Advanced Analytics Specialist Exam for Data Scientists v7.0
Question 1
What is the maximum number of edges in an undirected graph of 10 nodes?
- A. 45
- B. 90
- C. 100
- D. 9
Answer : A
Question 2
In multinomial logistic regression, what is used to calculate the probability of outcome occurring?
- A. Logistic function applied to a linear combination of the input and outcome variables
- B. Linear regression applied to a combination of input variables
- C. Linear regression applied to a combination of input and outcome variables
- D. Logistic function applied to a linear combination of the input variables
Answer : B
Question 3
What is a key beneficial characteristic of the Random Forest algorithm?
- A. Provides and explanatory model
- B. Distinguishes categorical from continuous variables
- C. Support for unstructured data
- D. Resiliency to complex, non-linear variable interactions
Answer : D
Question 4
What is an effective use of color in visualization?
- A. Use self-explanatory colors so a legend is unnecessary
- B. Maximize use of color to make a more lasting impression
- C. Use high contrast colors such as red and blue
- D. Minimize use of color except for emphasis
Answer : C
Question 5
After a client submits a job request to the YARN ResourceManager, what happens next?
- A. The scheduler allocates a container to run an ApplicationMaster
- B. The ResourceManager allocates containers to run map and reduce tasks
- C. The Resource Manager requests load data from the NodeManagers
- D. The ApplicationManager starts an ApplicationMaster
Answer : D
Question 6
Given an input vector of features, a Random Forests model performs a classification task and ends in a tie.
How does the model handle this outcome?
- A. The model will be rebuilt
- B. A winner is chosen at random
- C. The tree that caused the tie is discarded
- D. One more tree is added to the forest
Answer : B
Question 7
What best describes the meaning behind the phrase "Six Degrees of Separation'"?
- A. Ability to use about six hops to reach any other node in an extremely large social network
- B. Erdos number of all scholars having written papers with Paul Erdos
- C. Maximum number of edges between nodes in a graph with a diameter of six
- D. Typical distance between nodes that are connected by triadic closure
Answer : A
Question 8
What is NOT a category of a NoSQL data store?
- A. Columnar
- B. Document
- C. Key/Value
- D. Flat File
Answer : D
Question 9
Which representation is most suitable for a small and highly connected network?
- A. Edge list
- B. Adjacency matrix
- C. Eigenvector centrality
- D. Adjacency list
Answer : B
Question 10
In a social network, what does it mean for a node to have a high degree but low betweenness?
- A. The node is adjacent to a few nodes, each of each has high Page Ranks.
- B. The node has the only edge connecting its community to the rest of the graph.
- C. The node can be easily bypassed by communications taking other shorter paths.
- D. The node acts as the hub of the graph.
Answer : D
Question 11
A hotel chain runs a simul-ation on room pricing. They want to estimate revenue, per hotel, within +/- $10 with 95% confidence (Za/2=1.96). The estimated revenue standard deviation is $5000 based on previous booking data.
What is the optimal number of simulation trials to run?
- A. A 32-bit operating system was used
- B. The same number of trials was used
- C. A linear congruential generator (LCG) was used (or pseudo-random number generation
- D. Different seeds tor the random number generator were used.
Answer : C
Question 12
What process must address acoustic ambiguity in NLP?
- A. Part-of-speech tagging
- B. Word sense disambiguation
- C. Speech recognition
- D. Discourse
Answer : C
Question 13
What runs more efficiently because of Apache Tez?
- A. Pig and Hive
- B. Hive and HBase
- C. Yarn and Spark
- D. All MapReduce jobs
Answer : D
Question 14
Assuming the node index starts at 1, what is the out-degree of node 3 in the adjacency matrix shown?
Refer to the exhibit.
- A. 0
- B. 1
- C. 2
- D. 3
Answer : A
Question 15
Which scenario would be ideal for processing Hadoop data with Hive?
- A. Structured data, real-time processing
- B. Unstructured data; batch processing
- C. Unstructured data; real-time processing
- D. Structured data; batch processing
Answer : B