Icon

Dissertation

Feature Engineerin Jupyter Crime rate actually decreases as District populationincreases... Maybe fewer police officers per person,i.e. police are relatively understaffed? Crime rate increases significantly as Black pop %increases (because of over-policing). Crime Rate and Black Pop % follow extremelysimilar trend (would look better with 3 axes anddifferent colours, and maybe as a scatter plot...). Significant variation in crime count between Districts(provide T test).Also use this for Chicago Map. 4th of July seems to affect somecrimes - Battery increases, whileTheft, Narcotics and Prostitutiondecrease.Huge variation in Gambling -increases during summer. Prostitution increases around 7ammaybe because they're apprehendedafter leaving?Significantly different trends, seeingcrimes peak at different times of theday. Redistribute Dates inJupyter - FixingAnomalies N.B. Only full year data used, therefore 2017 not included.Sharp dip at 2004 - why? Possibly due to either change inpractice regarding crime reporting or an actual decrease incrime, due perhaps to introduction of policing initiatives, oran improvement in socioeconomic conditions. Slight increase at Friday - people going out - increasedlevels of socialisation as well as alcohol and drug intake.Slight decrease at Sunday because people tend to stay inand relax, therefore less opportunity for crime. Small peak at 12pm might be artificial -reporting officers default to the time outof convenience...Going to continue model-buildingwithout adjustment; in future, couldinvestigate reporting practices, thenjudge whether adjustment is needed. Crime Rate by YearVisualisation of Fill-rateFinal Model DataNode 103Node 104Node 105Node 106Node 108Node 119Day Length by Day of YearBox Plot of Crime Count by Day of YearCrime Count by Hour of DayCrime Count by DistrictCrime Rate vs Black Pop %Crime rate vs Pop SizeCrime Rate and Black Pop % by DistrictCrime Proportion by Type and Day of YearCrime Count by Type and Hour of DayCrime Count by Type and Day of YearCrime Proportion by Type and Hour of DayNode 180Node 181Node 184Node 185Box Plot of Crime Count by Hour of DayNode 188Fully Cleaned DataCrime Rate by Day of WeekNode 193Crime Count by Day of Year - *Before Fixing Anomalies*Node 199Node 200Crime Count by Day of Year - *After Fixing Anomalies*Node 202Node 203Node 204Node 205Node 206Node 207Count of Crime Count ValuesFinal DataNode 228Crime Rate by MonthNode 230Node 231Node 232Node 233Node 234Node 235Node 236Node 237Node 238Node 239Node 240Node 245Node 247Node 248Node 250Node 251Node 254Node 257Node 258Node 259 Line Plot Line Plot CSV Writer Filter for location Column Filter CSV Writer Loading, Cleaning,and Merging Raw Data CSV Reader General Tidying Scatter Plot Box Plot Line Plot Scatter Plot Scatter Plot Scatter Plot Line Plot Line Plot Line Plot Line Plot Line Plot Miscellaneous Misc Metanode Ranking Districtsby Crime Count Box Plot CSV Writer CSV Reader Line Plot CSV Reader Scatter Plot Getting Crime Rate Getting Crime Rate Scatter Plot Getting Crime Rate Getting Crime Rate Getting Crime Rate Row Filter Row Filter String Manipulation GroupBy Scatter Plot Grouping forPoisson Analysis Getting Crime Rate Line Plot Rule-basedRow Filter GroupBy Rule Engine CSV Writer GroupBy Math Formula Column Rename CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer Feature Engineerin Jupyter Crime rate actually decreases as District populationincreases... Maybe fewer police officers per person,i.e. police are relatively understaffed? Crime rate increases significantly as Black pop %increases (because of over-policing). Crime Rate and Black Pop % follow extremelysimilar trend (would look better with 3 axes anddifferent colours, and maybe as a scatter plot...). Significant variation in crime count between Districts(provide T test).Also use this for Chicago Map. 4th of July seems to affect somecrimes - Battery increases, whileTheft, Narcotics and Prostitutiondecrease.Huge variation in Gambling -increases during summer. Prostitution increases around 7ammaybe because they're apprehendedafter leaving?Significantly different trends, seeingcrimes peak at different times of theday. Redistribute Dates inJupyter - FixingAnomalies N.B. Only full year data used, therefore 2017 not included.Sharp dip at 2004 - why? Possibly due to either change inpractice regarding crime reporting or an actual decrease incrime, due perhaps to introduction of policing initiatives, oran improvement in socioeconomic conditions. Slight increase at Friday - people going out - increasedlevels of socialisation as well as alcohol and drug intake.Slight decrease at Sunday because people tend to stay inand relax, therefore less opportunity for crime. Small peak at 12pm might be artificial -reporting officers default to the time outof convenience...Going to continue model-buildingwithout adjustment; in future, couldinvestigate reporting practices, thenjudge whether adjustment is needed. Crime Rate by YearVisualisation of Fill-rateFinal Model DataNode 103Node 104Node 105Node 106Node 108Node 119Day Length by Day of YearBox Plot of Crime Count by Day of YearCrime Count by Hour of DayCrime Count by DistrictCrime Rate vs Black Pop %Crime rate vs Pop SizeCrime Rate and Black Pop % by DistrictCrime Proportion by Type and Day of YearCrime Count by Type and Hour of DayCrime Count by Type and Day of YearCrime Proportion by Type and Hour of DayNode 180Node 181Node 184Node 185Box Plot of Crime Count by Hour of DayNode 188Fully Cleaned DataCrime Rate by Day of WeekNode 193Crime Count by Day of Year - *Before Fixing Anomalies*Node 199Node 200Crime Count by Day of Year - *After Fixing Anomalies*Node 202Node 203Node 204Node 205Node 206Node 207Count of Crime Count ValuesFinal DataNode 228Crime Rate by MonthNode 230Node 231Node 232Node 233Node 234Node 235Node 236Node 237Node 238Node 239Node 240Node 245Node 247Node 248Node 250Node 251Node 254Node 257Node 258Node 259 Line Plot Line Plot CSV Writer Filter for location Column Filter CSV Writer Loading, Cleaning,and Merging Raw Data CSV Reader General Tidying Scatter Plot Box Plot Line Plot Scatter Plot Scatter Plot Scatter Plot Line Plot Line Plot Line Plot Line Plot Line Plot Miscellaneous Misc Metanode Ranking Districtsby Crime Count Box Plot CSV Writer CSV Reader Line Plot CSV Reader Scatter Plot Getting Crime Rate Getting Crime Rate Scatter Plot Getting Crime Rate Getting Crime Rate Getting Crime Rate Row Filter Row Filter String Manipulation GroupBy Scatter Plot Grouping forPoisson Analysis Getting Crime Rate Line Plot Rule-basedRow Filter GroupBy Rule Engine CSV Writer GroupBy Math Formula Column Rename CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer CSV Writer

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: curtu
Created at: 2023-07-10
On NodePit since: 2023-08-26
Last update: 2024-04-27
Created with KNIME version: v4.7.5

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!