CoChat

Interactive machine learning for health informatics: when do we need the human-in-the-loop?

Open access
Holzinger, Andreas · 2016 · Brain Informatics

The DOI matches its Crossref record; 105 of the 109 references it cites were found in a registry or an index.

Abstract

Machine learning (ML) is the fastest growing field in computer science, and health informatics is among the greatest challenges. The goal of ML is to develop algorithms which can learn and improve over time and can be used for predictions. Most ML researchers concentrate on automatic machine learning (aML), where great advances have been made, for example, in speech recognition, recommender systems, or autonomous vehicles. Automatic approaches greatly benefit from big data with many training sets. However, in the health domain, sometimes we are confronted with a small number of data sets or rare events, where aML-approaches suffer of insufficient training samples. Here interactive machine learning (iML) may be of help, having its roots in reinforcement learning, preference learning, and active learning. The term iML is not yet well used, so we define it as "algorithms that can interact with agents and can optimize their learning behavior through these interactions, where the agents can also be human." This "human-in-the-loop" can be beneficial in solving computationally hard problems, e.g., subspace clustering, protein folding, or k-anonymization of health data, where human expertise can help to reduce an exponential search space through heuristic selection of samples. Therefore, what would otherwise be an NP-hard problem, reduces greatly in complexity through the input and the assistance of a human agent involved in the learning phase.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1007/s40708-016-0042-6

Cites

105 of 105 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Four different ML-pipelines: A unsupervised, B supervised—e.g., humans are providing labels for training data sets and/or select features, C semi-supervised, D shows the iML human-in-the-loop approach: the important issue is that humans are not only involved in pre-processing, by selecting data or features, but actually during the learning phase, directly interacting with the algorithm, thus shifting away the black-box problem to a wished glass-box, 1 input data, 2 pre-processing phase, 3 human agent(s) interacting with the computational agent(s), allowing for crowdsourcing or gamification approaches, 4 final check done by the human expert
Fig. 1 Four different ML-pipelines: A unsupervised, B supervised—e.g., humans are providing labels for training data sets and/or select features, C semi-supervised, D shows the iML human-in-the-loop approach: the important issue is that humans are not only involved in pre-processing, by selecting data or features, but actually during the learning phase, directly interacting with the algorithm, thus shifting away the black-box problem to a wished glass-box, 1 input data, 2 pre-processing phase, 3 human agent(s) interacting with the computational agent(s), allowing for crowdsourcing or gamification approaches, 4 final check done by the human expertCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 105 references

Research works printed but not linked: 5

  1. Mitchell TM (1997) Machine learning. McGraw Hill, New York
  2. Fahlman SE, Hinton GE, Sejnowski TJ (1983) Massively parallel architectures for Al: NETL, Thistle, and Boltzmann machines. In: Genesereth MR (ed) AAAI-83. AAAI, Washington, DC, pp 109–113
  3. Wilson AG, Dann C, Lucas, Xing EP (2015) The human kernel. arXiv preprint arXiv:1510.07389 · arXiv
  4. Agarwal S, Cortes C, Herbrich R (2005) Learning to rank. In: Weiss Y, Schlkopf B, Platt JC (eds) NIPS Workshop
  5. Lynch NA (1996) Distributed algorithms. Morgan Kaufmann, San Francisco

Cited by

See all 1,002

Metadata

Authors
Holzinger, A.
Affiliations
University of Graz (AT)
Corresponding author
Holzinger, Andreas
Venue
Brain Informatics
Published
2016-03-02
Volume
3
Issue
2
Pages
119-131
ISSN
  • 2198-4018 (ISSN-L)
  • 2198-4026
OpenAlex
W2300445845
Cited by
906
References
109
Citations by year
2026: 93 · 2025: 85 · 2024: 87 · 2023: 74 · 2022: 111 · 2021: 132 · 2020: 110 · 2019: 77 · 2018: 53 · 2017: 46 · 2016: 38
FWCI
58.63
Type
journal article
Language
en
Open access
Locations
  • Brain Informatics (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Algorithms
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
Keywords
interactive machine learning, human-in-the-loop, health informatics, active learning (machine learning), preference learning, reinforcement learning, limited training data, rare events, human-agent interaction, NP-hard problems, k-anonymization, protein folding, subspace clustering

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Interactive machine learning for health informatics: when do we need the human-in-the-loop? https://app.cochat.ai/papers/10.1007/s40708-016-0042-6
The paper
Holzinger, A. (2016). Interactive machine learning for health informatics: when do we need the human-in-the-loop? Brain Informatics. https://doi.org/10.1007/s40708-016-0042-6

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.