?
Social Aspects of Machine Learning Model Evaluation: Model Interpretation and Justification from ML-practitioners' Perspective
Machine Learning (ML) is now widely applied in various life spheres. Experts from different domains become involved in the decision-making on the basis of complex machine learning models that causes in-creased interest in the research in model explainability. However, little is known about the ways that ML-practitioners use to describe and justify their models to others. This work aims to fill the research gap in understanding how data specialists evaluate machine learning models and how they communicate results to third parties. To explore that, the qualitative research design is suggested and semi-structured interviews with ML-practitioners are conducted. The decision-making process will be explored from a sociological perspective according to which data specialists are considered as actors who tend to construct knowledge rather than passively take it. The potential result of this work is to reveal the role of data specialists in model explanation and justification and describe methods they could use to explain complex models to domain experts with non-technical backgroundsMachine Learning (ML) is now widely applied in various life spheres. Experts from different domains become involved in the decision-making on the basis of complex machine learning models that causes in-creased interest in the research in model explainability. However, little is known about the ways that ML-practitioners use to describe and justify their models to others. This work aims to fill the research gap in understanding how data specialists evaluate machine learning models and how they communicate results to third parties. To explore that, the qualitative research design is suggested and semi-structured interviews with ML-practitioners are conducted. The decision-making process will be explored from a sociological perspective according to which data specialists are considered as actors who tend to construct knowledge rather than passively take it. The potential result of this work is to reveal the role of data specialists in model explanation and justification and describe methods they could use to explain complex models to domain experts with non-technical backgroundsMachine Learning (ML) is now widely applied in various life spheres. Experts from different domains become involved in the decision-making on the basis of complex machine learning models that causes in-creased interest in the research in model explainability. However, little is known about the ways that ML-practitioners use to describe and justify their models to others. This work aims to fill the research gap in understanding how data specialists evaluate machine learning models and how they communicate results to third parties. To explore that, the qualitative research design is suggested and semi-structured interviews with ML-practitioners are conducted. The decision-making process will be explored from a sociological perspective according to which data specialists are considered as actors who tend to construct knowledge rather than passively take it. The potential result of this work is to reveal the role of data specialists in model explanation and justification and describe methods they could use to explain complex models to domain experts with non-technical backgroundsMachine Learning (ML) is now widely applied in various life spheres. Experts from different domains become involved in the decision-making on the basis of complex machine learning models that causes in-creased interest in the research in model explainability. However, little is known about the ways that ML-practitioners use to describe and justify their models to others. This work aims to fill the research gap in understanding how data specialists evaluate machine learning models and how they communicate results to third parties. To explore that, the qualitative research design is suggested and semi-structured interviews with ML-practitioners are conducted. The decision-making process will be explored from a sociological perspective according to which data specialists are considered as actors who tend to construct knowledge rather than passively take it. The potential result of this work is to reveal the role of data specialists in model explanation and justification and describe methods they could use to explain complex models to domain experts with non-technical backgroundsMachine Learning (ML) is now widely applied in various life spheres. Experts from different domains become involved in the decision-making on the basis of complex machine learning models that causes in-creased interest in the research in model explainability. However, little is known about the ways that ML-practitioners use to describe and justify their models to others. This work aims to fill the research gap in understanding how data specialists evaluate machine learning models and how they communicate results to third parties. To explore that, the qualitative research design is suggested and semi-structured interviews with ML-practitioners are conducted. The decision-making process will be explored from a sociological perspective according to which data specialists are considered as actors who tend to construct knowledge rather than passively take it. The potential result of this work is to reveal the role of data specialists in model explanation and justification and describe methods they could use to explain complex models to domain experts with non-technical backgroundsMachine Learning (ML) is now widely applied in various life spheres. Experts from different domains become involved in the decision-making on the basis of complex machine learning models that causes in-creased interest in the research in model explainability. However, little is known about the ways that ML-practitioners use to describe and justify their models to others. This work aims to fill the research gap in understanding how data specialists evaluate machine learning models and how they communicate results to third parties. To explore that, the qualitative research design is suggested and semi-structured interviews with ML-practitioners are conducted. The decision-making process will be explored from a sociological perspective according to which data specialists are considered as actors who tend to construct knowledge rather than passively take it. The potential result of this work is to reveal the role of data specialists in model explanation and justification and describe methods they could use to explain complex models to domain experts with non-technical backgrounds
In book
