Safety and Risk Analytics Quiz
Uploaded by
Vidhi GabaSafety and Risk Analytics Quiz
Uploaded by
Vidhi GabaAssignment 3
Safety and Risk Analytics
1. The extent to which the data are well documented verifiable and easily attributed to a source
is defined as
a) Relevancy of data
b) Believability of data
c) Flexibility of data
d) Traceability of data
Ans: (d) Traceability of data
Solution: Refer to Lecture 11
2. Interpretability of data is a part of
a) Intrinsic data quality
b) Contextual data quality
c) Representational data quality
d) Accessibility data quality
Ans: (c) Representational data quality
Solution: Refer to Lecture 11
3. Fill in the blank. is a type of missing data mechanism.
a) NCAR
b) NAR
c) NMAR
d) MMAR
Ans: (c) NMAR
Solution: Refer to Lecture 12
4. The method that involves imputing missing data on a variable for a given case by matching
that case with other cases in the data set on several other key variables that have complete
information for those cases.
a) Hot deck imputation
b) Maximum Likelihood Expectation-Maximization Imputation
c) Regression Imputation
d) All the above
Ans: (d) All of the above
Solution: Refer to Lecture 12
5. Suppose the data quality value of the dimensions, interpretability, and ease of understanding,
representational consistency and concise representation of data are 0.70, 0.85, 0.55, and 0.60,
respectively. If the average of the data quality scores of the dimensions gives the category
score, which of the following is true?
a) Intrinsic quality = 0.675
b) Representational quality = 0.575
c) Representational quality =0.675
d) Contextual data quality = 0.775
Ans: Representational data quality = (0.70+0.85+0.55+0.60)/4 = 0.675
Solution: Refer to Lecture 11
Question 6-8 are based on the following description.
Consider the following data table of sample observations.
Incident path Probability Consequence
(IP) (P) (C)
IP1 VH 1
IP2 M 4
IP3 H 2
IP4 L ?
IP5 VL 5
Here, data for probability of occurrence of incident paths are provided in linguistic terms.
Also, consider the following fuzzy scale
Linguistic variable Trapezoidal fuzzy numbers
Very Low ( VL) (0,0,0.1,0.2)
Low (L) (0.2,0.3,0.3,0.4)
Medium (M) (0.3,0.4,0.5,0.6)
High (H) (0.6,0.7,0.7,0.8)
Very High (VH) (0.8,0.9,1,1)
6. Using the mean imputation method, the missing value of consequence (C) for IP4 is
a) 3
b) 1
c) 4
d) 5
Ans: Option (a)
Solution: Refer to Lecture 12
(1+4+2+5)/4 = 3
7. Consider the complete data table by including the imputed value obtained in Q.6. Using the
z-score normalization, the transformed values of consequences (C) for IP2 and IP4,
respectively are
a) 0, 0.63
b) -0.63, 0
c) 0, -0.63
d) 0.63, 0
Ans: Option (d)
Solution: Refer to Lecture 13
4−3
(CIP 2 ) Normalized =
(CIP 2 − meanC ) / st.divC = =
0.63
1.58
3−3
(CIP 4 ) Normalized =
(CIP 4 − meanC ) / st.divC = =
0
1.58
8. Consider the data table and fuzzy scale in Q.6. The transformed defuzzified values of
probability (P) for IP2 and IP4, respectively are
a) 0.15, 0.08
b) 0.3, 0.16
c) 0.45, 0.3
d) 0.08, 0.15
Ans: Option (c)
Solution: Refer to Lecture 14
1 1
−a1a2 + a3 a4 − (a2 − a1 ) 2 + (a4 − a3 ) 2
3 3
−a1 − a2 + a3 + a4
1 1
−(0.3 × 0.4) + (0.5 × 0.6) − (0.4 − 0.3) 2 + (0.6 − 0.5) 2
3 3 0.18
=
P = = 0.45
−0.3 − 0.4 + 0.5 + 0.6
IP 2
0.4
1 1
−(0.2 × 0.3) + (0.3 × 0.4) − (0.3 − 0.2) 2 + (0.4 − 0.3) 2
3 3 0.06
=
PIP 4 = = 0.3
−0.2 − 0.3 + 0.3 + 0.4 0.2
9. The non-parametric approach for replacing the original data volume by alternative smaller
forms of data representation is/are
a) Principal Component Analysis
b) Histograms
c) Both a and b
d) None of the above
Ans: Option (b)
Solution: Refer to Lecture 15
10.
Consider the following contingency table.
I1 I2 I3 I4
Q1 2 4 7 6
Q2 5 1 1 2
Q3 7 3 2 3
Q4 3 2 2 1
Where, Q1, Q2, Q3, and Q4 are quarter of a year
I1, I2, I3, and I4 are type of incidents in the year.
The similarities and differences amongst the four type of incidents with respect to the four
quarters can be obtained by computing
a) Row profile of the contingency table
b) Column profile of the contingency table
c) Weighted Chi-square distances
d) Singular value decomposition
Ans: Option (b).
Solution: Refer to Lecture 16