Hang tight while we fetch the video data and transcripts. This only takes a moment.
Connecting to YouTube player…
Fetching transcript data…
We’ll display the transcript, summary, and all view options as soon as everything loads.
Next steps
Loading transcript tools…
Spatial Database Management System SDBMS Spatial Data Science and Applications | MB | YouTubeToText
YouTube Transcript: Spatial Database Management System SDBMS Spatial Data Science and Applications
Skip watching entire videos - get the full transcript, search for keywords, and copy with one click.
Share:
Video Transcript
from the previous lecture you studied on
powerful functionalities of dbms at the
same time it was clearly noted that
conventional relational dbms is not
designed for complex data types such as spatial
spatial
data in order to fully utilize the
proposed framework you should have a
firm understanding of spatial database
management system which is the topic of this
structure now let's start with the
question what is spatial database
management system in other words what is
the difference of spatial dbms from
conventional relational
dbms literally speaking spatial dbms is
designed for handling spatial data such
as Vector data and restor data for
handling spatial data it needs object
relational Ms in which abstract data
type in other words user defined new
data types can be
managed because spatial data is saved so
that spatial dbms needs spatial quality
language equipped with spatial
operations and spatial indexing for
quity optimization and so on so spatial
dbms can be defined as a dbms with
additional capability of handling
spatial data
now let's think about why conventional
relational dbms cannot deal with spatial
data in fact relational dbms can handle
spatial data but the problem is it is
not efficient the figure shows the
limitation the cadest map can be modeled
and the data can be saved in real
relational dbms however the special
components polygon Edge and point should
be save in three different tables which
makes simple spatial operation very
complicated because all the three tables
should be visited using time consuming
operation so we needed another way to
deal with spatial data one simple
solution was so-called dual architecture
in which relation dbms manage only at
data and a separate file system is used
for spatial
data it can overcome the inefficiency of
relational dbms however the Dual
architecture cannot provide powerful
dbms features such as transaction
management and many
others it could be used for only single user
applications when object relation tbms was
was
introduced and it can deal with abstract
data types in
1990s spatial data can be tightly
integrated with dbns in other words
column data type can be polygon line or
Point as a result full functionalities
architecture I mentioned object
relational dbms quite a few times already
already
let's discuss more about what it is in
more detail world dbms can be considered
as the middle ground to bridge the gap
between relational dbms and the objectoriented
objectoriented
programming which supports three
functionalities complex data with user
defined Class Type inheritance object
Behavior with method World dbms present
the best of the two consequently can
deal with spatial
data among many or dbms object relational
relational
dbms throughout this course I will make
use of post s square and S extension for
special data types post
GIS which are opsource software
discussed in the second
week the example is showing how to
create userdefined data type address is
the case which is composed of load C and
zip code and then using a square we can
create a table named restaurant which
has a column address the final s Square
actually inserts a restaurant named
spaga to restaurant table as you can see
or dbms can store a complex data type in
table which was not possible in
dbms the figure shows a series of
processing in SQL window of postgis to
create a data type create a table and
insert a record which was described just
slide now we have a method to store
spatial data in
dbms the next issue is how to retrieve
spatial data from dbms as you studied it
for the retrieval process square is also
used that means that you need the square
with spatial operations for spatial data
SQL 3 which is a standard SQL
established in
1999 allows to support spal data type
such as Point curve surface geometri correction
correction
and operations such as spal reference
envelope some booing operations of Pop
topology and spatial functions such as
distance buffer insertion intersection
others joint operation is to connect two
tables based on clance check on common
field join is a key operation at the
same time a very expensive operation in
dbms C
processing actually indexing is to
basically speed up joint
operation likewise spatial joint to
connect two table based on spatial
relationship is also very powerful at
the same time
expensive in other words timeconsuming
operation instead of Correspondence or
exact matching of in regular joint
operation space joint is based on
special relationship such as intersect
contain covers and so
on let's take a look at an example of
spatial joint the quy is what subway
stations are located in Little Italy in
New York
City for the quy we need two tables New
York City neighborhood table which is
based on polygon and Subway table based on
on
point the query sets two variables n and
S to point the two tables and then find
Lally from New York City neighborhood
table and find the subway station of
which location is within Little Italy
then finally Lally and the subway station
station
names please note that for spatial
relationship a bulling function s within
is used for spatial joint
here the figures on the side illustrate
the query processing and the result of
view I brought in another example of
spatial joint the previous example was
spal joint between polygon and point
this time it is between polygon and
line the query is what are the RADS
which cross the boundary of Corona
neighborhood the S square is similar to
the previous one but this time the
spatial relationship is defined by a
bullan function s
crosses from all the street of New York
City the Quarry retrieves only Street
which meet the boundary of Corona neighborhood
and now you are looking at the Quality
result in table View and map
view now let's think about the same
query with different data set the query
is what are the
laws which cross the boundary of D
County in
Wisconsin for the query we are using
county map and load network of the whole
country for the query all the Road from
Alaska to Florida should be checked if
they meet the boundary of B County how
about that would it work yes it works
time that's why spatial indexing is
required in spatial dvms for efficient
quity processing of the previous spal
joint what if we have to check the rows
only around D count in
Wisconsin the quality processing can be
done very
quickly spatial indexing is to provide a
better search performance of spatial
context for that R Tree which is a
extension of B tree was
introduced there are many other spal
indexing method but Archer is considered
as the standard additionally it's worth
mentioning geohash which is a powerful
method for spatial data searching and
organization which is going to be used
Data the figure is composed of point and
rectangles for the given spatial data
you can apply Arch tree based on MBR
which stands for minimum bounding
rectangles you are looking at the given
spal data set which are organized by two
different level of MBR m n o and P
represent four higher level mbrs and
each MBR has three or four lower level
mbrs for example M has AB b c d as a
lower level
MBR hierarchical structure can be
transferred to three structure as you
can see now which is
R3 similarly to B Tre it stores data
sets on the Le noes and it is a balanced
tree quad tree is another spatial data
indexing method as you can see the
example qu tree partitions to this space
into four quadrants
recursively and save the data to the end
nodes the figure shows the corresponding
point region quad
tree each node has exactly four child
nodes and the data is saved to the end
nodes which one do you think is better
than the other R Tree or quod tree of
course it depends on your spatial data
set generally speaking if you have
unevenly distributed data set R Tree
would be better because it can balance
the shape of the tree structure
on the other hand if you have evenly
distributed Point data quad Tre would be
choice now let's take a look at real
example of applying spal data indexing
and its performance the given query is
what are the name of the rows in
manhatton that is more than 30 ft width
in which SD contain function is used for
spatial relationship for the spatial
without indexing on the left and with
indexing on the right the difference of
the Quality Pro performance is 47 second
versus 1
second it is significant
Improvement what if we deal with the
load networks of the whole country the
phenomenal now rest the difference
between GIS and spatial dbms again we
discussed it in the second week GIS has
a variety of spatial data handling
capability but only to a certain degree
s dbms spal dbms is basically designed
for spatial data management with all the
wonderful features inherited from
dbns in this lecture
we started with the definition of
spatial dbms and object or relational
dbms was introduced as a dbms solution
for complex data types such as spatial
data Square for spatial data and spatial
joint was discussed finally two
different spatial indexing methods were
introduced and performance enhancement
with real example was given for your
understanding of the value of indexing method
method
all right this is the end of this
Click on any text or timestamp to jump to that moment in the video
Share:
Most transcripts ready in under 5 seconds
One-Click Copy125+ LanguagesSearch ContentJump to Timestamps
Paste YouTube URL
Enter any YouTube video link to get the full transcript
Transcript Extraction Form
Most transcripts ready in under 5 seconds
Get Our Chrome Extension
Get transcripts instantly without leaving YouTube. Install our Chrome extension for one-click access to any video's transcript directly on the watch page.
Chrome Extension