Home / Current Issue / Paper 1722574
Whether the Data Scientist Role Survives the Rise of Automated Analysis
Subject area: Science,Engineering and Technology · Area of research: Data Science
Abstract
In a familiar scenario, fear that the technology that made an occupation possible and popular will make the job superfluous is now re-emerging in a new arena: automated analysis. With more and more technical tasks being done by an automated machine learning system, commentators wonder whether or not the data scientist will exist! It is the aim of this paper to try to answer this question, and to suggest that the role may remain, but it must be redefined, and that the redefinition has a pattern which is well-studied in the history of automation. The paper builds on a body of research dating back to 2015 and up to 2023 covering the automation of employment, the economics of artificial intelligence, and how the role of data scientist has itself evolved, to explain what automated analysis replaces and what it does not. It argues that automation is for the components of data science, model building, model tuning, and routine analysis that are executable, well-defined, and repeatable; and that the framing of the problem, the exercise of judgment, and the integration of the analysis into organizational decision-making are left to humans. Since a job consists of a set of tasks, some of which can be automated, automation of some of the tasks results in a different mix of tasks in the job, not its disappearance, and that the human contribution to the tasks that cannot be automated should be redirected toward tasks that can be automated, thereby effectively making them more valuable, in line with the principle of complementarity. The end result of the paper is that the data scientist will not disappear, but will shift from a technical execution to a judgmental, problem-defining, and organizational-translating role, and that the work of the data scientist and the profession must embrace this change, not fear it.
Keywords
data scientist, automated analysis, automation of employment, prediction and judgment, complementarity, role redefinition
How to cite this paper
@article{1722574,
author = {Ozoya Amos Ohilebo},
title = {Whether the Data Scientist Role Survives the Rise of Automated Analysis},
journal = {Iconic Research And Engineering Journals},
year = {2023},
volume = {7},
number = {5},
pages = {481-487},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1722574.pdf},
abstract = {In a familiar scenario, fear that the technology that made an occupation possible and popular will make the job superfluous is now re-emerging in a new arena: automated analysis. With more and more technical tasks being done by an automated machine learning system, commentators wonder whether or not the data scientist will exist! It is the aim of this paper to try to answer this question, and to suggest that the role may remain, but it must be redefined, and that the redefinition has a pattern which is well-studied in the history of automation. The paper builds on a body of research dating back to 2015 and up to 2023 covering the automation of employment, the economics of artificial intelligence, and how the role of data scientist has itself evolved, to explain what automated analysis replaces and what it does not. It argues that automation is for the components of data science, model building, model tuning, and routine analysis that are executable, well-defined, and repeatable; and that the framing of the problem, the exercise of judgment, and the integration of the analysis into organizational decision-making are left to humans. Since a job consists of a set of tasks, some of which can be automated, automation of some of the tasks results in a different mix of tasks in the job, not its disappearance, and that the human contribution to the tasks that cannot be automated should be redirected toward tasks that can be automated, thereby effectively making them more valuable, in line with the principle of complementarity. The end result of the paper is that the data scientist will not disappear, but will shift from a technical execution to a judgmental, problem-defining, and organizational-translating role, and that the work of the data scientist and the profession must embrace this change, not fear it.},
keywords = {data scientist, automated analysis, automation of employment, prediction and judgment, complementarity, role redefinition},
month = {November},
}