Assigning sentiment to themes

1. A system, comprising:

a processor configured to:

receive reputation data extracted from at least one data source, wherein the reputation data includes user-authored reviews pertaining to an entity and wherein a review comprises text and an accompanying rating;

for a first review included in the reputation data, identify a theme in the first review that is associated with an aspect of the entity at least in part by determining at least one keyword using the first review's text;

assign a first sentiment for the identified theme at least in part by using the first review's accompanying rating;

assign a sentiment for a parent theme of the identified theme at least in part by using the first review's accompanying rating;

for a second review included in the reputation data that includes the determined at least one keyword and in which the theme is identified, assign a second sentiment for the identified theme based at least in part on the second review's accompanying rating; and

provide as output a report that includes the identified theme, a combination of the respective first and second sentiments for that identified theme, the parent theme, and the sentiment for the parent theme; and

a memory coupled to the processor and configured to provide the processor with instructions.

2. The system of claim 1 wherein assigning the first sentiment for the identified theme includes assigning a sentiment for the determined at least one keyword.

3. The system of claim 1 wherein the combination comprises an aggregation.

4. The system of claim 1 wherein the at least one keyword determined included in the first review is a first keyword and wherein, for a third review that includes a second keyword that is associated with the identified theme and is included in the reputation data, the processor is configured to:

assign a third sentiment for the identified theme based at least in part on the third review's accompanying rating, wherein the report includes a combination of the respective first, second, and third sentiments for the identified theme.

5. The system of claim 1 wherein the processor is further configured to detect whether the review data includes cloned reviews.

6. The system of claim 5 wherein the detection is based at least in part on a determination that the first review and a second review include an identical clause.

7. The system of claim 1 wherein determining the at least one keyword includes applying a blacklist filter.

8. The system of claim 1 wherein the processor is further configured to generate a report that indicates, for a given entity, a set of keywords present in a plurality of user-authored reviews.

9. The system of claim 1 wherein the processor is further configured to generate a report that indicates, for the entity, a comparison of a sentiment score for a given keyword compared against an industry average sentiment score for the keyword.

10. The system of claim 1 wherein the processor is further configured to generate a report that indicates, for the entity, a comparison of the entity's score for a given keyword against a score of a competitor for the keyword.

11. The system of claim 1 wherein the processor is further configured to generate a report that indicates, for two locations associated with the entity, a comparison of the first location's score for a given keyword against the second locations' score for the keyword.

12. The system of claim 1 wherein the processor is further configured to locate the determined at least one keyword within a hierarchy of keywords and to assign a sentiment to at least one parent of the keyword in the hierarchy based at least in part on the first review's accompanying rating.

13. The system of claim 1 wherein the processor is further configured to perform positivity smoothing prior to generating a report.

14. The system of claim 1 wherein assigning a first sentiment includes assigning a sentiment score.

15. The system recited in claim 1 , wherein assigning the sentiment includes assigning a value to the identified theme, the value corresponding at least in part to the first review's accompanying rating.

16. A method, comprising:

receiving reputation data extracted from at least one data source, wherein the reputation data includes user-authored reviews pertaining to an entity and wherein a review comprises text and an accompanying rating;

for a first review included in the reputation data, identifying a theme in the first review that is associated with an aspect of the entity at least in part by determining, using a processor, at least one keyword using the first review's text;

assigning a first sentiment for the identified theme at least in part by using the first review's accompanying rating;

assigning a sentiment for a parent them of the identified theme at least in part by using the first review's accompanying rating;

for a second review included in the reputation data that includes the determined at least one keyword and in which the theme is identified, assigning a second sentiment for the identified theme based at least in part on the second review's accompanying rating; and

providing as output a report that includes the identified theme, a combination of the respective first and second sentiments for that identified theme, the parent theme, and the sentiment for the parent theme.

17. A computer program product embodied in a non-transitory tangible computer readable storage medium and comprising computer instructions for:

receiving reputation data extracted from at least one data source, wherein the reputation data includes user-authored reviews pertaining to an entity and wherein a review comprises text and an accompanying rating;

for a first review included in the reputation data, identifying a theme in the first review that is associated with an aspect of the entity at least in part by determining at least one keyword using the first review's text;

assigning a first sentiment for the identified theme at least in part by using the first review's accompanying rating;

assigning a sentiment for a parent theme of the identified theme at least in part by using the first review's accompanying rating;

for a second review included in the reputation data that includes the determined at least one keyword and in which the theme is identified, assign a second sentiment for the identified theme based at least in part on the second review's accompanying rating; and

providing as output a report that includes the identified theme, a combination of the respective first and second sentiments for that identified theme, the parent theme, and the sentiment for the parent theme.