final exam

Question 1

Append

Answer 1

Add records from one dataset to another

Question 2

Merge

Answer 2

Add fields from one dataset to another

Question 3

Rectangular Data

Answer 3

Product of records and fields

Question 4

Stages of CRISPDM

Answer 4

Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, Deployment

Question 5

SuperNode

Answer 5

Condensing several nodes into a single node

Question 6

Histogram is used for which fields

Answer 6

Continuous Fields

Question 7

Distribution is used for which fields

Answer 7

Categorical Fields

Question 8

Direct link between 2 variables

Answer 8

Causation

Question 9

2 variables change at a certain rate in relationship to eachother

Answer 9

Correlation

Question 10

Data point that deviates so far from the other observations

Answer 10

Outlier/Extreme

Question 11

Values 3-5 SD from the mean

Answer 11

Outlier

Question 12

Values more then 5 SD away from the mean

Answer 12

Extreme (Outlier)

Question 13

2 or more categories that can be ranked (School ranking)

Answer 13

Ordinal Ranking

Question 14

2 or more categories that can be ranked, that have no order (peoples favorite color)

Answer 14

Nominal Ranking

Question 15

All the numbers added together then divided by how many number there are (average)

Answer 15

Mean

Question 16

The number in the middle when you arrange a set from least to greatest

Answer 16

Median

Question 17

The number that appears most often in the set

Answer 17

Mode

Question 18

1) Rank,
2) Fractional rank as %
3) Sum of case weights
4) Savage score
5)Fractional rank

Answer 18

Options for ranking models

Question 19

Estimates and compares models for continuous numeric range outcomes

Answer 19

AutoNumeric

Question 20

Estimates and compares models for either nominal or binary targers

Answer 20

AutoClassifier

Question 21

Testing multiple models, this node would present the results for all models, for both partitions so you can easily determine which model performed the best

Answer 21

Running an Analysis Node

Question 22

1)Undefined values represented as $null$
2) White Spaces
3)Values that are not in the allowed set of values,

Answer 22

Invalid Values

Question 23

Define an area of certain size (Space-time-box)

Answer 23

Geohash

Question 24

Models in this category predict a target field, using one or more predictors

Answer 24

Supervised Models

Question 25

These models create groups of records with similar values on the input field

Answer 25

Unsupervised Models (Segmentation)

Question 26

The process of extracting valuable insights from larger datasets, it helps organizations make data-driven decisions understand customer behavior. By uncovering patterns and relationships in data, businesses can gain a competitive edge and improve efficiency.

Answer 26

Data Mining

Question 27

Modify unit of analysis, remove duplicates, create a dataset with one record per customer

Answer 27

Distinct Node

Question 28

Only data from records present in all source datasets will be merged

Answer 28

Inner Join

Question 29

Automatically create new nominal fields based on the values of one or more existing continuous field

Answer 29

Binning Node

Question 30

1) First m – Returns/Discards the first M records in in dataset 2)1-in-n - Every nth record is selected/discarded 3)Random % - There is a r% probability of each record being selected/discarded

Answer 30

Simple Node

Question 31

Measures distance from the center (mean)

Answer 31

Standard Deviation

Question 32

Sample from groups of records rather than from individual records

Answer 32

Clustered Sample

Question 33

Sample independently within subgroups

Answer 33

Stratified Sample

Question 34

Relationship between 2 Categorical Fields

Answer 34

Matrix, Distribution

Question 35

Relationship between 1 Categorical Fields, 1 Continuous Field

Answer 35

Means, Histogram

Question 36

Relationship between 2 Continuous Fields

Answer 36

Statistics, Plot

Question 37

Process of reading or specifying information such as measurement levels and values for a field

Answer 37

Instantiating Data

Question 38

A field has unknown storage

Answer 38

Uninstantiated

Question 39

K-Means, Kohohen, Two-Step

Answer 39

3 segmentation methods