Selecting sentiment attributes for visualization

1. A method comprising:

receiving, by a system including a processor, data records containing user feedback regarding at least one offering;

from among candidate attributes in the received data records, selecting, by the system, a subset of attributes that relate to user sentiment regarding the at least one offering, wherein the selecting is according to selection criteria including frequency of occurrence of the candidate attributes, relative amounts of negative and positive feedback associated with the candidate attributes, and time density of feedback associated with the candidate attributes; and

presenting, by the system, for display a visualization of at least one sentiment characteristic of the selected subset of attributes that relate to user sentiment.

2. The method of claim 1 , wherein the selecting of the subset of attributes is further according to a selection criterion that places greater importance on negative feedback associated with the candidate attributes.

3. The method of claim 1 , wherein the selecting of the subset of attributes is further according to an application-specific selection criterion.

4. The method of claim 1 , wherein the selecting is further based on user selection or deselection made in the visualization.

5. The method of claim 1 , wherein presenting for display the visualization comprises presenting for display the visualization that includes cells representing the at least one sentiment characteristic of at least one of the attributes in the subset.

6. The method of claim 5 , further comprising:

assigning different colors to the cells to indicate respective negative feedback, positive feedback, and neutral feedback associated with the at least one attribute and degree of positive and negative.

7. The method of claim 5 , further comprising:

arranging the cells in an arrangement where different columns correspond to different first time intervals, and different rows correspond to different second time intervals, each second time interval having a different length than each first time interval.

8. The method of claim 1 , wherein presenting for display the visualization comprises presenting for display a bar chart having plural bars corresponding to different ones of the attributes in the subset, where each of the bars has cells assigned different visual indicators to indicate different types of feedback.

9. The method of claim 8 , wherein the visual indicators are assigned based on sentiment scores of respective attributes, the method further comprising:

ordering, in each of the bars, the cells according to the sentiment scores such that each bar includes plural portions, each of the plural portions including a respective group of cells having the same visual indicator.

10. The method of claim 1 , wherein presenting for display the visualization comprises presenting for display a graph chart having a first graph corresponding to positive feedback of the attributes in the subset, and a second graph corresponding to negative feedback of the attributes in the subset.

11. The method of claim 1 , further comprising using visualization analytics to select the selection criteria from among a larger collection of selection criteria to use.

12. An article comprising at least one non-transitory machine-readable storage medium storing instructions that upon execution cause a system having a processor to:

receive data records containing user feedback regarding at least one offering;

extract candidate attributes from the received data records;

select a subset of attributes from among the candidate attributes, where the attributes in the selected subset relate to user sentiment regarding the at least one offering, wherein the selecting is according to selection criteria including frequency of occurrence of the candidate attributes, relative amounts of negative and positive feedback associated with the candidate attributes, and time density of feedback associated with the candidate attributes; and

present for display a visualization containing a bar chart including plural bars corresponding to the attributes in the selected subset, wherein each of the plural bars depict visual indicators of at least one sentiment characteristic of the corresponding attribute in the selected subset.

13. The article of claim 12 , wherein each of the plural bars includes cells associated with individual ones of the data records, and wherein the cells are assigned the respective visual indicators.

14. The article of claim 13 , wherein the assigned visual indicators comprise colors.

15. The article of claim 13 , wherein the instructions upon execution cause the system to further:

order the cells in each of the plural bars according to the at least one sentiment characteristic, such that at least one of the plural bars has plural portions-cells, each of the portions having cells of the same respective color.

16. The article of claim 12 , wherein the at least one sentiment characteristic is a sentiment score indicating user sentiment for each of the attributes of the selected subset.

17. The article of claim 12 , wherein the instructions upon execution cause the system to further:

present for display a second visualization having a graph chart, the graph chart having a first graph for positive feedback regarding the attributes in the selected subset, and a second graph for negative feedback regarding the attributes in the selected subset.

18. The article of claim 17 , wherein each point along the first graph represents a percentage of positive reviews in the data records for a corresponding one of the attributes in the selected subset, and each point along the second graph represents a percentage of negative reviews in the data records for a corresponding one of the attributes in the selected subset.

19. A system comprising:

a network interface to receive data records containing user feedback regarding at least one offering;

at least one processor to:

from among candidate attributes in the received data records, select a subset of attributes that relate to user sentiment regarding the at least one offering, wherein the selecting is according to selection criteria including frequency of occurrence of the candidate attributes, relative amounts of negative and positive feedback associated with the candidate attributes, and time density of feedback associated with the candidate attributes; and

present for display a visualization of at least one sentiment characteristic of the selected subset of attributes that relate to user sentiment.

20. The system of claim 19 , wherein the visualization includes at least one from among:

a bar chart having plural bars containing visual indicators representing the at least one sentiment characteristic for the attributes in the selected subset; and

a graph chart having graphs corresponding to positive and negative reviews for the attributes in the selected subset.

21. The method of claim 1 , wherein the selecting according to the relative amounts of negative and positive feedback associated with the candidate attributes comprises selecting based on comparing the amount of negative feedback associated with a given one of the candidate attributes to the amount of positive feedback associated with the given candidate attribute.

22. The method of claim 1 , wherein the selecting according to the time density of feedback comprises selecting based on determining an average time gap between feedback for each of the candidate attributes.

23. The article of claim 12 , wherein the selecting according to the relative amounts of negative and positive feedback associated with the candidate attributes comprises selecting based on comparing the amount of negative feedback associated with a given one of the candidate attributes to the amount of positive feedback associated with the given candidate attribute.

24. The article of claim 12 , wherein the selecting according to the time density of feedback comprises selecting based on determining an average time gap between feedback for each of the candidate attributes.