic modeling and semantic search throughout the application.OnSET enables users without prior knowledge of the ontology or networked knowledge to start exploring topics of interest over knowledge graphs, including the retrieval and detailed exploration of prototypical sub-graphs and their instances.Existing systems either focus on direct graph exploration or do not foster further exploration of the result set.We, however, provide a node-based editor that can extend these missing properties of existing systems to support search over large ontologies with subgraph instances.Furthermore, OnSET combines efficient and open platforms to deploy the system on commodity hardware.

Ontologies contain semantic relationships and dependencies for, either, ontological studies of a subject or as a blueprint for connected data within s. The ontologies themselves, however, can become difficult to comprehend as specialized class diagrams or complete views. We present an improved interactive visualization for ontology exploration using circle packed hierarchical views within our Ontology and Semantic Exploration Toolkit (OnSET) as the base layer, with interactive visual lenses. The circle packed visualization is enriched in two novel ways incorporating s: first, by linking sparse datasets to the ontology using semantic matching. The second enrichment is performed by employing topic modelling on the ontology and its connection to find groups of topics that cluster the ontology in a hierarchical manner.

Visualizations of knitting and weaving patterns are often bound to two-dimensional visualization, while the resulting fabric piece is inherently three-dimensional. This need for a third dimension arises from (a) the property of the knitting process to create a fabric with offsets and (b) the resulting piece’s structure, i.e. round or closed pieces intended to be worn. Our system, KnitSim, should aid users in visualising possible color combinations, effects of purling or knitting, and the proportions and effects of knit patterns before they start a lengthy, manual knitting process. We propose a web-based interface that enables users to describe their piece as scripted code, allowing it to be generated as a three-dimensional piece. KnitSim, furthermore, enables the simulation of fabric relaxation effects directly in the browser, allowing for the assessment of the final form of the workpiece.