LSUS MHA 705 – Exam 5 Questions and Answers|
Updated| Pass Guaranteed
DATA MINING CONCEPTS AND PHASES
1. Knowledge discovery in databases (KDD) is best described as:
A. The same thing as storing data in a warehouse
B. The overall process of identifying valid, novel, useful, and
understandable patterns in data, of which data mining is one step
C. A method of destroying obsolete records
D. A coding system for diagnoses
Answer: B. The overall process of identifying valid, novel, useful, and
understandable patterns in data, of which data mining is one step
Rationale: KDD is the broader process that includes selection,
preprocessing, transformation, data mining, and interpretation or evaluation.
Data mining is the pattern-finding step within KDD.
2. Data mining is defined as:
A. The use of statistical, mathematical, and machine learning
techniques to discover patterns and relationships in large data sets
B. Retrieving a single patient's record
C. Backing up a database
D. Encrypting data for transmission
Answer: A. The use of statistical, mathematical, and machine learning
techniques to discover patterns and relationships in large data sets
Rationale: Data mining automates the search for meaningful patterns,
trends, and associations in large volumes of data to support prediction and
decision making.
MHA 705 Exam 5– Page 1
,3. Which sequence correctly lists the steps of the traditional KDD
process?
A. Data mining, selection, destruction, reporting, archiving
B. Interpretation, transformation, selection, billing, mining
C. Preprocessing, deployment, selection, mining, coding
D. Selection, preprocessing, transformation, data mining,
interpretation/evaluation
Answer: D. Selection, preprocessing, transformation, data mining,
interpretation/evaluation
Rationale: Fayyad's KDD process starts with selecting target data, cleans
and preprocesses it, transforms it into suitable form, applies data mining,
and then interprets and evaluates the patterns discovered.
4. In the four-phase process described for forecasting and data
mining projects, which phase comes FIRST?
A. Knowledge deployment
B. Pattern discovery
C. Problem identification (definition)
D. Data exploration
Answer: C. Problem identification (definition)
Rationale: The four phases are problem identification, data exploration,
pattern discovery, and knowledge deployment. Clear problem definition
drives data selection and technique choice.
5. Which activity belongs in the data exploration phase?
A. Implementing the final model in daily operations
B. Examining, summarizing, and cleaning the data to understand its
quality, distribution, and relationships
C. Deciding on the business problem only
D. Shredding the source data
Answer: B. Examining, summarizing, and cleaning the data to
understand its quality, distribution, and relationships
Rationale: Exploration includes profiling, visualizing, checking missing
values and outliers, and identifying useful variables before modeling.
MHA 705 Exam 5– Page 2
, 6. Applying algorithms such as clustering, classification, or
association analysis to find relationships in the data occurs in which
phase?
A. Pattern discovery
B. Problem identification
C. Knowledge deployment
D. Data destruction
Answer: A. Pattern discovery
Rationale: Pattern discovery (modeling) is when techniques are applied
and the resulting models or patterns are tested and refined.
7. Knowledge deployment in the four-phase process means:
A. Collecting the initial data
B. Cleaning duplicate records
C. Defining the research question
D. Putting the validated findings or models to use in decisions,
processes, or systems, and monitoring their performance
Answer: D. Putting the validated findings or models to use in
decisions, processes, or systems, and monitoring their performance
Rationale: Deployment applies discovered knowledge, such as embedding
a prediction model in workflow or creating reports. It also includes
evaluating results over time because models can degrade.
8. CRISP-DM stands for:
A. Clinical Records Information System for Data Management
B. Computerized Registry Index for Statistical Data Modeling
C. Cross-Industry Standard Process for Data Mining
D. Common Research Industry Standard for Database Maintenance
Answer: C. Cross-Industry Standard Process for Data Mining
Rationale: CRISP-DM is a widely used, industry-neutral process model for
data mining projects.
MHA 705 Exam 5– Page 3
Updated| Pass Guaranteed
DATA MINING CONCEPTS AND PHASES
1. Knowledge discovery in databases (KDD) is best described as:
A. The same thing as storing data in a warehouse
B. The overall process of identifying valid, novel, useful, and
understandable patterns in data, of which data mining is one step
C. A method of destroying obsolete records
D. A coding system for diagnoses
Answer: B. The overall process of identifying valid, novel, useful, and
understandable patterns in data, of which data mining is one step
Rationale: KDD is the broader process that includes selection,
preprocessing, transformation, data mining, and interpretation or evaluation.
Data mining is the pattern-finding step within KDD.
2. Data mining is defined as:
A. The use of statistical, mathematical, and machine learning
techniques to discover patterns and relationships in large data sets
B. Retrieving a single patient's record
C. Backing up a database
D. Encrypting data for transmission
Answer: A. The use of statistical, mathematical, and machine learning
techniques to discover patterns and relationships in large data sets
Rationale: Data mining automates the search for meaningful patterns,
trends, and associations in large volumes of data to support prediction and
decision making.
MHA 705 Exam 5– Page 1
,3. Which sequence correctly lists the steps of the traditional KDD
process?
A. Data mining, selection, destruction, reporting, archiving
B. Interpretation, transformation, selection, billing, mining
C. Preprocessing, deployment, selection, mining, coding
D. Selection, preprocessing, transformation, data mining,
interpretation/evaluation
Answer: D. Selection, preprocessing, transformation, data mining,
interpretation/evaluation
Rationale: Fayyad's KDD process starts with selecting target data, cleans
and preprocesses it, transforms it into suitable form, applies data mining,
and then interprets and evaluates the patterns discovered.
4. In the four-phase process described for forecasting and data
mining projects, which phase comes FIRST?
A. Knowledge deployment
B. Pattern discovery
C. Problem identification (definition)
D. Data exploration
Answer: C. Problem identification (definition)
Rationale: The four phases are problem identification, data exploration,
pattern discovery, and knowledge deployment. Clear problem definition
drives data selection and technique choice.
5. Which activity belongs in the data exploration phase?
A. Implementing the final model in daily operations
B. Examining, summarizing, and cleaning the data to understand its
quality, distribution, and relationships
C. Deciding on the business problem only
D. Shredding the source data
Answer: B. Examining, summarizing, and cleaning the data to
understand its quality, distribution, and relationships
Rationale: Exploration includes profiling, visualizing, checking missing
values and outliers, and identifying useful variables before modeling.
MHA 705 Exam 5– Page 2
, 6. Applying algorithms such as clustering, classification, or
association analysis to find relationships in the data occurs in which
phase?
A. Pattern discovery
B. Problem identification
C. Knowledge deployment
D. Data destruction
Answer: A. Pattern discovery
Rationale: Pattern discovery (modeling) is when techniques are applied
and the resulting models or patterns are tested and refined.
7. Knowledge deployment in the four-phase process means:
A. Collecting the initial data
B. Cleaning duplicate records
C. Defining the research question
D. Putting the validated findings or models to use in decisions,
processes, or systems, and monitoring their performance
Answer: D. Putting the validated findings or models to use in
decisions, processes, or systems, and monitoring their performance
Rationale: Deployment applies discovered knowledge, such as embedding
a prediction model in workflow or creating reports. It also includes
evaluating results over time because models can degrade.
8. CRISP-DM stands for:
A. Clinical Records Information System for Data Management
B. Computerized Registry Index for Statistical Data Modeling
C. Cross-Industry Standard Process for Data Mining
D. Common Research Industry Standard for Database Maintenance
Answer: C. Cross-Industry Standard Process for Data Mining
Rationale: CRISP-DM is a widely used, industry-neutral process model for
data mining projects.
MHA 705 Exam 5– Page 3
