Data science
{{Short description|Field of study to extract knowledge from data}}
{{Distinguish|information science|computer science}}
{{Use dmy dates|date=August 2023}}
File: PIA23792-1600x1200(1).jpg (here depicted as a series of red dots) was discovered by analyzing astronomical survey data acquired by a space telescope, the Wide-field Infrared Survey Explorer.]]
Data science is an interdisciplinary academic field{{Cite journal |last1=Donoho |first1=David |title=50 Years of Data Science |doi=10.1080/10618600.2017.1384734 |journal=Journal of Computational and Graphical Statistics |year=2017 |volume=26 |issue=4 |pages=745–766 |s2cid=114558008 |doi-access=free}} that uses statistics, scientific computing, scientific methods, processing, scientific visualization, algorithms and systems to extract or extrapolate knowledge from potentially noisy, structured, or unstructured data.{{Cite journal |last1=Dhar |first1=V. |title=Data science and prediction |doi=10.1145/2500499 |journal=Communications of the ACM |volume=56 |issue=12 |pages=64–73 |year=2013 |s2cid=6107147 |url=http://cacm.acm.org/magazines/2013/12/169933-data-science-and-prediction/fulltext |access-date=2 September 2015 |archive-url=https://web.archive.org/web/20141109113411/http://cacm.acm.org/magazines/2013/12/169933-data-science-and-prediction/fulltext |archive-date=9 November 2014 |url-status=live}}
Data science also integrates domain knowledge from the underlying application domain (e.g., natural sciences, information technology, and medicine).{{cite report |last1=Danyluk |first1=A. |last2=Leidig |first2=P. |date=2021 |title=Computing Competencies for Undergraduate Data Science Curricula |work=ACM Data Science Task Force Final Report |url=https://dstf.acm.org/DSTF_Final_Report.pdf}} Data science is multifaceted and can be described as a science, a research paradigm, a research method, a discipline, a workflow, and a profession.{{Cite journal |last1=Mike |first1=Koby |last2=Hazzan |first2=Orit |date=2023-01-20 |title=What is Data Science? |journal=Communications of the ACM |volume=66 |issue=2 |pages=12–13 |doi=10.1145/3575663 |issn=0001-0782|doi-access=free }}
Data science is "a concept to unify statistics, data analysis, informatics, and their related methods" to "understand and analyze actual phenomena" with data.{{Cite book |chapter-url=https://www.springer.com/book/9784431702085 |title=Data Science, Classification, and Related Methods |last=Hayashi |first=Chikio |chapter=What is Data Science ? Fundamental Concepts and a Heuristic Example |date=1998-01-01 |publisher=Springer Japan |isbn=9784431702085 |editor-last=Hayashi |editor-first=Chikio |series=Studies in Classification, Data Analysis, and Knowledge Organization |pages=40–51 |language=en |doi=10.1007/978-4-431-65950-1_3 |editor-last2=Yajima |editor-first2=Keiji |editor-last3=Bock |editor-first3=Hans-Hermann |editor-last4=Ohsumi |editor-first4=Noboru |editor-last5=Tanaka |editor-first5=Yutaka |editor-last6=Baba |editor-first6=Yasumasa}} It uses techniques and theories drawn from many fields within the context of mathematics, statistics, computer science, information science, and domain knowledge.{{Cite journal |last=Cao |first=Longbing |date=2017-06-29 |title=Data Science: A Comprehensive Overview |journal=ACM Computing Surveys |volume=50 |issue=3 |pages=43:1–43:42 |doi=10.1145/3076253 |s2cid=207595944 |issn=0360-0300|doi-access=free |arxiv=2007.03606 }} However, data science is different from computer science and information science. Turing Award winner Jim Gray imagined data science as a "fourth paradigm" of science (empirical, theoretical, computational, and now data-driven) and asserted that "everything about science is changing because of the impact of information technology" and the data deluge.{{cite book |author1=Tony Hey |author2=Stewart Tansley |author3=Kristin Michele Tolle |title=The Fourth Paradigm: Data-intensive Scientific Discovery |url=https://books.google.com/books?id=oGs_AQAAIAAJ |year=2009 |publisher=Microsoft Research |isbn=978-0-9825442-0-4 |archive-url=https://web.archive.org/web/20170320193019/https://books.google.com/books?id=oGs_AQAAIAAJ |archive-date=20 March 2017 |url-status=live}}{{cite journal |last1=Bell |first1=G. |last2=Hey |first2=T. |last3=Szalay |first3=A. |title=Computer Science: Beyond the Data Deluge |journal=Science |volume=323 |issue=5919 |year=2009 |pages=1297–1298 |issn=0036-8075 |doi=10.1126/science.1170411 |pmid=19265007 |s2cid=9743327}}
A data scientist is a professional who creates programming code and combines it with statistical knowledge to summarize data.{{Cite journal |title=Data Scientist: The Sexiest Job of the 21st Century |url=https://hbr.org/2012/10/data-scientist-the-sexiest-job-of-the-21st-century/ |journal=Harvard Business Review |date=October 2012 |access-date=2016-01-18 |last1=Davenport |first1=Thomas H. |last2=Patil |first2=D. J. |volume=90 |issue=10 |pages=70–76, 128 |pmid=23074866}}
Foundations
Data science is an interdisciplinary field{{Cite journal |title=Defining data science by a data-driven quantification of the community |journal=Machine Learning and Knowledge Extraction |year=2018 |doi=10.3390/make1010015 |doi-access=free |last1=Emmert-Streib |first1=Frank |last2=Dehmer |first2=Matthias |volume=1 |pages=235–251 }} focused on extracting knowledge from typically large data sets and applying the knowledge from that data to solve problems in other application domains. The field encompasses preparing data for analysis, formulating data science problems, analyzing data, and summarizing these findings. As such, it incorporates skills from computer science, mathematics, data visualization, graphic design, communication, and business.{{Cite web |url=https://www.oreilly.com/library/view/doing-data-science/9781449363871/ch01.html |title=1. Introduction: What Is Data Science? |work=Doing Data Science [Book] |publisher=O’Reilly |language=en |access-date=2020-04-03}}
Vasant Dhar writes that statistics emphasizes quantitative data and description. In contrast, data science deals with quantitative and qualitative data (e.g., from images, text, sensors, transactions, customer information, etc.) and emphasizes prediction and action.{{Cite journal |author=Vasant Dhar |date=2013-12-01 |title=Data science and prediction |journal=Communications of the ACM |volume=56 |issue=12 |pages=64–73 |language=en |doi=10.1145/2500499 |s2cid=6107147|url=http://archive.nyu.edu/handle/2451/31553 }} Andrew Gelman of Columbia University has described statistics as a non-essential part of data science.{{Cite web |url=https://statmodeling.stat.columbia.edu/2013/11/14/statistics-least-important-part-data-science/ |title=Statistics is the least important part of data science « Statistical Modeling, Causal Inference, and Social Science |website=statmodeling.stat.columbia.edu |access-date=2020-04-03}} Stanford professor David Donoho writes that data science is not distinguished from statistics by the size of datasets or use of computing and that many graduate programs misleadingly advertise their analytics and statistics training as the essence of a data-science program. He describes data science as an applied field growing out of traditional statistics.
Etymology
= Early usage =
In 1962, John Tukey described a field he called "data analysis", which resembles modern data science.{{Cite web |url=http://courses.csail.mit.edu/18.337/2015/docs/50YearsDataScience.pdf |title=50 years of Data Science |last=Donoho |first=David |date=18 September 2015 |access-date=2 April 2020}} In 1985, in a lecture given to the Chinese Academy of Sciences in Beijing, C. F. Jeff Wu used the term "data science" for the first time as an alternative name for statistics.{{Cite journal |url=https://www2.isye.gatech.edu/~jeffwu/publications/fazhan.pdf |title=Future directions of statistical research in China: a historical perspective |last1=Wu |first1=C. F. Jeff |journal=Application of Statistics and Management |volume=1 |year=1986 |pages=1–7 |access-date=29 November 2020}} Later, attendees at a 1992 statistics symposium at the University of Montpellier II acknowledged the emergence of a new discipline focused on data of various origins and forms, combining established concepts and principles of statistics and data analysis with computing.{{Cite book |title=Data science and its applications |publisher=Academic Press/Harcourt Brace |editor-last1=Escoufier |editor-first1=Yves |editor-last2=Hayashi |editor-first2=Chikio |editor-last3=Fichet |editor-first3=Bernard |year=1995 |isbn=0-12-241770-4 |location=Tokyo |oclc=489990740}}{{Cite journal |last1=Murtagh |first1=Fionn |last2=Devlin |first2=Keith |date=2018 |title=The Development of Data Science: Implications for Education, Employment, Research, and the Data Revolution for Sustainable Development |journal=Big Data and Cognitive Computing |language=en |volume=2 |issue=2 |pages=14 |doi=10.3390/bdcc2020014 |doi-access=free}}
The term "data science" has been traced back to 1974, when Peter Naur proposed it as an alternative name to computer science. In 1996, the International Federation of Classification Societies became the first conference to specifically feature data science as a topic. However, the definition was still in flux. After the 1985 lecture at the Chinese Academy of Sciences in Beijing, in 1997 C. F. Jeff Wu again suggested that statistics should be renamed data science. He reasoned that a new name would help statistics shed inaccurate stereotypes, such as being synonymous with accounting or limited to describing data.{{Cite web |url=http://www2.isye.gatech.edu/~jeffwu/presentations/datascience.pdf |title=Statistics=Data Science? |last=Wu |first=C. F. Jeff |access-date=2 April 2020}} In 1998, Hayashi Chikio argued for data science as a new, interdisciplinary concept, with three aspects: data design, collection, and analysis.
= Modern usage =
In 2012, technologists Thomas H. Davenport and DJ Patil declared "Data Scientist: The Sexiest Job of the 21st Century",{{cite magazine |last=Davenport |first=Thomas |date=2012-10-01 |title=Data Scientist: The Sexiest Job of the 21st Century |url=https://hbr.org/2012/10/data-scientist-the-sexiest-job-of-the-21st-century |magazine=Harvard Business Review |access-date=2022-10-10}} a catchphrase that was picked up even by major-city newspapers like the New York Times{{cite news |last=Miller |first=Claire |date=2013-04-04 |title=Data Science: The Numbers of Our Lives |url=https://www.nytimes.com/2013/04/14/education/edlife/universities-offer-courses-in-a-hot-new-field-data-science.html |work=New York Times |location=New York City |access-date=2022-10-10}} and the Boston Globe.{{cite news |last=Borchers |first=Callum |date=2015-11-11 |title=Behind the scenes of the 'sexiest job of the 21st century' |url=https://www.bostonglobe.com/business/2015/11/11/behind-scenes-sexiest-job-century/Kc1cvXIu31DfHhVmyRQeIJ/story.html |work=Boston Globe |location=Boston |access-date=2022-10-10}} A decade later, they reaffirmed it, stating that "the job is more in demand than ever with employers".{{cite magazine |last=Davenport |first=Thomas |date=2022-07-15 |title=Is Data Scientist Still the Sexiest Job of the 21st Century? |url=https://hbr.org/2022/07/is-data-scientist-still-the-sexiest-job-of-the-21st-century |magazine=Harvard Business Review |access-date=2022-10-10}}
The modern conception of data science as an independent discipline is sometimes attributed to William S. Cleveland.{{Cite web |url=https://www.stat.purdue.edu/~wsc/ |title=William S. Cleveland |last=Gupta |first=Shanti |date=11 December 2015 |access-date=2 April 2020}} In 2014, the American Statistical Association's Section on Statistical Learning and Data Mining changed its name to the Section on Statistical Learning and Data Science, reflecting the ascendant popularity of data science.{{Cite news |last=Talley |first=Jill |url=https://magazine.amstat.org/blog/2016/06/01/datascience-2/ |title=ASA Expands Scope, Outreach to Foster Growth, Collaboration in Data Science |date=1 June 2016 |work=Amstat News |publisher=American Statistical Association}}. In 2013 the first European Conference on Data Analysis (ECDA2013) started in Luxembourg the process which founded the European Association for Data Science (EuADS) www.euads.org in Luxembourg in 2015.
The professional title of "data scientist" has been attributed to DJ Patil and Jeff Hammerbacher in 2008.{{Cite news |last1=Davenport |first1=Thomas H. |url=https://hbr.org/2012/10/data-scientist-the-sexiest-job-of-the-21st-century |title=Data Scientist: The Sexiest Job of the 21st Century |date=2012-10-01 |work=Harvard Business Review |access-date=2020-04-03 |last2=Patil |first2=D. J. |issue=October 2012 |issn=0017-8012}} Though it was used by the National Science Board in their 2005 report "Long-Lived Digital Data Collections: Enabling Research and Education in the 21st Century", it referred broadly to any key role in managing a digital data collection.{{Cite web |url=https://www.nsf.gov/pubs/2005/nsb0540/ |title=US NSF – NSB-05-40, Long-Lived Digital Data Collections Enabling Research and Education in the 21st Century |website=www.nsf.gov |access-date=2020-04-03}}
Data science and data analysis
File:EDA example - Always plot your data.jpg as demonstrated using the Datasaurus dozen data set]]
File:Data Science.png and domain expertise.]]
Data analysis typically involves working with structured datasets to answer specific questions or solve specific problems. This can involve tasks such as data cleaning and data visualization to summarize data and develop hypotheses about relationships between variables. Data analysts typically use statistical methods to test these hypotheses and draw conclusions from the data.{{Cite book |title=An Introduction to Statistical Learning: with Applications in R. |last1=James |first1=Gareth |author1-link=Gareth M. James |last2=Witten|first2=Daniela| author2-link=Daniela Witten |last3=Hastie |first3=Trevor |author3-link=Trevor Hastie |last4=Tibshirani |first4=Robert | author4-link=Robert Tibshirani |date=2017-09-29 |publisher=Springer |language=en}}
Data science involves working with larger datasets that often require advanced computational and statistical methods to analyze. Data scientists often work with unstructured data such as text or images and use machine learning algorithms to build predictive models. Data science often uses statistical analysis, data preprocessing, and supervised learning.{{Cite web |url=https://www.researchgate.net/publication/256438799|title=Data Science for Business: What You Need to Know about Data Mining and Data-Analytic Thinking. |last1=Provost |first1=Foster |date=2013-08-01 |website=O'Reilly Media, Inc. |language=en|last2=Tom Fawcett }}{{Cite book |url=https://www.sciencedirect.com/book/9780123814791/data-mining-concepts-and-techniques|title=Data Mining: Concepts and Techniques. |last1=Han |first1=Kamber |date=2011 |language=en|last2=Pei |isbn=9780123814791 }}
Cloud computing for data science
File:Cloud computing in enabling data science at scale.jpg, laptops, and smart phones, through cloud services for processing and analysis, finally leading to various big data applications.]]
Cloud computing can offer access to large amounts of computational power and storage.{{Cite journal |last1=Hashem |first1=Ibrahim Abaker Targio |last2=Yaqoob |first2=Ibrar |last3=Anuar |first3=Nor Badrul |last4=Mokhtar |first4=Salimah |last5=Gani |first5=Abdullah |last6=Ullah Khan |first6=Samee |date= 2015|title=The rise of "big data" on cloud computing: Review and open research issues |url=https://linkinghub.elsevier.com/retrieve/pii/S0306437914001288 |journal=Information Systems |language=en |volume=47 |pages=98–115 |doi=10.1016/j.is.2014.07.006}} In big data, where volumes of information are continually generated and processed, these platforms can be used to handle complex and resource-intensive analytical tasks.{{Cite journal |last1=Qiu |first1=Junfei |last2=Wu |first2=Qihui |last3=Ding |first3=Guoru |last4=Xu |first4=Yuhua |last5=Feng |first5=Shuo |date= 2016|title=A survey of machine learning for big data processing |journal=EURASIP Journal on Advances in Signal Processing |language=en |volume=2016 |issue=1 |doi=10.1186/s13634-016-0355-x |doi-access=free |issn=1687-6180}}
Some distributed computing frameworks are designed to handle big data workloads. These frameworks can enable data scientists to process and analyze large datasets in parallel, which can reduce processing times.{{Cite book |last1=Armbrust |first1=Michael |last2=Xin |first2=Reynold S. |last3=Lian |first3=Cheng |last4=Huai |first4=Yin |last5=Liu |first5=Davies |last6=Bradley |first6=Joseph K. |last7=Meng |first7=Xiangrui |last8=Kaftan |first8=Tomer |last9=Franklin |first9=Michael J. |last10=Ghodsi |first10=Ali |last11=Zaharia |first11=Matei |chapter=Spark SQL: Relational Data Processing in Spark |date=2015-05-27 |title=Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data |chapter-url=https://dl.acm.org/doi/10.1145/2723372.2742797 |language=en |publisher=ACM |pages=1383–1394 |doi=10.1145/2723372.2742797 |isbn=978-1-4503-2758-9}}
Ethical consideration in data science
Data science involves collecting, processing, and analyzing data which often includes personal and sensitive information. Ethical concerns include potential privacy violations, bias perpetuation, and negative societal impacts.{{Cite journal |last1=Floridi |first1=Luciano |author1-link=Luciano Floridi |last2=Taddeo |first2=Mariarosaria |author2-link=Mariarosaria Taddeo |date=2016-12-28 |title=What is data ethics? |journal=Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences |language=en |volume=374 |issue=2083 |pages=20160360 |doi=10.1098/rsta.2016.0360 |issn=1364-503X |pmc=5124072 |pmid=28336805|bibcode=2016RSPTA.37460360F }}{{Cite journal |last1=Mittelstadt |first1=Brent Daniel |last2=Floridi |first2=Luciano |date= 2016|title=The Ethics of Big Data: Current and Foreseeable Issues in Biomedical Contexts |url=http://link.springer.com/10.1007/s11948-015-9652-2 |journal=Science and Engineering Ethics |language=en |volume=22 |issue=2 |pages=303–341 |doi=10.1007/s11948-015-9652-2 |pmid=26002496 |issn=1353-3452}}
Machine learning models can amplify existing biases present in training data, leading to discriminatory or unfair outcomes.{{Cite journal |last1=Barocas|first1=Solon|last2=Selbst|first2=Andrew D |via=Berkeley Law Library Catalog |date=2016 |title=Big Data's Disparate Impact |url=https://lawcat.berkeley.edu/record/1127463 |journal=California Law Review |doi=10.15779/Z38BG31}}{{Cite journal |last1=Caliskan |first1=Aylin |last2=Bryson |first2=Joanna J. |author2-link=Joanna Bryson |last3=Narayanan |first3=Arvind |author3-link=Arvind Narayanan |date=2017-04-14 |title=Semantics derived automatically from language corpora contain human-like biases |url=https://www.science.org/doi/10.1126/science.aal4230 |journal=Science |language=en |volume=356 |issue=6334 |pages=183–186 |doi=10.1126/science.aal4230 |arxiv=1608.07187 |bibcode=2017Sci...356..183C |issn=0036-8075}}