Below is the link to the Visualization:
Motor Vehicle Collisions in New York City
New York City is one of the most populous and big cities in the world. In a city such as New York, traffic incidents are more common and the concerned authorities work on ways to make city more safer for the people. The purpose of the application is to analyze the traffic incidents occurring in and around the New York City. The application visualizes accidents involving Motorists, Cyclist and Pedestrians separately. This enables Police departments and other city authorities to see where most of the accidents occur and find out the reason behind these incidents. Based on that the authorities can come up with solutions such as diverting the traffic, installing more traffic signals and other regulations or construction modifications to avoid these incidents in future making New York roads more safe for people.
The New York Police Department provides data for every motor vehicle collision in NYC since July 2012. This data is collected because the NYC Council passed Local Law #11 in 2011. This data is manually run every month and reviewed by the TrafficStat Unit before being posted by the NYPD. The method on how the data is collected is not provided by the source. Each record represents a collision in NYC by city, borough, precinct and cross street, most notably the number of injuries and fatalities, segmented further by motorists, cyclists, and pedestrians. The data visualizes about 1.4 million collisions between July 2012 and January 2019.
Here is the link to the dataset: Link
1. The New York Police Department can use this visualization to see where the major accidents are occurring in the New York City and help them decide if they need to implement any restrictions on the traffic to avoid the accidents.
2. City Planners and architect to see where most of the accidents occur in New York City and analyze if these accidents are due to improper city layout and planning. This helps them avoiding the same mistakes and come up with a better city planning in their forthcoming projects.
3. General public can use this data to see the areas where it is safe to do cycling, walking and other activities.
4. Apart from this the data can be useful to media persons, government officials and other authorities who needs to see the accidents involving motorists, cyclists and pedestrians around New York Area.
The user has the option to choose the Type of Collisions/Accidents/People involved in the accidents, Boroughs and Year (From 2012 to 2019) to Visualize. The options can be picked from the drop down list provided. Based on the user's selection the application produces a heat map zoomed over the New York region. The user can zoom in and out the map provided to see the detailed values for the exact place required. The legend for the color coding is given at the bottom-right of the application. The user can hover over the map to see the detailed values of that location. The user can also switch between 2D and 3D mode. The 3D mode projects the visualization on the map in form of a histogram. The application auto adjusts the Hexagon radius of the each plot based on the zoom level on the map.
The Colors used to differentiate between various plots are good and stand out different from one other. The color of the heat map is made in such a way that brighter color has the greater value and the darker show the lesser in scale. The colors can be differentiated easy. The colors are useful when the plots are seen in 2D, when the 3D plot option is chosen the user can see the difference in values through histograms in 3D.
The map has been made interactive, has options to move around, zoom-in, zoom-out and change the angle as well. The map can be moved and the angle of the map can be changed (press shift + drag). This makes the user see the comparison of the 3-D plot. When we set the Hexagon radius of the minimum and zoom-in the map, we can see the exact location of the incidents and number. This makes it easy to pinpoint exact location in the map visualization rather than needing extra data to see the details.
The option to switch to 3-D histogram makes the users easier to compare the various location in the New York. The user can adjust the angle of the map to see the differences in the various regions. The 3D histogram is also made dynamic based on the zoom done on the map.
The user can use the hover option to see the details of the location. The user can place the cursor on each of the hexagons to see the details of that location with a pop-up that appears. The pop-up provides details about the total number of the collisions, type of person (Motorist, Cyclist and Pedestrians), count of injured and killed at that location.
The data processing and loading in the application is done so efficiently that we cannot find any lag in the application. Everything loads pretty fast. Processing around 1.4 million records of data such efficiently is a thing to appreciate. As the size of the hexagon auto adjusts every time the application recalculates the new value of location to which the hexagon adjusts to, this is a good feature since the values are dynamic based on the size of the location represented.
The background chosen for the map makes it difficult for the user to different smaller values. The gray background of the map makes it difficult for the user to differentiate between the black points in the visualization. This makes it even more difficult when the majority of the values fall in the lower range. This is a serious issue with the visualization. The black color in the plot matches with the river and other water bodies in the map, that can be confusing to the user. The user could have added different varieties for backgrounds to the map and colors as well.
We do not have the details about how the data is collected. The data is provided by the New York Police department and is published on the NYC Open Data website. They did not provide the details about how the data collection is done and the method behind it. If the data collector would have given these methods of collection then the users would know the credibility of the data.
The legend at the bottom right shows only the minimum and the maximum values. It doesn't show the intervals for each of the color coding in the legend. This makes it difficult for the user to see the range for each color in the legend.