METHODS AND APPARATUS TO ESTIMATE DEMOGRAPHICS OF AN AUDIENCE OF A MEDIA EVENT USING SOCIAL MEDIA MESSAGE SENTIMENT
Methods, apparatus, systems and articles of manufacture to estimate demographics of an audience of a media event using social media message sentiment are disclosed. An example method includes calculating a sentiment score for a media-exposure social media message received from a social media server. The media-exposure social media message is identified based on a media keyword. Demographic information associated with users who viewed the media-exposure social media message is retrieved. An estimated impact on a size of a demographic segment of viewers is calculated using the sentiment score and the demographic information.
1 . A method to estimate demographics for an audience of a media event, the method comprising:
calculating, with a processor, a sentiment score for a media-exposure social media message received from a social media server, the media-exposure social media message identified based on a media keyword;
retrieving demographic information associated with users who viewed the media-exposure social media message; and
calculating an estimated impact on a size of a demographic segment of viewers using the sentiment score and the demographic information.
2 . The method as defined in claim 1 , wherein calculating the sentiment score further comprises:
parsing a text of the media-exposure social media message to identify a keyword;
associating the keyword with a value; and
summing the value for a plurality of keywords identified in the text of the media-exposure social media message.
3 . The method as defined in claim 1 , further comprising identifying the media-exposure social media message by:
identifying a reference to the media event in a text of the media-exposure social media message; and
determining whether a characteristic of the media-exposure social media message satisfies a rule associated with the media event.
4 . The method as defined in claim 3 , wherein the rule associated with the media event is a broadcast time of the media event.
5 . The method as defined in claim 3 , further comprising calculating the sentiment score for the media-exposure social media message in response to determining that the characteristic of the media-exposure social media message satisfies the rule associated with the media event.
6 . The method as defined in claim 1 , wherein determining the demographics associated with users of social media exposed to the media-exposure social media message comprises:
identifying an impression associated with the social media message, the impression corresponding with exposure to the social media message;
identifying a user identifier associated with the impression; and
determining demographics associated with the user identifier.
7 . The method as defined in claim 1 , wherein the sentiment score is a first sentiment score, the media-exposure social media message is a first media-exposure social media message, the estimated impact is a first estimated impact, and further comprising:
calculating a sentiment score associated with a second media-exposure social media message; and
calculating an second estimated impact on the size of the demographic segment of viewers using the second sentiment score and the demographic information; and
calculating a total estimated impact on the size of the demographic segment of viewers using the first estimated impact and the second estimated impact.
8 . An apparatus to estimate demographics for an audience, the apparatus comprising:
a query transmitter to transmit, to a social media server, a query requesting a media-exposure social media message and demographic information associated with impressions of the media-exposure social media message, the media-exposure social media message associated with media;
a sentiment estimator to calculate a sentiment score for the media-exposure social media message; and
an audience demographic estimator to calculate an estimated impact on a size of a demographic segment based on the sentiment score and the demographic information.
9 . The apparatus as defined in claim 8 , wherein the sentiment estimator is to parse a text of the media-exposure social media message to identify a presence of a sentiment keyword within the text of the media-exposure social media message.
10 . The apparatus as defined in claim 8 , further comprising a query generator to generate the query based on a media profile identifying when the media is presented.
11 . The apparatus as defined in claim 8 , further comprising a query generator to generate the query based on a media keyword associated with the media.
12 . The apparatus as defined in claim 8 , wherein the sentiment score is a first sentiment score, the media-exposure social media message is a first media-exposure social media message, the demographic information is a first demographic information, the audience demographic estimator is further to calculate the estimated impact on the size of the demographic segment based on a second sentiment score associated with a second media-exposure social media message and second demographic information associated with the second media-exposure social media message.
13 . The apparatus as defined in claim 8 , further comprising a reporter to generate a report indicative of the size of the demographic segment.
14 . A tangible machine readable storage medium comprising instructions which, when executed, cause a machine to at least:
calculate a sentiment score for a media-exposure social media message received from a social media server, the media-exposure social media message identified based on a media keyword;
retrieve demographic information associated with users who viewed the media-exposure social media message; and
calculate an estimated impact on a size of a demographic segment of viewers using the sentiment score and the demographic information.
15 . The machine readable storage medium as defined in claim 14 , further comprising instructions which, when executed, cause the machine to at least:
parse a text of the media-exposure social media message to identify a keyword;
associate the keyword with a value; and
sum the value for a plurality of keywords identified in the text of the media-exposure social media message.
16 . The machine readable storage medium as defined in claim 14 , further comprising instructions which, when executed, cause the machine to at least:
identify a reference to the media event in a text of the media-exposure social media message; and
determine whether a characteristic of the media-exposure social media message satisfies a rule associated with the media event.
17 . The machine readable storage medium as defined in claim 16 , wherein the rule associated with the media event is a broadcast time of the media event.
18 . The machine readable storage medium as defined in claim 16 , further comprising instructions which, when executed, cause the machine to calculate the sentiment score for the media-exposure social media message in response to determining that the characteristic of the media-exposure social media message satisfies the rule associated with the media event.
19 . The machine readable storage medium as defined in claim 14 , further comprising instructions which, when executed, cause the machine to at least:
identify an impression associated with the social media message, the impression corresponding with exposure to the social media message;
identify a user identifier associated with the impression; and
determine demographics associated with the user identifier.
20 . The machine readable storage medium as defined in claim 14 , wherein the sentiment score is a first sentiment score, the media-exposure social media message is a first media-exposure social media message, the estimated impact is a first estimated impact, and further comprising instructions which, when executed, cause the machine to at least:
calculate a sentiment score associated with a second media-exposure social media message; and
calculate an second estimated impact on the size of the demographic segment of viewers using the second sentiment score and the demographic information; and
calculate a total estimated impact on the size of the demographic segment of viewers using the first estimated impact and the second estimated impact.