[Forum SIS] LET'S TALK ABOUT DATA SCIENCE

[Carlo Lauro]

Dear Paolo welcome to the  DS-sis Forum and thank you for raising two issues that allow me to precise my view.  

1- It is not easy to have a satisfactory  classification of all DS tools in the proposed infographics. It represents just a draft that can be surely improved and completed  taking into account of the suggestions that came from the Forum.  

I was undecided in positioning the machine learning among the indicative  list of  CS tools or at the intersection between CS and Knowledge Domain. Finally I preferred to call this intersection as Pattern Recognition as a research approach  lsimilarly to the one of Data Analysis crssing Statistics and Knowledge Domain 

2- As regard your preference to speak about Data Sciences  rather than Data Science  I fully agree with your view that corresponds to what I stated in my preface to the monograph on Social Data Science published by Springer in 2016. 

The role played by the Knowledge Domain in the Data Science  discovery process represent the major  issue to give sense to the research results  and actionate them. 

As consequence of the large  variety of knowledge domains I believe that  is more appropriate to speak about Data Sciences.

Therefore Data Scientist are active in different areas:  Business Data Science, Social Data Science, Health Data Science, Environment Data Science and so on.. 

This imply that Data Scientist should not only be skilled in Computer Science and Statistics bur also in a Knowledge Domain. 

In addition a suitable coding of the knowledge of a domain (in term logical rules; hierarchical rules, graphs,  metadata; or ontologies ....) could improve a knowledge discovery process by developing in perspective new statistical methods knowledge based.



⁣   Carlo Lauro
Università di Napoli Federico Ii
carlo.lauro at unina.it







Inviato da TypeApp ​

Il giorno 18 Apr 2017, 08:53, alle ore 08:53, Paolo Giudici <paolo.giudici at unipv.it> ha scritto:
>Ciao Carlo, thank you very much for your contribution, on which I
>mostly
>agree.
>I have two remarks.
>
>1. Where do "Machine Learning" fit into your picture? Perhaps along
>"pattern recognition"? or "computational statistics"
>
>
>2. I do not believe that ONE "data science" discipline exist but
>rather,
>than many "DATA SCIENCES" disciplines do, depending on the underlying
>domain knowledge.For example we may have "Social data science"
>"Economics
>data science" "Financial data science" "Medical data science".  Within
>this
>perspective, STATISTICS maintains its "existence" as the logic of
>(learning) in data science, and COMPUTER SCIENCE the "operational"
>language
>by which such logic is implemented.
>
>Looking forward to the debate,
>Paolo Giudici
>
>
>
>Paolo Giudici
>Professor of Statistics and Data Science
>http://economia.unipv.it/pagp/pagine_personali/giudici/
>
>
>
>
>
>
>
>
>
>
>
>Il giorno 14 aprile 2017 18:01, Carlo Lauro <clauro at unina.it> ha
>scritto:
>
>>
>> Data Science aims to transform data into actionable knowledge to
>perform
>> predictions as well to support and validate decisions.
>>
>> Computer Science represents the language of the Data Science whereas
>> Statistics is the Logic of the Data Science itself.
>>
>> However, in this process the domain expertise constitutes the
>catalytic
>> element in the absence of which the transformation can not be
>achieved
>>
>>
>> ---------
>> Sto organizzando una tavola rotonda per la prossima riunione della
>Sis a
>> Firenze sul tema LET'S TALK ABOUT DATA SCIENCE, in un momento in cui
>la
>> Statistica viene spesso posta in discussione dai cosiddetti Data
>Scientist.
>> Vi propongo la mia definizione e una grafica che in cui prospetto il
>ruolo
>> della nostra disciplina.
>> Mi piacerebbe ottenere una vostra opinione o una vostra definizione o
>se
>> ne avete una preferita tra le tante proposte in letteratura e poterne
>> discutere insieme in questo forum.
>> Ringraziandovi per l attenzione colgo l'occasione per augurarvi una
>Santa
>> e serena Pasqua
>>
>>    Carlo Lauro
>> Prifessore Emerito di Statistica
>> Università di Napoli Federico Ii
>> carlo.lauro at unina.it
>>
>>
>>
>>
>>
>>
>>
>> Inviato da TypeApp <http://www.typeapp.com/r?b=9447>
>>
>> _______________________________________________
>> Sis mailing list
>> Sis at stat.unipg.it
>> http://www.stat.unipg.it/mailman/listinfo/sis
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20170418/94364a35/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: photo1.png
Type: image/png
Size: 22044 bytes
Desc: not available
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20170418/94364a35/attachment-0001.png>