Data Mining the City @dmc-fa14 - Tumblr Blog

DepecheMood - a high-precision lexicon of roughly 37 thousand terms annotated with emotion scores, free for download on the Github. From this paper: Staiano, J., & Guerini, M. (2014). "DepecheMood: a Lexicon for Emotion Analysis from Crowd-Annotated News". Proceedings of ACL-2014.

Population Growth by County, 2012-2013.

Each marker is centered in a county. Its area is determined by its estimated population in 2013. The color of each marker is on a spectrum of dark red to dark blue, representing population loss to population growth respectively.

Source: American FactFinder

Map of access and utilization for Healthspaces in Shenzhen/Hong Kong Metro area based on Weibo check-ins for the past 12 months. Points are averaged over a grid of 1000 meter cells. Point color represents the number of checkins at healthspaces (Hospitals, Clinics, Physician and Dentist Offices). Point size represents the number of unique users checking in. The heatmap in gray is a representation of 10,000 meter catchment areas around large facilities (facilities with more than 100 unique users checking in during the study period). I used the bottom option for the rendering mode of the heatmap (the long Russian name), which essentially multiplies the gradients, this is nice (but still not quite right) because the folks in the overlaps have a different kind of access (to multiple locations). The normal heatmap rendering doesn't visually distinguish between being x distance from 1 point and being the same distance from 2 or more points. Next steps with this dataset include mapping a weighted centroid of each user’s previous check-in history in order to get a sense of how far out their way people go to reach healthspaces. I plan to assign a heavier weight to checkin points that occurred on same day as a healthspace checkin if they also occurred on days the user didn't check in at a healthspace. I hope to use those findings to re-plot the existing healthspace locations, visualizing a notional “de-stressed” heathspace network for the region. Jim Moffet

This map visualizes the major locations of NYC studio/ 1 bedroom on sale. The color indicates the price range, and the size of the circle represent number of listing in that area.

-Bruno Ma

This map shows the concentration of check-in points within the Art/Entertainment category within Shenzen. Using the Weibo check-in data set, I narrowed down on the check-in points that were under the Art/Entertainment category. Then I used grid counting to determine the concentration of check-in points throughout Shenzen. - Patrick Jalasco

This map visualizes shopping activities in Hong Kong. This heat map on the top right corner represents the total check-in density of weibo users. The main map represents total check-ins density( size of the circle) vs phototag density(color of circle).

- Ye Zhang (Phoebe)

Weibo Check In Count

Xiaobei Yang

Map Description: The image contains three maps, the left two visualize all Weibo Check-in data in Shenzhen, with a category of dinning (food & beverage).

The upper left one is dot density map based on the number of check-in user and number of photo posts of each check-in location. The lower left one is kernal density map based on check-in number of each check-in location.

The right one is an overview of primary development zones of Shenzhen. The left two maps respond to the development zones to some extent.

Techniques: For the upper left dot density map, it's created in QGIS. The graduated color is according to number of check-in users. The size of the dot illustrates the number of photo posts of each check-in location.

For the upper lower map, it's created in ArcGIS with Kernal Density, according to check-in number of each check-in location. The settings are: cell size = 100 meters, radius = 500 meters.

Name: Qihao Wang (qw2173)

This heat map uses the data retrieved from Baidu Map API. It shows the density of creative places in Shenzhen. From the map, we can see that the agglomeration of creative places are mainly in downtown areas, which are Luohu, Futian and Shekou.

----Juting Xu

1) the interpolation map try to analysis price for the renting apartment based on different location. The size of the circle indicate the renting price per sqft in Manhattan.

2} Based on the interpolation map build a contour map based on the price per sqft

Huiwen Zhu

The map shows the density of weibo users’ comments include creative keywords. (The comments include creative keywords are imported to QGIS and made a grid of 500*500. Then it is normalized by the total amount of checkins) The data derived from Sina weibo API(join Danil’s data-sets of weibo users and checkins. )

Ying Li

dmc-fa14

This map joins the data of density and price of rental apartment in New York City, comparing the areas of relatively high price and density in lower Manhattan.

dmc-fa14

- Jessica

dmc-fa14

This map shows the price interpolation of rental apartment in New York City, with the data scrapping from Street Easy.

- Jessica cheung

This map joins the data of density and price of rental apartment in New York City, comparing the areas of relatively high price and density in lower Manhattan.

This map shows the price interpolation of rental apartment in New York City, with the data scrapping from Street Easy.

This heatmap identifies areas of low and high affordability in Guangzhou and Shenzhen based on a calculation of price per room from Soufan data. Areas where the price per room is very high indicate low affordability and are highlighted in red. Conversely, areas where the price per room are low indicate high affordability and are highlighted in green.

-Houman Saberi

Trending Blogs

Recently Viewed Blogs

Data Mining the City