0:02
[Music]
0:06
thank you
0:09
[Music]
0:11
Hello friends welcome back to our
0:13
channel so in the previous session we
0:15
have discussed about the anomaly and
0:17
types of anomalies
0:19
and coming to the conclusion in order to
0:23
avoid the faults in the database that
0:25
means anomalies the only one solution is
0:29
to achieve the normalization right so
0:32
that means splitting the tables so
0:35
before going to the normalization first
0:37
let us discuss about one topic called a
0:40
functional dependency
0:45
functional dependency
0:51
so if you are aware about this
0:53
functional dependency it would be very
0:55
easy to understand this normalization
0:56
process that means splitting of tables
0:59
so first let us discuss about this
1:00
functional dependency in this session I
1:03
will I will give you what exactly is
1:05
functional functional dependency mean
1:07
and in our next sessions we'll see
1:09
different types of functional
1:10
dependencies and some exim properties
1:12
right then we'll start the normalization
1:15
process so coming to the first one
1:17
functional dependency
1:18
so this is nothing but a relation among
1:21
one attribute with another attribute in
1:24
a table
1:25
so it's a relationship
1:30
it's a relationship
1:32
between
1:34
one attribute
1:38
with another
1:43
in a table
1:46
in a table so the relationship between
1:48
different attributes we know that atable
1:51
contents of different attributes so the
1:52
relationship and it is usually denoted
1:58
denoted as
2:00
Arrow mark
2:02
okay so if X and Y are the attributes
2:09
attributes
2:11
then the functional dependency
2:14
is X tends to y
2:19
so if you X and Y are the attributes of
2:21
a relation r that means a table okay
2:23
then the functional dependency is
2:26
represented as X tends to Y X Arrow Mark
2:29
y so here the error mark represents this
2:31
functional dependency so here that
2:34
implies that implies so X
2:37
determines
2:40
determines y or
2:43
we can also say Y is
2:46
functionally
2:50
dependent
2:54
on X so this is the conclusion that
2:58
means if you represent X tends to Y that
3:02
implies X determines y or Y is
3:05
functionally dependent on X right so
3:08
here it's a table
3:14
let us take this table
3:16
so one attribute is X another attribute
3:19
is y okay so here this this implies if
3:23
we know the value of VX we can identify
3:26
the values of Y uniquely
3:29
okay if you know the value of x if you
3:33
know the value of x we can determine the
3:36
value of y right so for example
3:40
okay
3:48
let us take some values okay some values
3:51
so after that I will I will give you an
3:52
example right so let us take some values
3:54
if you know the value of x we can get
3:57
the value of y so for example if x is
4:00
equal to 20 that implies from the table
4:03
Y is equal to
4:05
6.
4:07
okay
4:08
so such a type of dependency we call it
4:11
as a x tends to Y and here x
4:15
is known as determinant
4:19
determinant and Y is called as dependent
4:25
dependent because y value is dependent
4:27
and here one question is whether X is a
4:30
single attribute and Y is a single
4:32
attribute no X can be
4:35
a set of attributes and Y can be a set
4:38
of attributes so that means
4:40
a set of attributes can determine some
4:44
set of attributes of Y so here so X and
4:49
Y can be single attribute
4:53
can be single attribute or set of
4:57
attributes
4:59
set of attributes
5:03
set of attitudes
5:04
right and one thing when we call it as
5:07
an functional dependency so
5:10
here
5:13
c x tends to Y is functionally dependent
5:19
if it satisfies
5:26
a constraint
5:28
so it should satisfy a constraint so
5:30
what is the constraints so consider any
5:33
random tuples two tuples so let us take
5:36
a random two tuples
5:38
so if
5:40
T1 dot X is equal to T2 dot X so if you
5:44
consider two tuples
5:46
so if both the Tuple values X values are
5:49
equal so for example if you consider
5:52
here
5:54
consider these two tuples so this is a
5:57
triple one and two pull two randomly you
6:00
can select randomly you can select two
6:01
tuples so if both the X values of both
6:04
the tuples are equal
6:06
then
6:08
T1 dot y must be equal to T2 of Y so the
6:13
corresponding y values also should be
6:15
equal so here you can observe T1 of v x
6:17
is 10 T 1 of Y is 5. and Team 2 of X is
6:21
10 still T2 of Y is 5 so we can say it
6:25
as a functionally dependent for example
6:27
if you are not having this value and if
6:29
you are having some another value now
6:31
what what happens t one of X is 10 T 2
6:34
of X is 10 both are equal so we have to
6:37
compare the corresponding y values so
6:39
here y values we are having a two
6:41
possibilities if x is equal to 10 then Y
6:44
is equal to either 5 or 8.
6:48
right so in this case we can't consider
6:52
this x tends to Y as a sorry X tends to
6:55
Y as in functionally dependent so in
6:59
this case in this case X and Y are not
7:01
functionally dependent because by
7:04
getting by by getting a value I mean by
7:06
using the value of v x by knowing the
7:08
value of x we can't determine the value
7:10
of y because here y value is if two
7:13
values one is having two values for the
7:15
same x value
7:16
right so this is a very important
7:19
constraint so
7:20
if you want to finalize the functional
7:23
dependency we have to find out this one
7:25
so X tends to Y so here X and Y are may
7:29
be the single attributes or the multiple
7:30
attributes so X tends to Y is
7:33
functionally dependent if it should
7:36
satisfy the constraints that randomly
7:38
select the two to pull T1 and T2 so the
7:40
T1 dot X and T2 dot ax both are equal
7:43
then corresponding y values should also
7:45
be equal for example here I am not
7:47
having this one so I am having this one
7:49
15. so now the X and Y are functionally
7:53
dependent because
7:54
if you are knowing value of x we can get
7:56
a y so if you know X is 10 5 26 37 50 15
8:01
18. so here T1 dot X is equal to T2 dot
8:05
X this doesn't happens here
8:08
so here both X values are not same okay
8:11
so we are not having the more than one I
8:15
mean the same X values duplications so
8:18
that's why we can say this as a
8:21
functionally dependent X tends to Y is a
8:23
functionally dependent here
8:25
right so hope you understood X tends to
8:28
Y is a functionally dependent when
8:31
X determines y uniquely X determines y
8:35
unitly u that means if it satisfies the
8:37
constraint T if both the X values are
8:40
equal for a different tuples the it
8:43
should contain or it should determines
8:45
the same y value also the corresponding
8:48
y values also should be same if both are
8:51
different automatically
8:53
it will be
8:55
not a functionally dependent
8:57
right so
9:00
then
9:02
here we can say a functionally dependent
9:05
fully functionally dependent or a
9:07
partial functionally dependent
9:08
fully functionally dependent or partial
9:11
functional dependent see
9:15
full functional dependent
9:24
and
9:26
partial
9:28
functional
9:31
dependent so when we call the functional
9:34
dependency as a fully functionally
9:36
dependency and a partial functional
9:38
dependency
9:39
so as we discussed now X and Y can be
9:43
set of attributes
9:46
set of attributes so
9:50
X tends to y
9:53
is said to be fully functionally
9:55
dependent
9:57
is a full
9:59
functionally dependent if
10:02
if
10:04
we remove
10:08
any attribute of x
10:18
any attribute of x
10:24
violates the functional dependency rule
10:30
violates the functional independency
10:32
rule that means we said that X and Y are
10:34
the set of attributes so for example a
10:37
comma B
10:40
function determines C
10:44
right a comma B determines C that means
10:48
the combination of A and B will
10:51
determine the C so C is a functionally
10:53
dependent on both A and P right so here
10:56
if you remove B
10:58
so then if a and C is a functionally
11:02
dependent
11:03
okay is functionally dependent
11:06
sorry sorry
11:08
if it is not a functionally dependent
11:10
okay if it is not functionally dependent
11:12
or B tends to C that means if you remove
11:15
a it's not a functionally dependent then
11:18
we call it as a
11:20
full functional dependency so even
11:22
though if you are removing one attribute
11:24
from X and still it doesn't obey the
11:27
functional dependency rule that implies
11:30
this x tends to Y will be the full full
11:33
functional dependency so that means we
11:35
can determine C only by the combination
11:38
of A and B so if you remove any one of
11:40
the attribute either b or a it doesn't
11:42
obey the functional dependency rule so
11:46
this is called a fully functional
11:47
dependency and coming to the partial
11:49
functional dependency the quite opposite
11:51
so
11:53
so he extends to Y is partial
11:59
functional dependent if
12:02
we remove
12:07
any attribute
12:11
of x
12:15
doesn't violence
12:20
FD rule
12:22
then we can say it as a partial
12:23
independent for example so let us
12:26
consider a comma B tends to C so in
12:30
order to get the C we we have to know
12:32
the values of A and B right and if a
12:35
tends to C
12:37
it's fine still functionally dependent
12:40
that means if you remove B the remaining
12:42
functional dependency a a tends to C is
12:45
also a functionally dependent that
12:47
implies this is a partial functional
12:50
dependent okay so if a tends to see F
12:54
and D all
12:56
B tends to C also a functionality
12:58
dependency
13:01
then we can say it as a partial
13:03
functional dependencies that means
13:05
removing a part of that means an
13:08
attribute of 3x still the function there
13:11
it obeys the functionality dependency
13:13
then we call it as a partial functional
13:16
dependent
13:17
right so hope you understood the
13:19
functional dependency the full full
13:21
functional dependency and the partial
13:23
functional dependencies
13:24
right so let's move on with one small
13:27
example and I will show you the
13:29
functional dependency what is the full
13:31
functional dependency and what is the
13:32
partial functional dependency
13:36
so let us take this example so student
13:39
details now we'll see what is a
13:41
functional dependency so what is the
13:43
functional dependency and what are the
13:46
partial functional dependencies and full
13:48
functional dependency now
13:50
as per our discussions we have to check
13:54
the functional dependency with the
13:55
following constraint right so let us
13:57
check here the functionality
13:59
dependencies
14:01
so first one S ID tends to
14:05
s name
14:08
so let us check whether it is a
14:09
functionality dependence or not so how
14:10
we have to check we have to consider two
14:12
tuples okay such that the two tuples x
14:16
value should be equal to the
14:17
corresponding y values for example you
14:19
can consider these two tuples so T 1 of
14:22
v x
14:24
is equal to one zero one and T2 of X is
14:27
equal to one zero one so corresponding
14:29
T2 of sorry T1 of Y
14:32
is equal to Sandeep and T2 dot Y is
14:37
equal to Sunday so it satisfies this one
14:40
okay so here it satisfies the one
14:44
right so that means
14:47
it's not it satisfies the constant once
14:50
again go with the any other or
14:53
any other constraint see any other
14:54
tuples so there are no more repetitions
14:57
in Sid so we can consider yes ID tends
15:01
to s name as a functionally dependent so
15:03
s name it depends on Sid so if you give
15:05
hid we'll get the yes name uniquely
15:08
there will be no duplications here
15:11
then
15:13
so along with this ID we can also uh
15:16
give the address we get a unique address
15:19
okay what about Sid be the course
15:24
so this will get because 1 0 1 0 1 so
15:28
there are no repetitions here leave it
15:30
Sid tends to course so course one zero
15:33
one python 101 Java so here T1 dot X is
15:38
equal to 1.1 tm2 dot x equal to 1.1 T1
15:41
dot Y is
15:43
python
15:45
and T2 dot Y is Java so it differs
15:51
so we are not having the same value of
15:53
here by
15:54
with the X so this we cannot consider it
15:57
as a functionally dependent course
15:59
doesn't functionally depends on the sib
16:02
similarly you can go with any other so
16:06
course and yes name consider course and
16:10
S name python
16:12
python so different a y values okay
16:15
different y values so here it is X and
16:17
here it is y so different y values so
16:20
this also we can't consider it okay
16:23
similarly
16:25
yes ID tends to
16:29
any other yes name
16:33
tends to S ID so if you know the name
16:36
shall we get the Sid so Sandeep Sandeep
16:40
so if you take these two we are getting
16:42
this corresponding y value same y values
16:44
and here we are having a two Sardi
16:46
sarathi but there are different X values
16:49
sorry y values so we should not consider
16:52
this one so these two are the
16:54
functionally dependents so as we said
16:56
that X and Y are the not as single
16:58
attributes it can be a set of attributes
17:01
so now we can check with that uh
17:05
yes ID comma yes name
17:09
gives the
17:11
course
17:13
Sid comma yes name gives the course let
17:18
us check here so we have to check both
17:21
we need to check both Sid and
17:24
name one zero one Sandeep python 101
17:28
Sandeep Java so
17:32
both
17:33
no
17:34
it's not a functional dependence next
17:37
yes ID comma course
17:40
completely can get the yes name
17:44
combined link it get the S9 so 101
17:47
python
17:49
gives the name Sandeep and 101 Java both
17:53
are different so we should not divide
17:54
this one right we have to find out with
17:57
a combination so here the attributes if
17:59
you know both the Sid and course we can
18:02
get the s name so here 101 Java
18:06
so that is a unit one or two python it's
18:09
a unique one zero two and C you say
18:11
unique one zero four Java it's a unique
18:13
one zero five and python it's a unique
18:15
so this is a functionally dependent
18:19
this is a functionally dependent so like
18:21
this we have to find out all the
18:23
possibilities and we have to check for
18:24
the functional dependencies with uh
18:27
whether it is following the constraint
18:28
or not now coming to the fully partially
18:32
full functional dependent and partial
18:34
functional dependent
18:38
coming to the Sid course
18:40
consider this functional dependence
18:42
because it is a set of this one so
18:45
course tends to yes name and you can
18:48
observe if you remove this course
18:50
whether it is still in a functional
18:53
dependent if it is still a functionally
18:55
dependent yes so if you remove the
18:57
course still it is in a functionally
18:59
dependent so that's why this we call it
19:02
as a partial partial functional
19:06
dependency
19:07
partial functional dependency
19:10
right so if you remove any one or one
19:14
attribute from the axis till the
19:16
functional or I mean still it obeys the
19:19
functional dependency we call it as a
19:21
partial functional dependency so we are
19:23
removing course here still it is in a
19:26
functional dependent Sid tends to s name
19:28
still it is a functional independent now
19:30
coming to the full functional dependence
19:33
so
19:36
let us check on this one s name dot a
19:39
course
19:41
give the Sid let us check this one yes
19:44
name and codes Sandeep and python gives
19:46
101 Sunday paint Java Juice 101 and
19:50
Sardi and python gives 102 so are they
19:52
and c groups one zero three so there is
19:55
no repetition so this is a functional
19:57
dependency yes obviously it will be a
19:59
functional dependent but we have to
20:01
check whether it is a partial
20:02
functionally dependent or
20:04
a full functional dependent for example
20:06
remove this course so if the dependency
20:10
yes name
20:13
tends to Sid if Sid is functionally
20:17
dependent on the S9 you can observe now
20:18
so this is the x value and this is a y
20:21
value so T 1 of e x t 2 of x
20:25
T 1 of y t 2 of Y so what is the T1 of v
20:29
x here name
20:31
so if you consider these two tuples
20:34
Okay so
20:36
we'll get a Sandeep
20:38
T1 of v x
20:42
and here also we'll get Sandeep
20:46
so here we'll get 1 0 1 and 1 0 1 yes
20:49
yes right randomly select one more table
20:52
so let us take these two things now let
20:55
us consider these two things one zero
20:57
two one zero three so the name t1.x is
21:01
sarathi
21:03
so if you consider these two tables one
21:06
zero two and one zero three t one dot X
21:08
is equal to sarathi
21:11
which is is equal to T2 dot X also right
21:15
but T1 dot y
21:18
is equal to 1 0 2 T 2 dot Y is equal to
21:23
one zero three so the names are they can
21:25
get the two different SIDS so this is
21:28
not a functionally dependent right so
21:31
this is not a functional dependent now
21:34
cancel another thing so here after
21:37
removing this course yes name
21:40
tends to Sid which is not functionally
21:42
dependent now remove this course I mean
21:45
yes name so let us check with the course
21:47
name
21:51
course with Sid so if you know the
21:54
course shall we get the Sid or not so
21:57
let us consider the things
22:00
course take this one okay so of course
22:04
T1 dot X is equal to python
22:08
and T2 dot x equal to python so we have
22:11
to consider only these things if the 2
22:13
to put the x value of two tuples are
22:15
equal then only the corresponding y
22:18
value should be equal so here I am
22:20
taking 2 python python the corresponding
22:22
Sid is one zero one and one zero two so
22:25
it may get one zero one or one zero two
22:27
which is T1 dot X sorry T1 dot Y which
22:30
is T 2 dot y so again it is not a
22:33
functionally dependent so this complete
22:36
dependency is a full functional
22:39
dependency so if we remove one attribute
22:42
from X it it violates the functionality
22:45
dependency it violates it doesn't become
22:47
a functional dependent right so that's
22:49
why we can say
22:51
this one
22:58
this one as a
23:00
full functional
23:06
full functional div
23:08
so we have to find out the functional
23:11
dependencies and then we can find out
23:13
the full functional dependency or
23:15
partial functional dependencies
23:17
right so if you know the concept of this
23:20
functional dependencies as I said
23:22
earlier it would be very easy for you to
23:24
understand the normalization process
23:26
that means dividing the tables
23:29
right so hope you understood this
23:30
functional dependency and if you are
23:31
having any doubts you can post your
23:33
doubts in the comment section definitely
23:34
I will try to clarify all your doubts
23:36
and if you really enjoyed my session
23:38
like my session share my session with
23:40
your friends and don't forget to
23:41
subscribe to our Channel thanks for
23:43
watching thank you very much