TY - GEN
T1 - Data Science Failure
T2 - 2023 IEEE International Conference on Big Data, BigData 2023
AU - Lahiri, Sucheta
AU - Saltz, Jeff
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Data science is a multifaceted field that integrates statistics, computer science, social science, and other domains to generate valuable insights from data. Despite unprecedented development, many data science projects fail to achieve desired outcomes. This paper presents a work-in-progress systematic literature review of grey literature to explore the opinions of industry practitioners on data science failure. Specifically, this study reviews trade journals, news articles, blogs, and industry reports published from 2018-2023 to identify common data science failure themes outside of traditional academic literature. Initial findings reveal that technical, process, people, financial, and organizational frictions frequently undermine data science projects. Furthermore, risks related to AI governance, ethical considerations, CRM strategies, data quality, access, and team skills also contribute to data science failure. The analysis highlights the contextual nature of 'failure,' emphasizing the importance of critical thinking that must align with data science goals and business needs. In short, the results suggest that grey literature provides unique perspectives into data science failure, which can be complementary to peer-reviewed scholarship.
AB - Data science is a multifaceted field that integrates statistics, computer science, social science, and other domains to generate valuable insights from data. Despite unprecedented development, many data science projects fail to achieve desired outcomes. This paper presents a work-in-progress systematic literature review of grey literature to explore the opinions of industry practitioners on data science failure. Specifically, this study reviews trade journals, news articles, blogs, and industry reports published from 2018-2023 to identify common data science failure themes outside of traditional academic literature. Initial findings reveal that technical, process, people, financial, and organizational frictions frequently undermine data science projects. Furthermore, risks related to AI governance, ethical considerations, CRM strategies, data quality, access, and team skills also contribute to data science failure. The analysis highlights the contextual nature of 'failure,' emphasizing the importance of critical thinking that must align with data science goals and business needs. In short, the results suggest that grey literature provides unique perspectives into data science failure, which can be complementary to peer-reviewed scholarship.
KW - big data
KW - data science
KW - failure
KW - grey literature
KW - machine learning
UR - http://www.scopus.com/inward/record.url?scp=85184983189&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85184983189&partnerID=8YFLogxK
U2 - 10.1109/BigData59044.2023.10386265
DO - 10.1109/BigData59044.2023.10386265
M3 - Conference contribution
AN - SCOPUS:85184983189
T3 - Proceedings - 2023 IEEE International Conference on Big Data, BigData 2023
SP - 2431
EP - 2440
BT - Proceedings - 2023 IEEE International Conference on Big Data, BigData 2023
A2 - He, Jingrui
A2 - Palpanas, Themis
A2 - Hu, Xiaohua
A2 - Cuzzocrea, Alfredo
A2 - Dou, Dejing
A2 - Slezak, Dominik
A2 - Wang, Wei
A2 - Gruca, Aleksandra
A2 - Lin, Jerry Chun-Wei
A2 - Agrawal, Rakesh
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 15 December 2023 through 18 December 2023
ER -