A Framework for Provenance Analysis and Visualization

Oliveira, Weiner de; Ambrósio, Lenita M.; Braga, Regina; Ströele, Victor; David, José; Campos, Fernanda

doi:10.1016/j.procs.2017.05.216

Cited by 20 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The last group of categories is subsequent to visualization tools and techniques that oversee the ML pipeline, leading to knowledge generation in the overall workflow. Evaluation of models and meeting user expectations [CRMH12] is a key component for people to trust or not ML model(s) for a task. Agreement of colleagues is supported by visualizations with provenance [OAB*17, RESC16] and collaborative visualizations which facilitate, for instance, ten experts from diverse domains to agree that a model performed well. This purpose could be served by provenance features and specific glyphs or snapshots, along with web‐based online tools and platforms.…”

Section: Background: Levels Of Trustworthiness Of Machine Learningmentioning

confidence: 99%

“…Agreement of colleagues is supported by visualizations with provenance [OAB*17, RESC16] and collaborative visualizations which facilitate, for instance, ten experts from diverse domains to agree that a model performed well. This purpose could be served by provenance features and specific glyphs or snapshots, along with web‐based online tools and platforms.…”

Section: Background: Levels Of Trustworthiness Of Machine Learningmentioning

confidence: 99%

See 1 more Smart Citation

The State of the Art in Enhancing Trust in Machine Learning Models with the Use of Visualizations

Chatzimparmpas

Martins

Jusufi

et al. 2020

Computer Graphics Forum

112

View full text Add to dashboard Cite

Machine learning (ML) models are nowadays used in complex applications in various domains, such as medicine, bioinformatics, and other sciences. Due to their black box nature, however, it may sometimes be hard to understand and trust the results they provide. This has increased the demand for reliable visualization tools related to enhancing trust in ML models, which has become a prominent topic of research in the visualization community over the past decades. To provide an overview and present the frontiers of current research on the topic, we present a State‐of‐the‐Art Report (STAR) on enhancing trust in ML models with the use of interactive visualization. We define and describe the background of the topic, introduce a categorization for visualization techniques that aim to accomplish this goal, and discuss insights and opportunities for future research directions. Among our contributions is a categorization of trust against different facets of interactive ML, expanded and improved from previous research. Our results are investigated from different analytical perspectives: (a) providing a statistical overview, (b) summarizing key findings, (c) performing topic analyses, and (d) exploring the data sets used in the individual papers, all with the support of an interactive web‐based survey browser. We intend this survey to be beneficial for visualization researchers whose interests involve making ML models more trustworthy, as well as researchers and practitioners from other disciplines in their search for effective visualization techniques suitable for solving their tasks with confidence and conveying meaning to their data.

show abstract

Section: Background: Levels Of Trustworthiness Of Machine Learningmentioning

confidence: 99%

Section: Background: Levels Of Trustworthiness Of Machine Learningmentioning

confidence: 99%

The State of the Art in Enhancing Trust in Machine Learning Models with the Use of Visualizations

Chatzimparmpas

Martins

Jusufi

et al. 2020

Computer Graphics Forum

112

View full text Add to dashboard Cite

show abstract

“…There is a clear opportunity for human computer interaction (HCI) research to assist in ensuring that decision provenance approaches enable a representation that assists endusers regarding their accountability concerns. A number of techniques have been suggested as ways in which provenance data can be made more usable and understandable, including Natural Language Interfaces for Databases (NLIDBs) [53], data visualisation techniques (such as graphs and plots) [50], [51], [53], as well as using online games [54] and comics [55] as a means for describing captured provenance information to end users. Generally, more work is required to explore the presentation of such data for accountability purposes, in ways that support the various perspectives (including users, technical experts, regulators, auditors, etc.)…”

Section: Usability Of Captured Datamentioning

confidence: 99%

Decision Provenance: Harnessing Data Flow for Accountable Systems

2019

View full text Add to dashboard Cite

Demand is growing for more accountability regarding the technological systems that increasingly occupy our world. However, the complexity of many of these systems -often systems-of-systems -poses accountability challenges. A key reason for this is because the details and nature of the information flows that interconnect and drive systems, which often occur across technical and organisational boundaries, tend to be invisible or opaque. This paper argues that data provenance methods show much promise as a technical means for increasing the transparency of these interconnected systems. Specifically, given the concerns regarding ever-increasing levels of automated and algorithmic decision-making, and so-called 'algorithmic systems' in general, we propose decision provenance as a concept showing much promise. Decision provenance entails using provenance methods to provide information exposing decision pipelines: chains of inputs to, the nature of, and the flow-on effects from the decisions and actions taken (at design and run-time) throughout systems. This paper introduces the concept of decision provenance, and takes an interdisciplinary (tech-legal) exploration into its potential for assisting accountability in algorithmic systems. We argue that decision provenance can help facilitate oversight, audit, compliance, risk mitigation, and user empowerment, and we also indicate the implementation considerations and areas for research necessary for realising its vision. More generally, we make the case that considerations of data flow, and systems more broadly, are important to discussions of accountability, and complement the considerable attention already given to algorithmic specifics.

show abstract

“…These features form the upper layers of the semantic web layer-cake, namely, unified logic, proof, trust, and user-interface and application (Sizov, 2007). The purpose of these layers is to provide the provenance (Moreau et al , 2015; Oliveira et al , 2017) of the contents, incorporating the individuals, institutions, data sources, data manipulation methodologies and algorithms, and so forth regarding the contents. Using the provenance, the layers of unified logic, proof, trust, and user-interface are built.…”

Section: Case Studymentioning

confidence: 99%

Hidden Markov model-based digital twin construction for futuristic manufacturing systems

Ghosh

Ullah

Kubo

2019

AIEDAM

View full text Add to dashboard Cite

This paper addresses the construction of digital twins (exact mirror images of real-world in cyberspace) using hidden Markov models for the futuristic manufacturing systems known as Industry 4.0. The proposed digital twin consists of two components namely model component and simulation component. The model component forms a Markov chain that encapsulates the dynamics underlying the phenomenon by using some discrete states and their transition probabilities. The simulation component recreates the phenomenon using a Monte Carlo simulation process. The efficacy of the proposed digital twin construction methodology is shown by a case study, where the digital twin of the surface roughness of a surface created by successive grinding operations is described. The developers of the cyber-physical systems will be benefitted from the outcomes of this study because these systems need the computable virtual abstractions of the manufacturing phenomena to address the issues related to the maturity index of futuristic manufacturing systems (i.e., understand, predict, decide, and adopt).

show abstract

A Framework for Provenance Analysis and Visualization

Cited by 20 publications

References 8 publications

The State of the Art in Enhancing Trust in Machine Learning Models with the Use of Visualizations

The State of the Art in Enhancing Trust in Machine Learning Models with the Use of Visualizations

Decision Provenance: Harnessing Data Flow for Accountable Systems

Hidden Markov model-based digital twin construction for futuristic manufacturing systems

Contact Info

Product

Resources

About