Saturday, June 21, 2014

How safe is New York?

MIT has released a number of maps showing areas of perceived safety in New York, Boston, Chicago and Detroit. Using Street View images of the cities StreetScore assesses the perceived safety of locations throughout the city. Green dots on the map represent the areas which StreetScore has assigned as having a high perceived safety rating and the red dots are the locations with a low perceived rating score.

One thing missing from the MIT StreetScore maps is the ability to filter the results shown on the map by score. However MIT has made the data available for download so I decided to make my own map of New York and add some filters.

In this Safe New York map I've taken the data for Manhattan, New York. The map shows the safest and least safe locations as perceived by the MIT Street View algorithm. The MIT q-score assigns a value up to 43 for each Street View image. The higher the q-score assigned the higher the safety perception. The lower the q-score the lower the safety perception.

I've marked every location which has a score under 15 with a red marker and every location with a score over 30 with a green marker. All locations with a score between 15 and 30 are displayed with a yellow marker.

You can filter the markers shown on the map by using the four buttons in the map sidebar.

I think there is some value in being able to query the data in this way. Switching between the highest rated and lower rated locations on the map (green & red buttons) is quite effective in identifying areas where there are clusters of locations perceived as unsafe and with few safe areas.

For example Hunts Point and Wards Island can be clearly identified as areas that are perceived unsafe by the MIT algorithm. It also becomes clear by just looking at the lowest (red) scores that major roads are often perceived as unsafe.

Using the Google Maps API also means that you can use Google Maps Street View. You can drag and drop pegman onto the map to check out the Street View images for any location. Therefore if you are wondering about why a particular area has been given low scores you can check out the Street View imagery for yourself and make your own judgement.

No comments: