Jump to content

Data thinking

Add links
From Wikipedia, the free encyclopedia
This is an old revision of this page, as edited by Howldev (talk | contribs) at 16:51, 31 December 2022 (commas, spelling and linking). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Recent years have seen the integration of computer science, mathematics, and statistics, together with real-world domain knowledge, into a new research and applications field: data science. Just as data science integrates knowledge and skills from computer science, statistics, and a real-world application domain, data thinking integrates computational thinking, statistical thinking, and domain thinking.[1]

In the context of new product development and innovation, data thinking can be described as follows: Data thinking is a framework to explore, design, develop and validate data-driven solutions and user, data and future-focused businesses. Data thinking combines data science with design thinking and therefore, the focus of this approach does not lie only on data analytics technologies and data collection but also on the design of user-centered solutions with high business potential. [2][3][4][5]

The term was created by Mario Faria and Rogerio Panigassi in 2013 when they were writing a book about data science, data analytics, data management, and how data practitioners were able to achieve their goals.

Major Components of Data Thinking

Computational thinking was first introduced by Papert[6] and, a quarter of a century later, was illuminated and elaborated on by Wing.[7] The term statistical thinking was coined by Deming[8] and developed by Moore.[9] Examining the data science thinking skills in relation to the disciplines that make up data science, it can be seen that each discipline contributes its unique thinking skills: computer science brings computational thinking, statistics brings statistical thinking, with each domain bringing the thinking skills rooted in the said domain of knowledge as well. According to Mike et al.[1]:

  • Data thinking is the understanding that a solution to a real-life problem should not be based only on data and algorithms, but also on the domain knowledge-driven rules that govern them.
  • Data thinking asks whether the data offer a good representation of the real-life situation. It also addresses how data were collected and asks, “Can the data collection be improved?”.
  • Data thinking is the understanding that data are not just numbers to be stored in an adequate data structure, but that these numbers have a meaning that derives from the domain knowledge.
  • Data thinking is understanding that any process or calculation performed on the data should preserve the meaning of the relevant knowledge domain.
  • Data thinking analyzes the data not only logically but also statistically, using visualizations and statistical methods to find patterns as well as irregular phenomena.
  • Data thinking is understanding that problem abstraction is domain-depended, and generalization is subject to biases and variance in the data.
  • Data thinking is understanding that lab testing is not enough, and that real-life implementation will always encounter unexpected data and situations, and so improving the models and the solution for a given problem is a continuous process that includes, among other activities, constant and iterative monitoring and data collection

Major Phases of Data Thinking

Even though no standardized process for data thinking yet exists, the major phases of the process are similar in many publications and could be summarized as follows:

Clarification of the Strategic Context and definition of data-driven risks and opportunities focus areas

During this phase, the broader context of digital strategy is analyzed. Before starting with a concrete data project, it is essential to understand how the new data and AI-driven technologies are affecting the business landscape and the implications this has on the future of an organization. Trend analysis / technology forecasting and scenario planning/analysis as well as internal data capability assessments are the major techniques that are typically applied at this stage. [10][4]

Ideation/Exploration

The result of the earlier stage is a definition of the focus areas which are either the most promising or are at the highest risks for or due to data-driven transformation. At the Ideation/exploration phase, the concrete use cases are defined for the selected focus areas. For successful Ideation, it is important to combine information about organizational (business) goals, internal/external use needs, data and infrastructure needs as well as domain knowledge about the latest data-driven technologies and trends.  [11][3]

Design thinking principles in the context of data thinking can be interpreted as follows: when developing data-driven ideas, it is crucial to consider the intersection of technical feasibility, business impact and data availability. Typical instruments of design thinking (e.g. user research, personas, customer journey) are broadly applied at this stage. [12]

But not only the user, but customer and strategic needs of an organization must also be considered here. Data needs and data availability analysis as well as research on the AI technologies suitable for the data-based solution are essential parts of the successful development process. [13]

To scope data and the technological basement of the solution, practices from cross-industry standard processes for data mining (CRISP-DM) are typically utilized on this stage. [14]

Prototyping / Proof of Concept

During the previous stages, the major concept of the data solution was developed. At the current step, the proof of concept is conducted to check its feasibility. This stage also exploits the prototype framework of design thinking and includes testing, evaluation, iteration, and refinement.[15] Prototyping design thinking principles are also combined during this phase with process models that are applied in data science projects (e.g. CRISP-DM).[10]

Measuring business impact

Not only solution feasibility but also its profitability is proven during the data thinking process. Cost benefits analysis and business case calculation are commonly applied during this step.[16]

Implementation and Improvement

If the developed solution proves its feasibility and profitability during this phase, it will be implemented and operationalized. [2][4]

References

  1. ^ a b Mike, Koby; Ragonis, Noa; Rosenberg-Kima, Rinat B.; Hazzan, Orit (2022-07-21). "Computational thinking in the era of data science". Communications of the ACM. 65 (8): 33–35. doi:10.1145/3545109. ISSN 0001-0782.
  2. ^ a b "Why do companies need Data Thinking?". 2020-07-02.
  3. ^ a b "Data Thinking - Mit neuer Innovationsmethode zum datengetriebenen Unternehmen" [With new innovation methods to the data-driven company] (in German).
  4. ^ a b c "Data Thinking: A guide to success in the digital age".
  5. ^ Herrera, Sara (2019-02-21). "Data-Thinking als Werkzeug für KI-Innovation" [Data Thinking as a tool for KI-innovation]. Handelskraft (in German).
  6. ^ Papert, Seymour A. (1980). Mindstorms Children, Computers, and Powerful Ideas. New York: Basic Books. ISBN 978-1-5416-7510-0. OCLC 1314616065.
  7. ^ Wing, Jeannette M. (2006). "Computational thinking". Communications of the ACM. 49 (3): 33–35. doi:10.1145/1118178.1118215. ISSN 0001-0782.
  8. ^ Deming, W. Edwards (2000). Out of the crisis (1st ed.). Cambridge, Mass.: MIT Press. ISBN 0-262-54115-7. OCLC 46934182.
  9. ^ On the shoulders of giants : new approaches to numeracy. Lynn Arthur Steen, National Research Council. Mathematical Sciences Education Board. Washington, D.C.: National Academy Press. 1990. ISBN 0-585-26843-6. OCLC 45730083.{{cite book}}: CS1 maint: others (link)
  10. ^ a b Schnakenburg, Igor; Kuhn, Steffen. "Data Thinking: Daten schnell produktiv nutzen können". LÜNENDONK-Magazin "Künstliche Intelligenz" (in German). 05/2020: 42–46.
  11. ^ Nalchigar, Soroosh; Yu, Eric (2018-09-01). "Business-driven data analytics: A conceptual modeling framework". Data & Knowledge Engineering. 117: 359–372. doi:10.1016/j.datak.2018.04.006. ISSN 0169-023X. S2CID 53096729.
  12. ^ Woods, Rachel (2019-03-22). "A Design Thinking Mindset for Data Science". Medium. Retrieved 2020-07-08.
  13. ^ Fomenko, Elena; Mattgey, Annette (2020-05-12). "Was macht eigentlich … ein Data Thinker?". W & V. German.
  14. ^ Marbán, Óscar; Mariscal, Gonzalo; Menasalvas, Ernestina; Segovia, Javier (2007). Yin, Hujun; Tino, Peter; Corchado, Emilio; Byrne, Will; Yao, Xin (eds.). "An Engineering Approach to Data Mining Projects". Intelligent Data Engineering and Automated Learning - IDEAL 2007. Lecture Notes in Computer Science. 4881. Berlin, Heidelberg: Springer: 578–588. doi:10.1007/978-3-540-77226-2_59. ISBN 978-3-540-77226-2.
  15. ^ Brown, Tim Wyatt, Jocelyn (2010-07-01). "Design Thinking for Social Innovation". Development Outreach. 12 (1): 29–43. doi:10.1596/1020-797X_12_1_29. hdl:10986/6068. ISSN 1020-797X.{{cite journal}}: CS1 maint: multiple names: authors list (link)
  16. ^ "Data-Thinking – das Potenzial von Daten richtig nutzen". t3n Magazin (in German). 2018-09-08. Retrieved 2020-07-08.
Retrieved from "https://en.wikipedia.org/w/index.php?title=Data_thinking&oldid=1130719835"