Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

The framework for obtaining social media messages about “outdoor air pollution”.

The framework containing two phases: data collection and data pre-processing.

More »

Fig 1 Expand

Table 1.

Different types of microblogs collected in Beijing relevant to air pollution.

More »

Table 1 Expand

Fig 2.

Examples of high correlations between AQI and individual messages.

(A) The trends of AQI (red line) and individual messages (blue line) for January 2012 in Beijing. (B) The trends of AQI (red line) and individual messages (blue line) in April 2012 in Beijing.

More »

Fig 2 Expand

Table 2.

Correlation coefficient between AQI and individual messages.

More »

Table 2 Expand

Fig 3.

An example of the low correlation between AQI and individual messages.

The trends of AQI (red line) and the individual messages (blue line) for July 2012 in Beijing. Point A is the lowest AQI value and point B is the highest daily frequency of individual messages in July.

More »

Fig 3 Expand

Fig 4.

An example of the high correlation between AQI and negative individual messages.

The trends of AQI (red line) and negtive individual messages (blue line) for July 2012 in Beijing.

More »

Fig 4 Expand

Table 3.

Correlation coefficient between AQI and negative individual messages.

More »

Table 3 Expand

Fig 5.

An example of high correlation between AQI and pollution app messages.

The trends of AQI (red line) and pollution app messages (blue line) for July 2012 in Beijing.

More »

Fig 5 Expand

Table 4.

Correlation coefficient between AQI and app messages.

More »

Table 4 Expand

Fig 6.

An example of the low correlation between AQI and retweets.

The trends of AQI (red line) and retweet (blue line) for June 2012 in Beijing. Point C and point D are the highest two daily frequencies of retweets in June.

More »

Fig 6 Expand

Table 5.

Correlation coefficient between AQI and retweets.

More »

Table 5 Expand

Fig 7.

The inferred AQI and the observed AQI in each month in 2013.

The trends of observed AQI (red line) and inferred AQI (blue line) in each month in 2013.

More »

Fig 7 Expand

Table 6.

RMSE for each month in 2013.

More »

Table 6 Expand

Fig 8.

The page of Sina Weibo users’ profile.

Basic information of user’s profile. The user-defined geolocation of Sina Weibo users is selected from the drop-down list and different from the Twitter with open entry fields.

More »

Fig 8 Expand