The Need of Standardised Metadata to Encode Causal Relationships: Towards Safer Data-Driven Machine Learning Biological Solutions

Beatriz Garcia Santa Cruz; Carlos Vega; Frank Hertel

doi:10.1007/978-3-031-20837-9_16

The Need of Standardised Metadata to Encode Causal Relationships: Towards Safer Data-Driven Machine Learning Biological Solutions

Beatriz Garcia Santa Cruz^*, Carlos Vega, Frank Hertel

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

In this paper, we discuss the importance of considering causal relations in the development of machine learning solutions to prevent factors hampering the robustness and generalisation capacity of the models, such as induced biases. This issue often arises when the algorithm decision is affected by confounding factors. In this work, we argue that the integration of research assumptions as causal relationships can help identify potential confounders. Together with metadata information, it can enable meta-comparison of data acquisition pipelines. We call for standardised meta-information practices as a crucial step for proper machine learning solutions development, validation, and data sharing. Such practices include detailing the data acquisition process, aiming for automatic integration of causal relationships and actionable metadata.

Original language	English
Title of host publication	Computational Intelligence Methods for Bioinformatics and Biostatistics - 17th International Meeting, CIBB 2021, Revised Selected Papers
Editors	Davide Chicco, Angelo Facchiano, Erica Tavazzi, Enrico Longato, Martina Vettoretti, Anna Bernasconi, Simone Avesani, Paolo Cazzaniga
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	200-216
Number of pages	17
ISBN (Print)	9783031208362
DOIs	https://doi.org/10.1007/978-3-031-20837-9_16
Publication status	Published - 2022
Externally published	Yes
Event	17th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2021 - Virtual, Online Duration: 15 Nov 2021 → 17 Nov 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13483 LNBI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	17th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2021
City	Virtual, Online
Period	15/11/21 → 17/11/21

Keywords

Causality
Confounders
Machine learning
Metadata
Systems biology

Access to Document

10.1007/978-3-031-20837-9_16

Cite this

Garcia Santa Cruz, B., Vega, C., & Hertel, F. (2022). The Need of Standardised Metadata to Encode Causal Relationships: Towards Safer Data-Driven Machine Learning Biological Solutions. In D. Chicco, A. Facchiano, E. Tavazzi, E. Longato, M. Vettoretti, A. Bernasconi, S. Avesani, & P. Cazzaniga (Eds.), Computational Intelligence Methods for Bioinformatics and Biostatistics - 17th International Meeting, CIBB 2021, Revised Selected Papers (pp. 200-216). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13483 LNBI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-20837-9_16

Garcia Santa Cruz, Beatriz ; Vega, Carlos ; Hertel, Frank. / The Need of Standardised Metadata to Encode Causal Relationships : Towards Safer Data-Driven Machine Learning Biological Solutions. Computational Intelligence Methods for Bioinformatics and Biostatistics - 17th International Meeting, CIBB 2021, Revised Selected Papers. editor / Davide Chicco ; Angelo Facchiano ; Erica Tavazzi ; Enrico Longato ; Martina Vettoretti ; Anna Bernasconi ; Simone Avesani ; Paolo Cazzaniga. Springer Science and Business Media Deutschland GmbH, 2022. pp. 200-216 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{a89ed702987647408ea4588caac2dc16,

title = "The Need of Standardised Metadata to Encode Causal Relationships: Towards Safer Data-Driven Machine Learning Biological Solutions",

abstract = "In this paper, we discuss the importance of considering causal relations in the development of machine learning solutions to prevent factors hampering the robustness and generalisation capacity of the models, such as induced biases. This issue often arises when the algorithm decision is affected by confounding factors. In this work, we argue that the integration of research assumptions as causal relationships can help identify potential confounders. Together with metadata information, it can enable meta-comparison of data acquisition pipelines. We call for standardised meta-information practices as a crucial step for proper machine learning solutions development, validation, and data sharing. Such practices include detailing the data acquisition process, aiming for automatic integration of causal relationships and actionable metadata.",

keywords = "Causality, Confounders, Machine learning, Metadata, Systems biology",

author = "{Garcia Santa Cruz}, Beatriz and Carlos Vega and Frank Hertel",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 17th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2021 ; Conference date: 15-11-2021 Through 17-11-2021",

year = "2022",

doi = "10.1007/978-3-031-20837-9_16",

language = "English",

isbn = "9783031208362",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "200--216",

editor = "Davide Chicco and Angelo Facchiano and Erica Tavazzi and Enrico Longato and Martina Vettoretti and Anna Bernasconi and Simone Avesani and Paolo Cazzaniga",

booktitle = "Computational Intelligence Methods for Bioinformatics and Biostatistics - 17th International Meeting, CIBB 2021, Revised Selected Papers",

address = "Germany",

}

Garcia Santa Cruz, B, Vega, C & Hertel, F 2022, The Need of Standardised Metadata to Encode Causal Relationships: Towards Safer Data-Driven Machine Learning Biological Solutions. in D Chicco, A Facchiano, E Tavazzi, E Longato, M Vettoretti, A Bernasconi, S Avesani & P Cazzaniga (eds), Computational Intelligence Methods for Bioinformatics and Biostatistics - 17th International Meeting, CIBB 2021, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13483 LNBI, Springer Science and Business Media Deutschland GmbH, pp. 200-216, 17th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2021, Virtual, Online, 15/11/21. https://doi.org/10.1007/978-3-031-20837-9_16

The Need of Standardised Metadata to Encode Causal Relationships: Towards Safer Data-Driven Machine Learning Biological Solutions. / Garcia Santa Cruz, Beatriz; Vega, Carlos; Hertel, Frank.
Computational Intelligence Methods for Bioinformatics and Biostatistics - 17th International Meeting, CIBB 2021, Revised Selected Papers. ed. / Davide Chicco; Angelo Facchiano; Erica Tavazzi; Enrico Longato; Martina Vettoretti; Anna Bernasconi; Simone Avesani; Paolo Cazzaniga. Springer Science and Business Media Deutschland GmbH, 2022. p. 200-216 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13483 LNBI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - The Need of Standardised Metadata to Encode Causal Relationships

T2 - 17th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2021

AU - Garcia Santa Cruz, Beatriz

AU - Vega, Carlos

AU - Hertel, Frank

PY - 2022

Y1 - 2022

N2 - In this paper, we discuss the importance of considering causal relations in the development of machine learning solutions to prevent factors hampering the robustness and generalisation capacity of the models, such as induced biases. This issue often arises when the algorithm decision is affected by confounding factors. In this work, we argue that the integration of research assumptions as causal relationships can help identify potential confounders. Together with metadata information, it can enable meta-comparison of data acquisition pipelines. We call for standardised meta-information practices as a crucial step for proper machine learning solutions development, validation, and data sharing. Such practices include detailing the data acquisition process, aiming for automatic integration of causal relationships and actionable metadata.

AB - In this paper, we discuss the importance of considering causal relations in the development of machine learning solutions to prevent factors hampering the robustness and generalisation capacity of the models, such as induced biases. This issue often arises when the algorithm decision is affected by confounding factors. In this work, we argue that the integration of research assumptions as causal relationships can help identify potential confounders. Together with metadata information, it can enable meta-comparison of data acquisition pipelines. We call for standardised meta-information practices as a crucial step for proper machine learning solutions development, validation, and data sharing. Such practices include detailing the data acquisition process, aiming for automatic integration of causal relationships and actionable metadata.

KW - Causality

KW - Confounders

KW - Machine learning

KW - Metadata

KW - Systems biology

UR - http://www.scopus.com/inward/record.url?scp=85144316560&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-20837-9_16

DO - 10.1007/978-3-031-20837-9_16

M3 - Conference contribution

AN - SCOPUS:85144316560

SN - 9783031208362

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 200

EP - 216

BT - Computational Intelligence Methods for Bioinformatics and Biostatistics - 17th International Meeting, CIBB 2021, Revised Selected Papers

A2 - Chicco, Davide

A2 - Facchiano, Angelo

A2 - Tavazzi, Erica

A2 - Longato, Enrico

A2 - Vettoretti, Martina

A2 - Bernasconi, Anna

A2 - Avesani, Simone

A2 - Cazzaniga, Paolo

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 15 November 2021 through 17 November 2021

ER -

Garcia Santa Cruz B, Vega C, Hertel F. The Need of Standardised Metadata to Encode Causal Relationships: Towards Safer Data-Driven Machine Learning Biological Solutions. In Chicco D, Facchiano A, Tavazzi E, Longato E, Vettoretti M, Bernasconi A, Avesani S, Cazzaniga P, editors, Computational Intelligence Methods for Bioinformatics and Biostatistics - 17th International Meeting, CIBB 2021, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2022. p. 200-216. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-20837-9_16