arXiv subjects

George Papastefanatos

Publications and source records attributed to George Papastefanatos.

17 recordsLinked to original sources

arXiv2026

Human-Data Interaction, Exploration, and Visualization in the AI Era: Challenges and Opportunities

The rapid advancement of AI is transforming human-centered systems, with profound implications for human-AI interaction, human-data interaction, and visual analytics. In the AI era, data analysis increasingly involves large-scale, heterogeneous, and multimodal data that is predominantly unstructured, as well as foundation models such as LLMs and VLMs, which introduce additional uncertainty into analytical processes. These shifts expose persistent challenges for human-data interactive systems, including perceptually misaligned latency, scalability constraints, limitations of existing interaction and exploration paradigms, and growing uncertainty regarding the reliability and interpretability of AI-generated insights. Responding to these challenges requires moving beyond conventional efficiency and scalability metrics, redefining the roles of humans and machines in analytical workflows, and incorporating cognitive, perceptual, and design principles into every level of the human-data interaction stack. This paper investigates the challenges introduced by recent advances in AI and examines how these developments are reshaping the ways users engage with data, while outlining limitations and open research directions for building human-centered AI systems for interactive data analysis in the AI era.

cs.DB↗

arXiv2025

EmeraldMind: A Knowledge Graph-Augmented Framework for Greenwashing Detection

As AI and web agents become pervasive in decision-making, it is critical to design intelligent systems that not only support sustainability efforts but also guard against misinformation. Greenwashing, i.e., misleading corporate sustainability claims, poses a major challenge to environmental progress. To address this challenge, we introduce EmeraldMind, a fact-centric framework integrating a domain-specific knowledge graph with retrieval-augmented generation to automate greenwashing detection. EmeraldMind builds the EmeraldGraph from diverse corporate ESG (environmental, social, and governance) reports, surfacing verifiable evidence, often missing in generic knowledge bases, and supporting large language models in claim assessment. The framework delivers justification-centric classifications, presenting transparent, evidence-backed verdicts and abstaining responsibly when claims cannot be verified. Experiments on a new greenwashing claims dataset demonstrate that EmeraldMind achieves competitive accuracy, greater coverage, and superior explanation quality compared to generic LLMs, without the need for fine-tuning or retraining.

cs.AI↗

arXiv2025

Adaptive Indexing for Approximate Query Processing in Exploratory Data Analysis

Minimizing data-to-analysis time while enabling real-time interaction and efficient analytical computations on large datasets are fundamental objectives of contemporary exploratory systems. Although some of the recent adaptive indexing and on-the-fly processing approaches address most of these needs, there are cases, where they do not always guarantee reliable performance. Some examples of such cases include: exploring areas with a high density of objects; executing the first exploratory queries or exploring previously unseen areas (where the index has not yet adapted sufficiently); and working with very large data files on commodity hardware, such as low-specification laptops. In such demanding cases, approximate and incremental techniques can be exploited to ensure efficiency and scalability by allowing users to prioritize response time over result accuracy, acknowledging that exact results are not always necessary. Therefore, approximation mechanisms that enable smooth user interaction by defining the trade-off between accuracy and performance based on vital factors (e.g., task, preferences, available resources) are of great importance. Considering the aforementioned, in this work, we present an adaptive approximate query processing framework for interactive on-the-fly analysis (with out a preprocessing phase) over large raw data. The core component of the framework is a main-memory adaptive indexing scheme (VALINOR-A) that interoperates with user-driven sampling and incremental aggregation computations. Additionally, an effective error-bounded approximation strategy is designed and integrated in the query processing process. We conduct extensive experiments using both real and synthetic datasets, demonstrating the efficiency and effectiveness of the proposed framework.

cs.DB↗

arXiv2025

Visual Analytics Challenges and Trends in the Age of AI: The BigVis Community Perspective

This report provides insights into the challenges, emerging topics, and opportunities related to human-data interaction and visual analytics in the AI era. The BigVis 2024 organizing committee conducted a survey among experts in the field. They invite the Program Committee members and the authors of accepted papers to share their views. Thirty-two scientists from diverse research communities, including Databases, Information Visualization, and Human-Computer Interaction, participated in the study. These scientists, representing both industry and academia, provided valuable insights into the current and future landscape of the field. In this report, we analyze the survey responses and compare them to the findings of a similar study conducted four years ago. The results reveal some interesting insights. First, many of the critical challenges identified in the previous survey remain highly relevant today, despite being unrelated to AI. Meanwhile, the field's landscape has significantly evolved, with most of today's vital challenges not even being mentioned in the earlier survey, underscoring the profound impact of AI-related advancements. By summarizing the perspectives of the research community, this report aims to shed light on the key challenges, emerging trends, and potential research directions in human-data interaction and visual analytics in the AI era.

cs.DB↗

arXiv2024

Partial Adaptive Indexing for Approximate Query Answering

In data exploration, users need to analyze large data files quickly, aiming to minimize data-to-analysis time. While recent adaptive indexing approaches address this need, they are cases where demonstrate poor performance. Particularly, during the initial queries, in regions with a high density of objects, and in very large files over commodity hardware. This work introduces an approach for adaptive indexing driven by both query workload and user-defined accuracy constraints to support approximate query answering. The approach is based on partial index adaptation which reduces the costs associated with reading data files and refining indexes. We leverage a hierarchical tile-based indexing scheme and its stored metadata to provide efficient query evaluation, ensuring accuracy within user-specified bounds. Our preliminary evaluation demonstrates improvement on query evaluation time, especially during initial user exploration.

cs.DB↗

arXiv2023

Auditing for Spatial Fairness

This paper studies algorithmic fairness when the protected attribute is location. To handle protected attributes that are continuous, such as age or income, the standard approach is to discretize the domain into predefined groups, and compare algorithmic outcomes across groups. However, applying this idea to location raises concerns of gerrymandering and may introduce statistical bias. Prior work addresses these concerns but only for regularly spaced locations, while raising other issues, most notably its inability to discern regions that are likely to exhibit spatial unfairness. Similar to established notions of algorithmic fairness, we define spatial fairness as the statistical independence of outcomes from location. This translates into requiring that for each region of space, the distribution of outcomes is identical inside and outside the region. To allow for localized discrepancies in the distribution of outcomes, we compare how well two competing hypotheses explain the observed outcomes. The null hypothesis assumes spatial fairness, while the alternate allows different distributions inside and outside regions. Their goodness of fit is then assessed by a likelihood ratio test. If there is no significant difference in how well the two hypotheses explain the observed outcomes, we conclude that the algorithm is spatially fair.

cs.LG↗

arXiv2023

Data-driven soiling detection in PV modules

Soiling is the accumulation of dirt in solar panels which leads to a decreasing trend in solar energy yield and may be the cause of vast revenue losses. The effect of soiling can be reduced by washing the panels, which is, however, a procedure of non-negligible cost. Moreover, soiling monitoring systems are often unreliable or very costly. We study the problem of estimating the soiling ratio in photo-voltaic (PV) modules, i.e., the ratio of the real power output to the power output that would be produced if solar panels were clean. A key advantage of our algorithms is that they estimate soiling, without needing to train on labelled data, i.e., periods of explicitly monitoring the soiling in each park, and without relying on generic analytical formulas which do not take into account the peculiarities of each installation. We consider as input a time series comprising a minimum set of measurements, that are available to most PV park operators. Our experimental evaluation shows that we significantly outperform current state-of-the-art methods for estimating soiling ratio.

eess.SP↗

arXiv2022

QueryER: A Framework for Fast Analysis-Aware Deduplication over Dirty Data

In this work, we explore the problem of correctly and efficiently answering complex SPJ queries issued directly on top of dirty data. We introduce QueryER, a framework that seamlessly integrates Entity Resolution into Query Processing. QueryER executes analysis-aware deduplication by weaving ER operators into the query plan. The experimental evaluation of our approach exhibits that it adapts to the workload and scales on both real and synthetic datasets.

cs.DB↗

arXiv2018

Hierarchical Characteristic Set Merging for Optimizing SPARQL Queries in Heterogeneous RDF

Characteristic sets (CS) organize RDF triples based on the set of properties characterizing their subject nodes. This concept is recently used in indexing techniques, as it can capture the implicit schema of RDF data. While most CS-based approaches yield significant improvements in space and query performance, they fail to perform well in the presence of schema heterogeneity, i.e., when the number of CSs becomes very large, resulting in a highly partitioned data organization. In this paper, we address this problem by introducing a novel technique, for merging CSs based on their hierarchical structure. Our technique employs a lattice to capture the hierarchical relationships between CSs, identifies dense CSs and merges dense CSs with their ancestors, thus reducing the size of the CSs as well as the links between them. We implemented our algorithm on top of a relational backbone, where each merged CS is stored in a relational table, and we performed an extensive experimental study to evaluate the performance and impact of merging to the storage and querying of RDF datasets, indicating significant improvements.

cs.DB↗

arXiv2017

rdf:SynopsViz - A Framework for Hierarchical Linked Data Visual Exploration and Analysis

The purpose of data visualization is to offer intuitive ways for information perception and manipulation, especially for non-expert users. The Web of Data has realized the availability of a huge amount of datasets. However, the volume and heterogeneity of available information make it difficult for humans to manually explore and analyse large datasets. In this paper, we present rdf:SynopsViz, a tool for hierarchical charting and visual exploration of Linked Open Data (LOD). Hierarchical LOD exploration is based on the creation of multiple levels of hierarchically related groups of resources based on the values of one or more properties. The adopted hierarchical model provides effective information abstraction and summarization. Also, it allows efficient -on the fly- statistic computations, using aggregations over the hierarchy levels.

cs.DB↗

arXiv2016

A Query Language for Multi-version Data Web Archives

The Data Web refers to the vast and rapidly increasing quantity of scientific, corporate, government and crowd-sourced data published in the form of Linked Open Data, which encourages the uniform representation of heterogeneous data items on the web and the creation of links between them. The growing availability of open linked datasets has brought forth significant new challenges regarding their proper preservation and the management of evolving information within them. In this paper, we focus on the evolution and preservation challenges related to publishing and preserving evolving linked data across time. We discuss the main problems regarding their proper modelling and querying and provide a conceptual model and a query language for modelling and retrieving evolving data along with changes affecting them. We present in details the syntax of the query language and demonstrate its functionality over a real-world use case of evolving linked dataset from the biological domain.

cs.DB↗

arXiv2016

graphVizdb: A Scalable Platform for Interactive Large Graph Visualization

We present a novel platform for the interactive visualization of very large graphs. The platform enables the user to interact with the visualized graph in a way that is very similar to the exploration of maps at multiple levels. Our approach involves an offline preprocessing phase that builds the layout of the graph by assigning coordinates to its nodes with respect to a Euclidean plane. The respective points are indexed with a spatial data structure, i.e., an R-tree, and stored in a database. Multiple abstraction layers of the graph based on various criteria are also created offline, and they are indexed similarly so that the user can explore the dataset at different levels of granularity, depending on her particular needs. Then, our system translates user operations into simple and very efficient spatial operations (i.e., window queries) in the backend. This technique allows for a fine-grained access to very large graphs with extremely low latency and memory requirements and without compromising the functionality of the tool. Our web-based prototype supports three main operations: (1) interactive navigation, (2) multi-level exploration, and (3) keyword search on the graph metadata.

cs.HC↗

arXiv2016

A Hierarchical Aggregation Framework for Efficient Multilevel Visual Exploration and Analysis

Data exploration and visualization systems are of great importance in the Big Data era, in which the volume and heterogeneity of available information make it difficult for humans to manually explore and analyse data. Most traditional systems operate in an offline way, limited to accessing preprocessed (static) sets of data. They also restrict themselves to dealing with small dataset sizes, which can be easily handled with conventional techniques. However, the Big Data era has realized the availability of a great amount and variety of big datasets that are dynamic in nature; most of them offer API or query endpoints for online access, or the data is received in a stream fashion. Therefore, modern systems must address the challenge of on-the-fly scalable visualizations over large dynamic sets of data, offering efficient exploration techniques, as well as mechanisms for information abstraction and summarization. In this work, we present a generic model for personalized multilevel exploration and analysis over large dynamic sets of numeric and temporal data. Our model is built on top of a lightweight tree-based structure which can be efficiently constructed on-the-fly for a given set of data. This tree structure aggregates input objects into a hierarchical multiscale model. Considering different exploration scenarios over large datasets, the proposed model enables efficient multilevel exploration, offering incremental construction and prefetching via user interaction, and dynamic adaptation of the hierarchies based on user preferences. A thorough theoretical analysis is presented, illustrating the efficiency of the proposed model. The proposed model is realized in a web-based prototype tool, called SynopsViz that offers multilevel visual exploration and analysis over Linked Data datasets.

cs.HC↗

arXiv2015

Towards Scalable Visual Exploration of Very Large RDF Graphs

In this paper, we outline our work on developing a disk-based infrastructure for efficient visualization and graph exploration operations over very large graphs. The proposed platform, called graphVizdb, is based on a novel technique for indexing and storing the graph. Particularly, the graph layout is indexed with a spatial data structure, i.e., an R-tree, and stored in a database. In runtime, user operations are translated into efficient spatial operations (i.e., window queries) in the backend.

cs.HC↗

arXiv2015

A Framework for Managing Evolving Information Resources on the Data Web

The web of data has brought forth the need to preserve and sustain evolving information within linked datasets; however, a basic requirement of data preservation is the maintenance of the datasets' structural characteristics as well. As open data are often found using different and/or heterogeneous data models and schemata from one source to another, there is a need to reconcile these mismatches and provide common denominations of interpretation on a multitude of levels, in order to be able to preserve and manage the evolution of the generated resources. In this paper, we present a linked data approach for the preservation and archiving of open heterogeneous datasets that evolve through time, at both the structural and the semantic layer. We first propose a set of re-quirements for modelling evolving linked datasets. We then proceed on concep-tualizing a modelling framework for evolving entities and place these in a 2x2 model space that consists of the semantic and the temporal dimensions.

cs.DB↗

arXiv2012

Diachronic Linked Data: Towards Long-Term Preservation of Structured Interrelated Information

The Linked Data Paradigm is one of the most promising technologies for publishing, sharing, and connecting data on the Web, and offers a new way for data integration and interoperability. However, the proliferation of distributed, inter-connected sources of information and services on the Web poses significant new challenges for managing consistently a huge number of large datasets and their interdependencies. In this paper we focus on the key problem of preserving evolving structured interlinked data. We argue that a number of issues that hinder applications and users are related to the temporal aspect that is intrinsic in linked data. We present a number of real use cases to motivate our approach, we discuss the problems that occur, and propose a direction for a solution.

cs.DB↗

arXiv2012

Publishing Life Science Data as Linked Open Data: the Case Study of miRBase

This paper presents our Linked Open Data (LOD) infrastructures for genomic and experimental data related to microRNA biomolecules. Legacy data from two well-known microRNA databases with experimental data and observations, as well as change and version information about microRNA entities, are fused and exported as LOD. Our LOD server assists biologists to explore biological entities and their evolution, and provides a SPARQL endpoint for applications and services to query historical miRNA data and track changes, their causes and effects.

cs.DB↗