Quiz
CMSC 435 Midterm Review
Create a free account to finish this quiz
Save your progress and see your final score.
Already have an account? Sign in
Questions in this quiz (28)
1The typically largest bigdata that are analyzed (in contrast to being collected) in recent years by data science practitioners is between:
- A100 megabytes and 1 terabyte
- B1 megabyte and 100 megabytes
- C1 petabyte and 100 petabytes
2 Knowledge Discovery aims to identify patterns that are:
- Avalid, novel, and potentially useful
- Bautomated and compact
- Caccurate and multi-dimensional
3 Based on the six-step Data Mining and Knowledge Discovery process model, which of the
following is true:- AThe data mining step should be performed before the preparation of the data step
- BThe using the discovered knowledge step should be performed after the evaluation of the discovered knowledge step
- CThe data mining step should be performed before the understanding the data step
4Commonly used synonyms for features are
- Avariables, instances and data points
- Btuples, columns and data points
- Cattributes, dimensions and variables
5Datasets where objects (also called examples) are described by multiple features are called:
- Anominal
- Bmultivariate
- Cunivariate
6SMOTE algorithm is used to
- AOversample small/minority classes
- BReduce the amount of noise
- CImpute missing values
7Sequential processing of several disjoint subsets of a training dataset does not lower
runtime when compared to processing of the entire training dataset for algorithms that have:- AO(n3 ) asymptotic complexity
- BO(n2 ) asymptotic complexity
- CO(n) asymptotic complexity
8Common data quality issues do not include
- Aincompleteness
- Brandomness
- Cmissing values
9The top ten data mining algorithms according to the “Top 10 algorithms in data mining”
article do not include- APageRank
- BCART
- CDataSqueezer
10Which of the missing data types cannot be accurately imputed:
- Amissing completely at random
- Bmissing not at random
- Cmissing at random
11Asymptotic complexity of the hot deck imputation method is:
- AO(n)
- BO(nlogn)
- CO(n2 )
12A four-dimensional data cube contains the following number of distinct cuboids, including
the apex and the base cuboids- A16
- B4
- C8
13A Data Warehouse that uses galaxy schema has:
- Aone fact table and one dimension table
- Bmultiple fact tables and any number of dimension tables
- Cone fact table and any number of dimension tables
14In the following formula that is used for the information retrieval: tfij = fij / maxi(fij), fij is
- Afrequency of terms i and j in document f
- Bfrequency of term f in documents i and j
- Cfrequency of term i in document j
15The amount of data that are currently available worldwide is below one zettabyte.
- ATrue
- BFalse
16Project goals are translated into Data Mining goals during the Preparation of the Data step in the six-step Data Mining and Knowledge Discovery process.
- ATrue
- BFalse
17Datasets that have over 50% of missing data cannot be used to generate rules
- ATrue
- BFalse
18The parameter k used in the kNN algorithm, which is described in the “Top 10 algorithms in data
mining” article, defines the number of nearest neighbors- ATrue
- BFalse
19Some of the data science algorithms have log-linear complexity, i.e., O(nlog(n)).
- ATrue
- BFalse
20A nominal feature has natural ordering between its values
- ATrue
- BFalse
21Noise in a training dataset can be reduced with the help of Tomek links.
- ATrue
- BFalse
22The mean imputation is typically more accurate than the hot deck imputation.
- ATrue
- BFalse
23The key characteristics of Data Warehouses include the time-variant and nonvolatile nature of the
underlying data.- ATrue
- BFalse
24The drill down and roll up commands in OLAP are typically implemented with a SELECT statement
that includes GROUP BY clause.- ATrue
- BFalse
25The snowflake schema is less suitable to implement concept hierarchy in a Data Warehouse than the star schema.
- ATrue
- BFalse
26The top three most popular programming languages for Data Science include Python, C and
Javascript.- ATrue
- BFalse
27RapidMiner is among the most popular software tools for data science.
- ATrue
- BFalse
28WATSON system is capable of providing a single answer to a user query expressed in a natural
language.- ATrue
- BFalse
Information
- Description
- Engineering Quiz on CMSC 435 Midterm Review, created by Aatir Siddiqui on 12/10/2023.
- Author
- Aatir Siddiqui
- Language
- English
- Views
- 14
- Tags
- Rating
- No ratings yet Rate this quiz