3 ms·
In part 1/3, the author writes that there are 2 branches of data science: > Data science for people (Type A), i.e. analytics to support evidence-based decision
by mastazi 10y ago
In part 1/3, the author writes that there are 2 branches of data science:
> Data science for people (Type A), i.e. analytics to support evidence-based decision making
> Data science for software (Type B), for example: recommender systems as we see in Netflix and Spotify
Isn't "type A" business intelligence, and isn't "type B" machine learning? Why doesn't he use those more widely known terms? Or maybe he is referring to something else?
- preygel 10y agoThe distinction is being made on end outputs -- is it more like a Powerpoint deck, or more like a production data pipeline -- not on techniques. "Machine learning" includes plenty of activities that can be used to provide evidence for one-off human decision making (e.g., using a model to produce forecasts or to understand sensitivities).
- mastazi 10y agoMmmmh yes, often ML produces data which supports human decision making; I see what you mean. In fact, now that you mention it, in my company the BI team is starting to use ML techniques to recognise some particular patterns.
- SatvikBeri 10y agoBoth business intelligence and machine learning are narrower terms with a more specific meaning. Business intelligence has the connotation of a certain set of techniques (e.g. reporting, SQL, querying, PowerPoint presentations) while machine learning is a fairly specific set of tools that is much narrower than what most software data scientists do. E.g. I'm a software data scientist and I spend much more time on descriptive statistics than machine learning.
- mastazi 10y agoYes, I agree that both ML and BI are narrower definitions compared to that of data science.
- bigger_cheese 10y agoMy current job title is "Data Engineer" I'm not sure how that differs from a "Data Scientist". My Job basically aligns with Type A above. I mostly work on optimizing our industrial process through a combination of modelling and simulation work. Other than that I do quality/defect investigations when we have issues with defective batches etc and I do yield optimization work. I also oversee various plant trials as required. I use Business Intelligence software (like SAS and COGNOS) but I use other tools as well (including my own C/C++ code). I lean heavily on my own theoretical knowledge - in particular metallurgy and minerals processing. (I am a Materials Engineer by qualification). I think most BI people would lack the background theory to do my role. My Job title was more or less arbitrarily chosen by my manager (Other people in my team have 'Automation Engineer', 'Process Engineer' etc. as their role titles). I consider myself an Engineer above anything else.
- apahwa 10y agoThat seems like a pretty inaccurate job title then. Data Engineers are people working with data pipelines, storage, and schemas. They can lean towards more software engineering or towards analytics with dashboarding/machine learning but their primary responsibilities are the former.
- bigger_cheese 10y agoYeah it is a pretty generic title. Regardless it fits my job description pretty well I am an Engineer I work with data. The people who do the work you mention (schemas etc) are called Database Admins here. They work for Information Services - different department I'm not all that familiar with.