Explaining data to laymen
Sassoon Kosian
Corporate Vice President, Data Science @ New York Life | Data Science, AI solutions
As a society that has made data much more central in pretty much every area of life and has become more data-driven than ever, we are not as good at explaining data-driven decisions to every layer of society. COVID-19 is a case to the point.
It seems that the link between our actions and data pointing to the rise or decline in cases on one hand, and the link between rising cases leading to recommendations and decisions reacting to that specific spike seen in the data is straightforward to those of us who work with data. Yet we see many people across the globe who seriously doubt seemingly simple facts observed in the data. Not to mention widely spread conspiracy theories that defy factual data and common sense. Why is it so difficult to explain data to laymen? Are we even trying?
It seems that the rapid spread of big data in the last decade has not been followed by similar democratization of understanding and appreciating data and decisions made using it. As a data scientist working in an industry where it's extremely important to be able to explain my machine learning models and data analytics, my day to day job consists of breaking down complex data science insights into simple explanations that my business partners can understand. I contrast my experience with what's going on in society at large. I feel that we are doing a poor job explaining data to non-data people. It's a failure of data professionals and scientists that we haven't tried or not hard enough to really get the message across.
Using data is not going to stop. It will continue growing exponentially and we data professionals and scientists are very excited about it. To get the rest of the world on board we better start explaining it to them a little better. In that, the government has some role, but I believe most of the effort should come from the "data" industry. I would be curious to hear how other data professionals feel.
#data #analytics #dataanalytics #datascience #machinelearning #artificialintelligence
Data Quality Analyst | AI Activist, Evangelist, and Skeptic | Lifelong Learner | Husband & Father | Cross Trained Mind
4 年Well written and very much on point. I think that data communication is on the level of science communication; the quality of the data/science doesn't matter if people don't understand. This is why I'm interested in the data storytelling movement and related areas.
Sassoon Kosian great article! Do visit INDIAai the central hub for everything AI in India and beyond. A joint initiative of MeitY, NeGD and NASSCOM. You can browse through our selection of more than 1000 pieces of content on the platform, covering topics from?reinforcement learning?to?neuromorphic chips?to?key government AI initiatives.? Would be great to have you on board as one of the contributors.
Explaining AI is the key to industry success. Key decision makers understand the “ROI” language while Data Scientists talk “Accuracy”. There has to be a translator between “Accuracy” to “ROI” for any AI project to take off.
We need to design a system that converts Data into presentable and easily understood information using psychology to help the non data people in the market currently instead of having everyone go through an unnecessary learning curve. #PISIQ is here to help you with that. #AI for Everyone.
The word Data has been used in every context in every part of the organisation, but the questions is really do they know how they are supposed to be using that data to gain valuable business outcomes? I think understanding its value is far more crucial.