Chapter04 Review by 5R42zo5

VIEWS: 8 PAGES: 20

									                                    Database Processing, Ninth Edition



                                          CHAPTER 4
                  The Relational Model and Normalization

True-False Questions
1.    A relation is a three-dimensional table.
      Answer: False                                                Level: moderate
      Page: 120

2.    A characteristic of a relation is that the cells of the relation hold a single value.
      Answer: True                                                   Level: easy
      Page: 120

3.    A characteristic of a relation is that the rows of a relation may hold identical values.
      Answer: False                                                Level: moderate
      Page: 120

4.    The columns of a relation are sometimes called “tuples.”
      Answer: False                                            Level: easy
      Page: 122

5.    Keys are always unique.
      Answer: False                                                Level: easy
      Page: 122

6.    A tuple is a group of one or more columns that uniquely identifies a row.
      Answer: False                                            Level: easy
      Page: 122

7.    A row can be uniquely identified by a key.
      Answer: True                                                 Level: easy
      Page: 122

8.    A key can be composed of a group of attributes taken together.
      Answer: True                                            Level: easy
      Page: 122

9.    It is possible to have a relation that does not have a key.
      Answer: False                                               Level: hard
      Page: 123

10.   Attribute Y is functionally dependent on attribute X if the value of attribute X determines
      the value of Y.
      Answer: True                                               Level: moderate
      Page: 123




                                                   4-1
                       Chapter 4 - The Relational Model and Normalization


11.   The functional dependency noted as A  B, means that the value of A can be determined
      from the value of B.
      Answer: False                                         Level: easy
      Page: 123

12.   In the functional dependency shown as A  B, B is the determinant.
      Answer: False                                          Level: moderate
      Page: 123

13.   Functional dependencies can involve groups of attributes.
      Answer: True                                            Level: easy
      Page: 124

14.   A determinant of a functional dependency may or may not be unique in a relation.
      Answer: True                                          Level: moderate
      Page: 125

15.   A primary key functionally determines an entire row.
      Answer: True                                                Level: hard
      Page: 123

16.   If a table meets the minimum definition of a relation, it has an effective or appropriate structure.
      Answer: False                                              Level: moderate
      Page: 125

17.   Undesirable consequences of changing the data in a relation are called “modification anomalies.”
      Answer: True                                             Level: moderate
      Page: 125

18.   A deletion anomaly exists when deleting data about one entity results in the loss of data about
      another entity.
      Answer: True                                            Level: easy
      Page: 125

19.   A constraint, based on a business rule, that requires an instance of an entity to exist in one relation
      before it can be referenced in another relation is called an insertion anomaly.
      Answer: False                                               Level: hard
      Page: 126-127

20.   A referential integrity constraint is based on a business rule.
      Answer: True                                                Level: moderate
      Page: 127

21.   The essence of normalization is taking a relation that has more than one theme
      and breaking it into multiple relations that each have only one theme.
      Answer: True                                               Level: easy
      Page: 127




                                                  4-2
                       Chapter 4 - The Relational Model and Normalization


22.   Breaking a relation into two or more relations may create the need for a referential
      integrity constraint to be defined.
      Answer: True                                             Level: moderate
      Page: 127

23.   Relations are classified into “normal forms” based on the types of modification anomalies
      that they are vulnerable to.
      Answer: True                                             Level: hard
      Page: 127

24.   The only normal form that guarantees that a relation is free from every possible type of
      anomaly is 5NF.
      Answer: False                                             Level: hard
      Page: 127-128

25.   Any table that meets the definition of a relation is in 2NF.
      Answer: False                                              Level: easy
      Page: 128

26.   A relation is in first normal form if all of its non-key attributes are dependent on part of the key.
      Answer: False                                                 Level: moderate
      Page: 128

27.   To be in second normal form, a relation must contain no partial dependencies.
      Answer: True                                            Level: moderate
      Page: 128

28.   A relation is in second normal form if it is in first normal form and all of its non-key
      attributes are dependent on the entire key.
      Answer: True                                                Level: moderate
      Page: 128

29.   Relations that have a composite key and are in 1NF are automatically in 2NF.
      Answer: False                                           Level: easy
      Page: 128

30.   A transitive dependency exists when a non-key attribute is determined by only part of the key.
      Answer: False                                            Level: moderate
      Page: 129

31.   A relation is in 3NF if it is in 2NF and contains no transitive dependencies.
      Answer: True                                                Level: easy
      Page: 129

32.   A relation can have only one candidate key.
      Answer: False                                               Level: moderate
      Page: 130




                                                  4-3
                       Chapter 4 - The Relational Model and Normalization


33.   A relation is in Boyce-Codd Normal Form if it is in 3NF and every determinant
      is a candidate key.
      Answer: True                                            Level: easy
      Page: 131

34.   A multi-value dependency exists when a relation has at least three attributes, two of which
      are multi-value and their values depend only on the third attribute.
      Answer: True                                              Level: hard
      Page: 132

35.   A relation is in 4NF if it is in 3NF and contains no multi-value dependencies.
      Answer: True                                               Level: moderate
      Page: 132

36.   Fifth normal form deals with obscure problems with transitive dependencies.
      Answer: False                                           Level: easy
      Page: 133

37.   A relation is in DK/NF if every domain is the logical consequence of the definition of constraints
      and keys.
      Answer: False                                            Level: hard
      Page: 133

38.   In considering DK/NF, a constraint includes any rule governing static values in the relation
      except constraints that are time-based.
      Answer: True                                            Level: moderate
      Page: 133

39.   There is no known algorithm for converting a relation into DK/NF.
      Answer: True                                             Level: moderate
      Page: 134

40.   If A  B and B  A, then A and B have a many-to-many attribute relationship.
      Answer: False                                      Level: moderate
      Page: 138

41.   Attributes that have a one-to-one attribute relationship must occur together in at least one relation.
      Answer: True                                              Level: moderate
      Page: 138

42.   It is generally undesirable to have attributes that have a one-to-one attribute relationship appear in
      more than one relation together.
      Answer: True                                                Level: hard
      Page: 139

43.   In general, a row of a relation should have all of the data about one instance of the relation’s
      theme.
      Answer: True                                                Level: moderate
      Page: 141



                                                  4-4
                        Chapter 4 - The Relational Model and Normalization




44.    De-normalization is the process of constructing relations by synthesis of attribute relationships.
       Answer: False                                             Level: moderate
       Page: 141

45.    Relations may be unnormalized or de-normalized to improve database performance.
       Answer: True                                          Level: easy
       Page: 143-144


Multiple Choice Questions
46.    A relation is also known as                          .
       a.)     a table
       b.)     a tuple
       c.)     a relationship
       d.)     an attribute
       e.)     a field

Level: easy
Page: 122

47.    A tuple is also known as a                       .
       a.)     table
       b.)     relation
       c.)     row
       d.)     field
       e.)     file

Level: moderate
Page: 122

48.    An attribute is also known as a                          .
       a.)     table
       b.)     relation
       c.)     row
       d.)     field
       e.)     file

Level: moderate
Page: 122




                                                  4-5
                       Chapter 4 - The Relational Model and Normalization


49.    Which of the following, if any, is not true about keys?
       a.)    can be a single attribute
       b.)    can be a group of attributes taken together
       c.)    can uniquely identify a row
       d.)    must be functionally determined by another attribute
       e.)    All of the above are true.

Level: moderate
Page: 123-125

50.    A functional dependency in a relation                       .
       a.)     usually involves a formula
       b.)     always involves a primary key
       c.)     exists among tuples
       d.)     must always be removed through normalization
       e.)     none of the above

Level: moderate
Page: 123

51.    Which of the following is known to be true from the functional dependency shown as (A, B) 
       (C, D)?
       a.)     A is the determinant of C
       b.)     A and B together are determined by C and D together
       c.)     A and B together determine D
       d.)     C and D together determine A
       e.)     A determines B

Level: moderate
Page: 125

52.    Which of the following is known to be true from the functional dependency
       shown as A  (X, Y)?
       a.)    X is functionally dependent on A
       b.)    A determines Y
       c.)    A is a determinant
       d.)    X and Y are functionally dependent on A
       e.)    All of the above.

Level: moderate
Page: 123-125




                                                 4-6
                        Chapter 4 - The Relational Model and Normalization


53.    If the removal of facts about one entity results in the unintentional lose of data about another
       entity, this is referred to an a(n) ____________.
       a.)      normalization anomaly
       b.)      insertion anomaly
       c.)      entity anomaly
       d.)      deletion anomaly
       e.)      removal anomaly

Level: moderate
Page: 125

54.    The notation CUST-CAR[License]  CAR-RENT[License] indicates                             .
       a.)    License in CAR-RENT is a subset of License in CUST-CAR.
       b.)    Before a value can be entered in the License field of the CAR-RENT relation, it must
              have a matching value in the License field of the CUST-CAR relation.
       c.)    License in CUST-CAR is functionally dependent on License in CAR-RENT.
       d.)    Before a value can be entered in the License field of the CUST-CAR relation, it must
              have a matching value in the License field of the CAR-RENT relation.
       e.)    CUST-CAR and CAR-RENT are functionally dependent on the License field.

Level: hard
Page: 126

55.    Which of the following is not true about normalization?
       a.)    produces relations with a single theme
       b.)    may create referential integrity constraints
       c.)    reduces the number of functional dependencies in the schema
       d.)    reduces anomalies
       e.)    splits a poorly structured relation into two or more well-structured relations

Level: moderate
Page: 127

56.    A table that meets the definition of a relation is in                     .
       a.)     First Normal Form
       b.)     Second Normal Form
       c.)     Third Normal Form
       d.)     Boyce-Codd Normal Form
       e.)     Fourth Normal Form

Level: easy
Page: 128




                                                    4-7
                        Chapter 4 - The Relational Model and Normalization


57.    Which of the following is not a requirement for 1NF?
       a.)    cells must contain single values
       b.)    all entries in a column must be of the same kind
       c.)    no two rows may be identical
       d.)    rows must be ordered by the value of the primary key
       e.)    the order of the columns is insignificant

Level: moderate
Page: 120

58.    A relation is in 2NF if it is in 1NF and all its non-key attributes are   .
       a.)     dependent on part of the primary key
       b.)     dependent on all of the primary key
       c.)     independent of the primary key
       d.)     independent of each other
       e.)     independent of any other relation

Level: moderate
Page: 128

59.    Which of the following is a requirement of 2NF?
       a.)    must contain a partial dependency
       b.)    must contain a composite primary key
       c.)    must contain no partial dependencies
       d.)    must contain no transitive dependencies
       e.)    must contain a multi-valued dependency

Level: moderate
Page: 128

60.    A relation is in third normal form if it is in second normal form and         .
       a.)     is dependent on part of the key
       b.)     is dependent on all of the key
       c.)     is independent of the key
       d.)      has no transitive dependencies
       e.)     is independent of any other relation

Level: moderate
Page: 129

61.    Which of the following is a requirement of 3NF?
       a.)    must contain a partial dependency
       b.)    must contain a composite key
       c.)    must contain no partial dependencies
       d.)    must contain no transitive dependencies
       e.)    must contain a multi-valued dependency

Level: moderate
Page: 129



                                                   4-8
                       Chapter 4 - The Relational Model and Normalization



                         EMPLOYEE (SSN, SupervisorID, Department)
                         Key: SSN
                         Functional Dependencies: SSN  SupervisorID
                                                  SupervisorID  Department
                             SSN            SupervisorID         Department
                         123-45-6789            100              Engineering
                         234-56-7890            200               Marketing
                         345-67-8901            100              Engineering
                         456-78-9012            300               Production
                         567-89-0123            100              Engineering
                         678-90-1234            200               Marketing

62.    Which type of dependency is present in the EMPLOYEE relation above?
       a.)    partial dependency
       b.)    transitive dependency
       c.)    determinant dependency
       d.)    multi-valued dependency
       e.)    deletion dependency

Level: hard
Page: 129

63.    A relation is in Boyce-Codd normal form if it is in 3NF and              .
       a.)     every determinant is a candidate key
       b.)     every determinant is a primary key
       c.)     every attribute is a candidate key
       d.)     there is more than one candidate key
       e.)     there is more than one primary key

Level: moderate
Page: 131

64.    A “candidate key” is                       .
       a.)    a primary key.
       b.)    any group of attributes that are a determinant
       c.)    functionally dependent on the non-key attributes
       d.)    an attribute or group of attributes that can be the primary key
       e.)    the primary key selected to be the key of a relation

Level: easy
Page: 123




                                                 4-9
                        Chapter 4 - The Relational Model and Normalization


65.    A multi-value dependency exists when a relation has                            .
       a.)     at least two attributes, both of them are multi-value, and their values depend on each
               other
       b.)     at least two attributes, one of them is multi-value, and its value depends on the other
       c.)     at least three attributes, two of them are multi-value, and their values depend on
               only the third attribute
       d.)     at least three attributes, one of them is multi-value, and its value depends on the other two
               attributes
       e.)     at least three attributes, all of them are multi-value, and their values depend on each other

Level: hard
Page: 132

66.    A relation is in fourth normal form if it is in BCNF and it has no                      .
       a.)     transitive dependencies
       b.)     multi-value dependencies
       c.)     partial dependencies
       d.)     deletion dependencies
       e.)     referential integrity conflicts

Level: easy
Page: 132

67.    The anomalies addressed by moving from BCNF to 4NF generally deal with                            .
       a.)    excessive updates and redundancy of data for each entity
       b.)    inability to uniquely identify an entity
       c.)    inability to reconstruct relations once they have been decomposed
       d.)    creation of identical rows in a relation
       e.)    functional dependencies

Level: hard
Page: 132

68.    Fifth normal form deals with                      .
       a.)     excessive updates and redundancy of data for each entity
       b.)     inability to uniquely identify an entity
       c.)     inability to reconstruct relations once they have been decomposed
       d.)     creation of identical rows in a relation
       e.)     dependencies between referential integrity constraints

Level: moderate
Page: 133




                                                 4 - 10
                       Chapter 4 - The Relational Model and Normalization


69.    The advantage of having a relation in domain/key normal form is that                        .
       a.)    it takes less storage space than other normal forms
       b.)    it is easily obtained from Boyce-Codd normal form
       c.)    there is a relatively simple algorithm for obtaining DK/NF
       d.)    it is obtained by enforcing referential integrity constraints
       e.)    it is guaranteed to have no modification anomalies

Level: easy
Page: 133

70.    A relation is in domain/key normal form if                        .
       a.)     every key of the relation is a logical consequence of the definition of constraints and
               determinants
       b.)     every key of the relation is a logical consequence of the definition of constraints and
               domains
       c.)     every constraint on the relation is a logical consequence of the definition of keys and
               determinants
       d.)     every constraint on the relation is a logical consequence of the definition of keys and
               domains
       e.)     every domain of the relation is a logical consequence of the definition of keys and
               constraints

Level: moderate
Page: 133

71.    In the definition of DK/NF, which of the following is a type of constraint that is excluded from
       consideration?
       a.)     edit rules
       b.)     intra-relation constraints
       c.)     functional dependencies
       d.)     time-based constraints
       e.)     inter-relation constraints

Level: moderate
Page: 133

72.    Synthesizing relations means that the developers start with a                      .
       a.)    relation containing functional dependencies
       b.)    relation without any functional dependencies
       c.)    set of attributes with certain functional dependencies
       d.)    set of domains and their constraints
       e.)    set of anomalies based on dependencies

Level: moderate
Page: 138




                                                4 - 11
                        Chapter 4 - The Relational Model and Normalization


73.    If two attributes A and B have a one-to-one attribute relationship, it is shown as              .

       a.)     A  B and B  A
       b.)     A  B, but B not  A
       c.)     B  A, but A not  B
       d.)     A not  B and B not  A
       e.)     AB

Level: moderate
Page: 138

74.    If two attributes A and B have a many-to-one attribute relationship, it is shown as             .
       a.)      A  B and B  A
       b.)      A  B, but B not  A
       c.)      B  A, and A  B
       d.)      A not  B and B not  A
       e.)      AB

Level: moderate
Page: 138

75.    If two attributes A and B have a many-to-many attribute relationship, it is shown as            .
       a.)      A  B and B  A
       b.)      A  B, but B not  A
       c.)      B  A, and A  B
       d.)      A not  B and B not  A
       e.)      AB

Level: moderate
Page: 138

76.    If two attributes A and B have a one-to-one attribute relationship in the relation R1(A,B,C),
       then the primary key is                      .
       a.)      A
       b.)      B
       c.)      A or B
       d.)      A and B
       e.)      C

Level: easy
Page: 139




                                                 4 - 12
                       Chapter 4 - The Relational Model and Normalization


77.    If two attributes A and B have a many-to-one attribute relationship in the relation R2(A,B,C),
       then the primary key is                     .
       a.)      A
       b.)      B
       c.)      A or B
       d.)      A and B
       e.)      C

Level: easy
Page: 139

78.    When creating a database with attributes that have a one-to-one attribute relationship ________ .
       a.)    the two attributes must appear in a relation together
       b.)    the two attributes usually only appear in one relation together
       c.)    other attributes that are functionally dependent on either of them may appear in a relation
              with them
       d.)    either of the attributes can appear in relations without the other
       e.)    all of the above

Level: moderate
Page: 139-140

79.    If two attributes A and B have a many-to-many attribute relationship in the relation R3(A,B,C),
       then the primary key is                    .
       a.)      A
       b.)      B
       c.)      A or B
       d.)      A and B
       e.)      C

Level: easy
Page: 139

80.    If two attributes G and H have a many-to-one relationship exist in a relation R4(G, H, J) where
       J is a third attribute appearing in the relation, then                   .
       a.)       G  H and G  J
       b.)       G  H, but H  J
       c.)       (G, H)  J
       d.)       J  (G, H)
       e.)       J  G, and J  H

Level: moderate
Page: 139




                                                4 - 13
                        Chapter 4 - The Relational Model and Normalization


81.    If two attributes A and B have a many-to-many attribute relationship, it is shown as                 .
       a.)      A  B and B  A
       b.)      A  B, but B not  A
       c.)      B  A, and A  B
       d.)      A not  B and B not  A
       e.)      AB

Level: moderate
Page: 139

82.    When a relation has a key consisting of multiple attributes, you can add a new attribute to the
       relation ______________.
       a.)     without any restriction
       b.)     so long as it is functionally dependent on part of the key
       c.)     so long as it is functionally dependent on all of the key
       d.)     so long as it is functionally dependent on a non-key attribute
       e.)     so long as it is a candidate key

Level: moderate
Page: 140

83.    If two attributes A and B that have a many-to-many attribute relationship, and a third attribute C
       also appears in the relation expressing the many-to-many relationship, then ______________.
       a.)      A  B and B  C
       b.)      B  A, but A  C
       c.)      (A, B)  C
       d.)      C  (A, B)
       e.)      C A, and C  B

Level: moderate
Page: 139

84.    In general, a row in a relation should have all of the data about ___________.
       a.)     one instance of the relation's theme
       b.)     each instance of the relation's theme
       c.)     every instance of the relation's theme
       d.)     no instance of the relation's theme
       e.)     the relation's functional dependencies

Level: easy
Page: 141




                                                 4 - 14
                         Chapter 4 - The Relational Model and Normalization


85.     Database performance can be improved through appropriate use of ____________.
        a.)    uncontrolled redundancy
        b.)    de-normalization
        c.)    data integrity violations
        d.)    synchronous dependencies
        e.)    both a and b

Level: easy
Page: 143-144


Fill in the Blank Questions
86.    A(n) relation is a two-dimensional table.
Level: easy
Page: 120

87.    In relational terms, a row is called a(n)   tuple .
Level: easy
Page: 122

88.    In relational terms, a column is called a(n)    attribute .
Level: easy
Page: 122

89.    A(n) composite key       is a group of attributes that uniquely identifies a row.
Level: easy
Page: 122

90.    A(n) candidate key is one of a group of keys that may serve as the primary key.
Level: easy
Page: 123

91.    A(n)    primary key is a group of one or more attributes that uniquely identifies a row.
Level: easy
Page: 123

92.    A key functionally determines the entire       row .
Level: easy
Page: 123

93.    A(n) functional dependency is a relationship between attributes such that if we know the value
       of one attribute, we can determine the value of the other attribute.
Level: easy
Page: 123




                                                   4 - 15
                           Chapter 4 - The Relational Model and Normalization


94.    If by knowing the value of A, we can find the value of B, then we would say that
       B is functionally dependent on A.
Level: moderate
Page: 123

95.    In functional dependencies, the attribute whose value is known or given is
       referred to as the determinant .
Level: moderate
Page: 123

96.    Unlike determinants, primary keys are always unique.
Level: easy
Page: 125

97.    For some relations, changing the data can have undesirable consequences
       called modification anomalies .
Level: moderate
Page: 125

98.    Every time we break up a relation during the normalization process, we may
       create referential integrity constraints.
Level: moderate
Page: 127

99.    Normal forms are nested so that if a relation is in 2NF, it is also in 1NF.
Level: easy
Page: 127

100.   A relation that is in domain / key normal form is assured to be free from all anomalies.
Level: moderate
Page: 128

101.   Any table that meets the definition of a(n)      relation is said to be in first normal form.
Level: easy
Page: 128

102.   If a table is a relation then it is in   1NF .
Level: easy
Page: 128

103.   To be in 1NF, the cells of a table must contain a(n)       single   value.
Level: hard
Page: 120

104.   A relation is in 2NF if it is in 1NF and all of its non-key attributes are
       dependent on all of the key .
Level: moderate
Page: 128




                                                    4 - 16
                         Chapter 4 - The Relational Model and Normalization


105.   A relation is in second normal form if all its non-key attributes are dependent on the entire
       key.
Level: easy
Page: 128

106.   A relation is in 2NF if it is in 1NF and has no partial dependencies .
Level: easy
Page: 128

107.   A relation is in third normal form if it is in second normal form
       and has no transitive dependencies .
Level: moderate
Page: 129

108.   A defining requirement for     Boyce-Codd normal form is that every determinant must be a
       candidate key.
Level: moderate
Page: 131

109.   A relation is in BCNF if it is in 3NF and every     determinant is a candidate key.
Level: moderate
Page: 131

110.   A relation is in BCNF if it is in 3NF and every determinant is a(n)    candidate key .
Level: moderate
Page: 131

111.   Relations in BCNF have no anomalies in regard to functional dependencies .
Level: hard
Page: 131

112.   A(n) multi-value dependency exists when a relation has at least three attributes, two of them
       are multi-value, and their values depend on only the third attribute.
Level: easy
Page: 132

113.   A relation is in 4NF if it is in BCNF and it has no    multi-value dependencies .
Level: moderate
Page: 132

114.     Fifth normal form deals with obscure dependencies that may not have any practical
       consequences.
Level: moderate
Page: 133

115.   Domain / key normal form requires that every        constraint be a logical consequence of the
       definition of domains and keys.
Level: moderate
Page: 133



                                                  4 - 17
                         Chapter 4 - The Relational Model and Normalization




116.   In regard to DK/NF, a constraint is broadly defined to include all types of constraints
       except time-based constraints.
Level: hard
Page: 133

117.   If the functional dependencies between two attributes is A  B and B  A, then they
       have a(n) one-to-one attribute relationship.
Level: easy
Page: 138

118.   If the functional dependencies between two attributes is A  B, but B not  A, then they
       have a(n) many-to-one attribute relationship.
Level: moderate
Page: 138

119.   If the functional dependencies between two attributes is A not  B and B not  A, then they
       have a(n) many-to-many attribute relationship.
Level: moderate
Page: 138

120.   When synthesizing relations, if two attributes have the relationship A  B and B  A, then
       create a relation with A or B as the key.
Level: easy
Page: 139

121.   When synthesizing relations, if two attributes have the relationship A  B, but not B  A, then
       create a relation with A as the key.
Level: easy
Page: 139

122.   When synthesizing relations, if attributes have the dependencies (A, B)  C, then create
       a relation with (A, B) as the key.
Level: moderate
Page: 139

123.   Relations are sometimes left unnormalized to improve       performance .
Level: easy
Page: 141



Essay Questions
124.    Describe the requirements that a table must meet to qualify as a relation.

        In order for a table to meet the requirements of a relation, it must satisfy certain conditions.
        First, the rows contain data about an entity or some portion of the entity. Second, the columns
        contain data that represent attributes of the entity – each column must contain the same kind of



                                                  4 - 18
                        Chapter 4 - The Relational Model and Normalization


       data for all rows – and each column must have a unique name. Third, the cells must contain
       single values – no group or array values are allowed. Finally, the rows must be unique. It is
       worth noting that the order of the columns and the rows is insignificant.

125.   Explain the relationships among primary keys, functional dependencies, and uniqueness.

       All primary keys represent a functional dependency, since each primary key functionally
       determines the entire row. Since the primary key determines the entire row, if a primary key were
       duplicated in a relation, then the entire row would have to be duplicated. Since a relation can
       contain no duplicate rows, the primary key must be unique. However, all functional
       dependencies are not represented through primary keys. Therefore, the determinant of a
       functional dependency may or may not be unique.

126.   Explain the essence of normalization that is implemented through the use of normal forms.

       The essence of normalization is to create relations that contain data on a single theme, and for
       each row in a relation to contain all the data about that theme for a given entity. This essence is
       implemented in the normal forms by taking a relation that contains data on more than one theme
       and splitting it into multiple relations that each contain data on a single theme.

127.   Explain the concept of a transitive dependency and give an example not included in the book.

       A transitive dependency is evidenced through a functional dependency among non-key attributes.
       If the key of a relation determines the value of one attribute because it can determine the value of
       another attribute, then a transitive dependency exists. For example, if an employee is assigned a
       computer to work on and the computer has an assigned IP address for the corporate network, the
       relation EMPLOYEE (EmployeeID, ComputerID, IPaddress) with EmployeeID as the key,
       contains a transitive dependency. The IP address is not really determined by the EmployeeID,
       but rather by the ComputerID, such that EmployeeID determines IPaddress only because
       EmployeeID determines ComputerID which in turn determines IPaddress. The functional
       dependencies can be shown asEmployeeID  ComputerID
       ComputerID  IPaddress
       Therefore EmployeeID  ComputerID  Ipaddress

128.   Explain the practical value that DK/NF has in the real world, and why that practical
       value is limited.

       Domain key normal form has practice value as a database design objective. Since a relation in
       domain key normal form is guaranteed to contain no anomalies, it is the preferred normalization
       goal. Many relations can achieve DK/NF while others cannot. The fact that many relations
       cannot be placed in DK/NF, and there is no method for determining which can and which cannot,
       limits the practical value of DK/NF as a design goal. Further, there is no algorithm for
       achieving DK/NF even for relations that can reach this ultimate level of normalization.

129.   Contrast the analytic and synthetic perspectives of relational design.

       The analytic perspective of relational design starts from a relation and attempts to identify the
       anomalies to which it is vulnerable in order to determine which normal form it is in. The
       synthetic perspective starts with a collection of attributes and their functional dependencies. It



                                                 4 - 19
                Chapter 4 - The Relational Model and Normalization


then attempts to determine which relations should be formed from these attributes based on their
attribute relationships.

130.    Explain what it means to de-normalize, and why it may be appropriate.

De-normalization is the process of taking a normalized set of relations and converting them into
a smaller set of relations that is in a lower normal form and thus vulnerable to a greater number
of anomalies. While the fully normalized set of relations is more desirable from a theoretical
perspective, it does have some disadvantages. The greater the number of relations in the
database, the more work that is required by the database whenever data must be retrieved from
multiple relations to reconstruct the users’ view of the data. In some circumstances, the
anomalies introduced by de-normalizing a data model may be considered an acceptable cost for
the performance improvement from reducing the overhead work performed by the database.




                                         4 - 20

								
To top