Big Data - Data Lakes - Bancos de Dados Não SQL

 Big Data - Data Lakes - Bancos de Dados Não SQL 

Exercícios:

 Exercícios:

Exercício 1 A – Obrigatório

Satisfação do Cliente com Numero USP em sinais de interrogação (Rodar em SAS - Weka - Libre Office). DL: 8/5/2023

 

Bu_Unit	Sales	Price	Qu_level	Claims	NPS	PV	Satisfac
1	65,??108	97,??22	96,??419	13,??025	98,??11	19	97,??609
2	15,8371	98,9011	98,3871	12,34568	97,8022	29	98,91304
3	8,885232	100	100	11,11111	100	21	100
4	12,46401	98,9011	95,16129	12,34568	96,7033	94	96,73913
5	80,66639	21,97802	19,35484	100	2,197802	34	21,73913
6	32,16783	23,07692	22,58065	97,53086	3,296703	64	23,91304
7	23,44714	24,17582	24,19355	96,2963	2,747253	61	25
8	89,96298	24,17582	19,35484	95,06173	2,197802	25	26,08696
9	31,4274	64,83516	56,45161	50,61728	65,93407	10	65,21739
10	11,22995	65,93407	51,6129	49,38272	71,42857	3	66,30435
11	77,45784	70,32967	53,22581	46,91358	63,73626	56	68,47826
12	23,89963	68,13187	51,6129	45,67901	61,53846	4	67,3913
13	7,40436	86,81319	80,64516	25,92593	90,10989	90	86,95652
14	0,287947	87,91209	79,03226	24,69136	85,71429	48	85,86957
15	83,42246	87,91209	77,41935	22,22222	90,10989	78	88,04348
16	100	86,81319	75,80645	25,92593	84,61538	88	84,78261

 Exercício 1 B – Facultativo. Criar um exemplo proprio de voces. 

. DL: 15/5/2023 

SAS - Weka - Excel - L. Office - Conventional and Robust Data Science for SML to Prediction or Regression -

 SAS - Weka - Excel - L. Office - Conventional and Robust Data Science for SML to Prediction or Regression - 

Machine Learning Supervisionado para Predição ou Regressão - Excel ou Libre Office

Bu_Unit	Sales	Price	Qu_level	Claims	NPS	PV	Satisfac
1	65,98108	97,8022	96,77419	13,58025	98,9011	19	97,82609
2	15,8371	98,9011	98,3871	12,34568	97,8022	29	98,91304
3	8,885232	100	100	11,11111	100	21	100
4	12,46401	98,9011	95,16129	12,34568	96,7033	94	96,73913
5	80,66639	21,97802	19,35484	100	2,197802	34	21,73913
6	32,16783	23,07692	22,58065	97,53086	3,296703	64	23,91304
7	23,44714	24,17582	24,19355	96,2963	2,747253	61	25
8	89,96298	24,17582	19,35484	95,06173	2,197802	25	26,08696
9	31,4274	64,83516	56,45161	50,61728	65,93407	10	65,21739
10	11,22995	65,93407	51,6129	49,38272	71,42857	3	66,30435
11	77,45784	70,32967	53,22581	46,91358	63,73626	56	68,47826
12	23,89963	68,13187	51,6129	45,67901	61,53846	4	67,3913
13	7,40436	86,81319	80,64516	25,92593	90,10989	90	86,95652
14	0,287947	87,91209	79,03226	24,69136	85,71429	48	85,86957
15	83,42246	87,91209	77,41935	22,22222	90,10989	78	88,04348
16	100	86,81319	75,80645	25,92593	84,61538	88	84,78261
					média=	724

Conventional and Robust Data Science for SML to Prediction or Regression

  Conventional and Robust Data Science for SML to Prediction or Regression 

SAS Program

Data Customer;

Input Bu_Unit  Sales  Price Qu_level Claims NPS Satisfac;

Cards;

1 65.98107775 97.8021978 96.77419355 13.58024691 98.9010989 97.82608696

2 15.83710407 98.9010989 98.38709677 12.34567901 97.8021978 98.91304348

3 8.885232415 100 100 11.11111111 100 100

4 12.46400658 98.9010989 95.16129032 12.34567901 96.7032967 96.73913043

5 80.66639243 21.97802198 19.35483871 100 2.197802198 21.73913043

6 32.16783217 23.07692308 22.58064516 97.5308642 3.296703297 23.91304348

7 23.44714109 24.17582418 24.19354839 96.2962963 2.747252747 25

8 89.9629782 24.17582418 19.35483871 95.0617284 2.197802198 26.08695652

9 31.42739613 64.83516484 56.4516129 50.61728395 65.93406593 65.2173913

10 11.22994652 65.93406593 51.61290323 49.38271605 71.42857143 66.30434783

11 77.45783628 70.32967033 53.22580645 46.91358025 63.73626374 68.47826087

12 23.89962978 68.13186813 51.61290323 45.67901235 61.53846154 67.39130435

13 7.404360346 86.81318681 80.64516129 25.92592593 90.10989011 86.95652174

14 0.287947347 87.91208791 79.03225806 24.69135802 85.71428571 85.86956522

15 83.42245989 87.91208791 77.41935484 22.22222222 90.10989011 88.04347826

16 100 86.81318681 75.80645161 25.92592593 84.61538462 84.7826087

;

proc print; run;

/* Input Bu_Unit  Sales  Price Qu_level Claims NPS Satisfac; */

proc reg;

   model  Satisfac = Sales  Price Qu_level Claims NPS;

Run;

proc robustreg;

model Satisfac = Sales  Price Qu_level Claims NPS;

Run;

 MLS Causas - Efeito ou Regressão ou Predição - Weka

Download Arquivo para Weka

@RELATION Customer

@ATTRIBUTE U_Neg REAL

@ATTRIBUTE Vendas REAL

@ATTRIBUTE Preco REAL

@ATTRIBUTE Niv_Qual REAL

@ATTRIBUTE Reclama REAL

@ATTRIBUTE NPS REAL

@ATTRIBUTE Satisf REAL

@DATA

1,65.98107775,97.8021978,96.77419355,13.58024691,98.9010989,97.82608696

2,15.83710407,98.9010989,98.38709677,12.34567901,97.8021978,98.91304348

3,8.885232415,100,100,11.11111111,100,100

4,12.46400658,98.9010989,95.16129032,12.34567901,96.7032967,96.73913043

5,80.66639243,21.97802198,19.35483871,100,2.197802198,21.73913043

6,32.16783217,23.07692308,22.58064516,97.5308642,3.296703297,23.91304348

7,23.44714109,24.17582418,24.19354839,96.2962963,2.747252747,25

8,89.9629782,24.17582418,19.35483871,95.0617284,2.197802198,26.08695652

9,31.42739613,64.83516484,56.4516129,50.61728395,65.93406593,65.2173913

10,11.22994652,65.93406593,51.61290323,49.38271605,71.42857143,66.30434783

11,77.45783628,70.32967033,53.22580645,46.91358025,63.73626374,68.47826087

12,23.89962978,68.13186813,51.61290323,45.67901235,61.53846154,67.39130435

13,7.404360346,86.81318681,80.64516129,25.92592593,90.10989011,86.95652174

14,0.287947347,87.91208791,79.03225806,24.69135802,85.71428571,85.86956522

15,83.42245989,87.91208791,77.41935484,22.22222222,90.10989011,88.04347826

16,100,86.81318681,75.80645161,25.92592593,84.61538462,84.7826087

Linear Regression Model

Satisf =

      0.2481 * Preco +

      0.1063 * Niv_Qual +

     -0.4275 * Reclama +

      0.1135 * NPS +

     57.4038