GIS Policy Map for Local Government in Korea: Story of Dobong-gu, Seoul
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 183 | |
| Author | ||
| License | CC Attribution - NonCommercial - ShareAlike 3.0 Germany: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this | |
| Identifiers | 10.5446/32096 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language | ||
| Producer | ||
| Production Year | 2015 | |
| Production Place | Seoul, South Korea |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
FOSS4G Seoul 2015167 / 183
7
8
47
53
54
65
73
74
79
82
84
92
102
103
105
124
126
127
130
141
142
143
156
161
162
170
176
178
181
183
00:00
Level (video gaming)Local ringMereologyInformationData analysisArchaeological field surveyCurve fittingPhase transitionSummierbarkeitInheritance (object-oriented programming)Archaeological field surveyStudent's t-testTraffic reportingMathematical analysisInformation privacyProjective planeResultantRaw image formatSlide ruleMereologyInsertion lossCausalityAutomatic differentiationLine (geometry)Computer animation
02:40
FrequencyDevice driverContext awarenessLibrary (computing)Point (geometry)Type theoryNP-hardSpecial unitary groupPort scannerTablet computerBinary filePerformance appraisalCellular automatonTheory of relativityPhysical systemReading (process)Inheritance (object-oriented programming)Point cloudRight angleDevice driverDependent and independent variablesType theoryDecision tree learningChromosomal crossoverStudent's t-testWritingInterrupt <Informatik>Arithmetic meanMathematical analysisText miningSlide ruleMultiplication signMetreCoefficient of determinationAverageMeasurementAreaTable (information)System callProcess (computing)Level (video gaming)Line (geometry)Constructor (object-oriented programming)Vector spaceBinary codeGraph coloringEndliche ModelltheorieNetwork topologyAssociative propertyFrequencyLibrary (computing)Rule of inferenceRaw image formatView (database)NP-hardTotal S.A.ResultantWordPolygonComplex (psychology)Pattern languageComputer animation
09:22
Performance appraisalSpecial unitary groupAdditionDreizehn3 (number)Projective planeLevel (video gaming)ResultantDisk read-and-write headStudent's t-testInformationOffice suiteBlock (periodic table)NumberDependent and independent variablesCellular automatonPresentation of a groupLine (geometry)Inheritance (object-oriented programming)Patch (Unix)Reflection (mathematics)Uniform resource locatorRaw image formatAreaMeasurementProcess (computing)WorkloadNP-hardFile formatSet (mathematics)PlanningMetropolitan area networkArchaeological field surveyCountingWeightReal numberArithmetic meanFrequencyLocal ringType theoryMedical imagingPattern languageFilm editingComputer animation
16:04
Computer animation
Transcript: English(auto-generated)
00:05
Hello, my name is Yongjae Park. Nice to meet you. I'm working at GIS United in Korea, Seoul, and we provide consulting report based on GIS analysis. Today, I'm going to introduce about
00:22
child safety on school walkway. Introduce Tobongu. Tobongu is here. This is Seoul district, and we are here, K Hotel. Tobongu is at edge of the Seoul north.
00:43
And there is about 20 primary schools in Tobongu. This project concept, from next slide,
01:01
I'm going to introduce and explain how participation and big data analysis, that's all joined. We do not focus on administrative public data to analyze children's safety,
01:22
but we mainly use survey data. This is our paper questionnaires. We ask to draw the line, you, the student, walk to school,
01:40
and where the spot they feel fear or dangerous. About 4,000 students and their parents participated this survey in 20 primary schools. This is examples, results.
02:00
Some students draw this spot because illegal parking causes some dangerous result. Some students said street lamp doesn't work here.
02:21
Some students didn't answer about dangerous spot, just only draw the line, he walk through, walk to school. This is where he live. This is the main entrance of the school. This paper raw data, we start to construct all the data
02:43
from paper with four women who run their home. These are all the paper we gathered. These are process using QGIS.
03:03
These narrow lines are, all lines are students go to school, walk, and I select one feature, ID 10015 student, wait to school, from here to main entrance of school.
03:21
This is the dangerous spot. We gather the data, not point data. We gather the polygon data because they draw where they feel dangerous spot, like road or area. Sometimes they draw a whole parking, a park,
03:44
so we construct data in polygon data. This is the result. They, every day, every morning in Dobon district, Dobon-gu, students walk about 1,766 kilometer,
04:05
and they feel fear in area, I find some, and about 14.3 square kilometer. We think this is big data because
04:21
it's more, longer than trip from Shanghai to Sakhapia, so we always think this is big data. We, and then, before GIS analysis,
04:43
we try text, we try to text analysis using R package, two kinds of library, a word cloud and associate rules. Using word cloud, we extract keywords in complex responses. The R package shows us 312 keywords
05:02
with 12,914 frequencies. Here, most frequent keywords was many cars, and the second one is drive too fast and dark traffic light, so we divide data
05:23
each 20 primary schools and make it a word cloud. Most schools, top keywords look similar, but some primary schools show different pattern. Then we try to analyze relations between keywords
05:44
because we couldn't understand the reason why the students and their parents describe some keywords. For example, can you imagine why supermarket dangerous? Through relation analysis and read raw sentences,
06:01
we could understand the keyword supermarket relate with big car. The respondent said, sometimes full-size car or trucks are reversing when children go to school, so we understand this keyword means big car.
06:23
Based on text analysis, we categorize the typology of dangerous situation. Nine types of situation is on the slide. I'm going to explain each type. First, too many cars, it includes keywords like apartment, supermarket, church.
06:41
The student's parents dislike when they see cars on the school road just because. Second type is bad driving, bad drivers. It includes illegal return and drive too fast and so on. Keywords in view interrupted, it means interrupted,
07:03
interrupt to see each other, drivers, and children, so three illegal parking, standing signboard are selected. Here, hard to notice means on frequent road or narrow dark street or alleys.
07:24
We could know what kind of dangerous student and their parents concern. These five types are about traffic accident. Some of these percentages, 75%, so we knew that.
07:42
It also different between each schools. For example, respondents in C school have stressed about bad drivers more than other schools or total average, about 30%. After text mining analysis, we start to analyze
08:02
using GIS, especially QGIS. These are research process. I'm going to introduce each step from next slide. This is the raw data we constructed. Leveling all the responses, actually complaints from young citizen.
08:22
We open the table, next we open the table, complaints written, and insert nine columns and write binary values, each type, zero or one. And then these maps are reserved after write binary values.
08:42
We can check respondents of each type as you see the slide. First one is all the type of dangerous spot. Second one is want to avoid situation spot. And then bad drivers are here. Hard to notice dangerous spot is here. Nine type of maps.
09:05
And then we put the 10 meter by 10 meter grid on our raw data. We try to join in to vector data set, special join, and find some of the binary values.
09:21
Like this. Through this process, we evaluate each grid cell to know which location is more dangerous to others relatively. This is the results map. You may see the numbers in the cell. That is some of the responses. This cell is called highest value, about 69.
09:42
And these cells count only one or two. It also, this kind of, this map is about too many cars. This map is hard to notice.
10:02
This map is want to avoid situation, very different. And then we recheck keywords and full sentence is hotspot of each line type. You may see this bold blocks. This is about, this is hotspot of this type.
10:24
Hard to notice hotspot. And then we check raw data of keywords. It means lack of CCTV cameras and unfrequency people. And then it also same map about transport infrastructure
10:45
hotspot, this bold areas hotspot. And we gathered keywords again and try to pick top keywords in hotspot. So next, after this process,
11:03
then we also use patch drawing using the school walkway and grid. So this cell is where students, most students walk through, 144.
11:28
This is final map, summarize safety information map. This is, we designed map manually like tour map. Pull the keywords and hotspot,
11:40
only using top keywords and top hotspot. So we designed this map every each 20 primary schools. Provide them this map. After this analysis, Tobong officers carried out spot inspection
12:07
based on the map. CCTV check and narrow road, sheet lamp and reflector. And then we check the safety info map
12:22
and result of inspection they describe. So now they plan to project to solve these problems. Thank you for listening.
12:46
Okay, thank you very much. Are there any questions or comments? Well, I do have a question because this was very interesting for me because only a few weeks ago, they opened a new school in my village and everybody's complaining about safety
13:02
because of the heavy traffic. So I really like your approach of announcing where the spots are, where people find themselves unsafe. My question to you, because you did a nice survey, did you went to the local government and presented them and say, well, did they took any measures or how did they respond?
13:24
Have you been to the local government with your findings or is that the next step? I'm sorry. Officers in ultimate public really like this process
13:45
but especially who runs budget really like this and who runs spot inspection didn't like this because they have additional workload from this time.
14:02
And they do not provide real map to primary school student and their parents because they fight, we want CCTV more than you. No, no, we are more dangerous, they fight.
14:23
So this is only for officers who run budget.
14:59
Yes, actually this is second project.
15:01
First project was just for one pilot project for one primary school and the results was really, really clapped. So the head of this Togumu office, head of Togumu office want to 20 older primary school.
15:26
So actually the end of this project is next week. The end of this project is next week but for this presentation I finished this project
15:43
last night.