arXiv subjects

Maria-Cruz Villa-Uriol

Publications and source records attributed to Maria-Cruz Villa-Uriol.

6 recordsLinked to original sources

arXiv2026

Critical Challenges and Guidelines in Evaluating Synthetic Tabular Data: A Systematic Review

Generating synthetic tabular health data is challenging, and evaluating their quality is equally, if not more, complex. This systematic review highlights the critical importance of rigorous evaluation of synthetic health data to ensure reliability, clinical relevance, and appropriate use. From an initial identification of 2067 relevant papers published in the last ten years, 134 studies were selected for detailed analysis. Our review identifies key challenges, including lack of consensus on evaluation methods, inconsistent application of evaluation metrics, limited involvement of domain experts, inadequate reporting of dataset characteristics, and limited reproducibility of results. In response, we provide a structured consolidation of synthetic data generation and evaluation methods into taxonomies, alongside practical guidelines to support more robust and standardised evaluation practices. These findings aim to support the responsible development and use of synthetic health data, aligned with emerging expectations around transparency, reproducibility, and governance, ultimately enabling the community to fully harness its transformative potential and accelerate innovation.

cs.LG↗

arXiv2025

Beyond surface form: A pipeline for semantic analysis in Alzheimer's Disease detection from spontaneous speech

Alzheimer's Disease (AD) is a progressive neurodegenerative condition that adversely affects cognitive abilities. Language-related changes can be automatically identified through the analysis of outputs from linguistic assessment tasks, such as picture description. Language models show promise as a basis for screening tools for AD, but their limited interpretability poses a challenge in distinguishing true linguistic markers of cognitive decline from surface-level textual patterns. To address this issue, we examine how surface form variation affects classification performance, with the goal of assessing the ability of language models to represent underlying semantic indicators. We introduce a novel approach where texts surface forms are transformed by altering syntax and vocabulary while preserving semantic content. The transformations significantly modify the structure and lexical content, as indicated by low BLEU and chrF scores, yet retain the underlying semantics, as reflected in high semantic similarity scores, isolating the effect of semantic information, and finding models perform similarly to if they were using the original text, with only small deviations in macro-F1. We also investigate whether language from picture descriptions retains enough detail to reconstruct the original image using generative models. We found that image-based transformations add substantial noise reducing classification accuracy. Our methodology provides a novel way of looking at what features influence model predictions, and allows the removal of possible spurious correlations. We find that just using semantic information, language model based classifiers can still detect AD. This work shows that difficult to detect semantic impairment can be identified, addressing an overlooked feature of linguistic deterioration, and opening new pathways for early detection systems.

cs.CL↗

arXiv2025

EventBox: A Novel Visual Encoding for Interactive Analysis of Temporal and Multivariate Attributes in Event Sequences

The rapid growth and availability of event sequence data across domains requires effective analysis and exploration methods to facilitate decision-making. Visual analytics combines computational techniques with interactive visualizations, enabling the identification of patterns, anomalies, and attribute interactions. However, existing approaches frequently overlook the interplay between temporal and multivariate attributes. We introduce EventBox, a novel data representation and visual encoding approach for analyzing groups of events and their multivariate attributes. We have integrated EventBox into Sequen-C, a visual analytics system for the analysis of event sequences. To enable the agile creation of EventBoxes in Sequen-C, we have added user-driven transformations, including alignment, sorting, substitution and aggregation. To enhance analytical depth, we incorporate automatically generated statistical analyses, providing additional insight into the significance of attribute interactions. We evaluated our approach involving 21 participants (3 domain experts, 18 novice data analysts). We used the ICE-T framework to assess visualization value, user performance metrics completing a series of tasks, and interactive sessions with domain experts. We also present three case studies with real-world healthcare data demonstrating how EventBox and its integration into Sequen-C reveal meaningful patterns, anomalies, and insights. These results demonstrate that our work advances visual analytics by providing a flexible solution for exploring temporal and multivariate attributes in event sequences.

cs.HC↗

arXiv2021

Sequen-C: A Multilevel Overview of Temporal Event Sequences

Building a visual overview of temporal event sequences with an optimal level-of-detail (i.e. simplified but informative) is an ongoing challenge - expecting the user to zoom into every important aspect of the overview can lead to missing insights. We propose a technique to build a multilevel overview of event sequences, whose granularity can be transformed across sequence clusters (vertical level-of-detail) or longitudinally (horizontal level-of-detail), using hierarchical aggregation and a novel cluster data representation Align-Score-Simplify. By default, the overview shows an optimal number of sequence clusters obtained through the average silhouette width metric - then users are able to explore alternative optimal sequence clusterings. The vertical level-of-detail of the overview changes along with the number of clusters, whilst the horizontal level-of-detail refers to the level of summarization applied to each cluster representation. The proposed technique has been implemented into a visualization system called Sequence Cluster Explorer (Sequen-C) that allows multilevel and detail-on-demand exploration through three coordinated views, and the inspection of data attributes at cluster, unique sequence, and individual sequence level. We present two case studies using real-world datasets in the healthcare domain: CUREd and MIMIC-III; which demonstrate how the technique can aid users to obtain a summary of common and deviating pathways, and explore data attributes for selected patterns.

cs.HC↗

arXiv2021

Gradeer: An Open-Source Modular Hybrid Grader

Automated assessment has been shown to greatly simplify the process of assessing students' programs. However, manual assessment still offers benefits to both students and tutors. We introduce Gradeer, a hybrid assessment tool, which allows tutors to leverage the advantages of both automated and manual assessment. The tool features a modular design, allowing new grading functionality to be added. Gradeer directly assists manual grading, by automatically loading code inspectors, running students' programs, and allowing grading to be stopped and resumed in place at a later time. We used Gradeer to assess an end of year assignment for an introductory Java programming course, and found that its hybrid approach offers several benefits.

cs.CY↗

arXiv2019

Analyzing Time Attributes in Temporal Event Sequences

Event data is present in a variety of domains such as electronic health records, daily living activities and web clickstream records. Current visualization methods to explore event data focus on discovering sequential patterns but present limitations when studying time attributes in event sequences. Time attributes are especially important when studying waiting times or lengths of visit in patient flow analysis. We propose a visual analytics methodology that allows the identification of trends and outliers in respect of duration and time of occurrence in event sequences. The proposed method presents event data using a single Sequential and Time Patterns overview. User-driven alignment by multiple events, sorting by sequence similarity and a novel visual encoding of events allows the comparison of time trends across and within sequences. The proposed visualization allows the derivation of findings that otherwise could not be obtained using traditional visualizations. The proposed methodology has been applied to a real-world dataset provided by Sheffield Teaching Hospitals NHS Foundation Trust, for which four classes of conclusions were derived.

cs.HC↗