Summarizing reviews
Summarizing a set of reviews is disclosed. In some embodiments, a set of reviews is analyzed, e.g., by an at least partially automated process. A summary of the information included in the set of reviews is provided. The summary includes a visual indication of a range and distribution of opinions expressed in the set of reviews. In some embodiments, the set of reviews includes reviews from one or more members of an online or other user community, such as customers of an online store, subscribers to a podcast, blog, or other online source of content, etc.
1. A computer-implemented method for summarizing a set of reviews, comprising:
extracting a plurality of keywords from a set of reviews that are associated with a product, each review containing content expressing opinions of the product, wherein the plurality of keywords are selected from the content based on a relevance score of multiple tuples of words in the content, wherein the multiple tuples include various permutations of the same root words, and wherein the relevance score is based on the spacing between two words in each tuple in the multiple tuples;
selecting a subset of reviews from the set of reviews, each review in the subset of reviews containing a keyword from the plurality of keywords;
generating a keyword rating value based on the content belonging to the subset of reviews; and
presenting a summary of the set of reviews, the summary being configured to graphically display the keyword and the corresponding keyword rating value along with another keyword and another corresponding keyword rating value simultaneously.
2. The computer-implemented method of claim 1 , wherein the content includes a reviewer-specified rating value rating the product and wherein generating the keyword rating value further comprises calculating the average or median of the reviewer-specified rating value belonging to each review in the subset of reviews.
3. The computer-implemented method of claim 1 , wherein the tuples of words are segments of the content that do not contain a word from a predefined stop list of words.
4. The computer-implemented method of claim 1 , wherein the relevance score indicates a degree of relevance of the keyword to the set of reviews.
5. The computer-implemented method of claim 1 , wherein a user to whom the summary is presented is provided an opportunity to vote on a usefulness of the summary.
6. The computer-implemented method of claim 1 , wherein the set of reviews is hierarchically categorized by the plurality of keywords to result in at least each of the one or more keywords being associated with a subset of the set of reviews.
7. The computer-implemented method of claim 1 , wherein at least one of the plurality of keywords comprises a link which when selected directs to a set of one or more other keywords hierarchically categorized under that keyword.
8. A system for summarizing a set of reviews, comprising:
a processor configured to:
extract a plurality of keywords from a set of reviews that are associated with a product, each review containing content expressing opinions of the product, wherein the plurality of keywords are selected from the content based on a relevance score of tuples of words in the content, wherein the multiple tuples include various permutations of the same root words, and wherein the relevance score is based on the spacing between two words in each tuple in the multiple tuples;
select a subset of reviews from the set of reviews, each review in the subset of reviews containing a keyword from the plurality of keywords;
generate a keyword rating value based on the content belonging to the subset of reviews; and
present a summary of the set of reviews, the summary being configured to graphically display the keyword and the corresponding keyword rating value along with another keyword and another corresponding keyword rating value simultaneously; and
a memory coupled to the processor and configured to provide instructions to the processor.
9. The system of claim 8 , wherein the content includes a reviewer-specified rating value rating the product and wherein generating the keyword rating value further comprises calculating the average or median of the reviewer-specified rating value belonging to each review in the subset of reviews.
10. The system of claim 8 , wherein the tuples of words are segments of the content that do not contain a word from a predefined stop list of words.
11. The system of claim 8 , wherein the relevance score indicates a degree of relevance of the keyword to the set of reviews.
12. The system of claim 8 , wherein a user to whom the summary is presented is provided an opportunity to vote on a usefulness of the summary.
13. The system of claim 8 , wherein the set of reviews is hierarchically categorized by the plurality of keywords to result in at least each of the one or more keywords being associated with a subset of the set of reviews.
14. The system of claim 8 , wherein at least one of the plurality of keywords comprises a link which when selected directs to a set of one or more other keywords hierarchically categorized under that keyword.
15. A computer program product for summarizing a set of reviews, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
extracting a plurality of keywords from a set of reviews that are associated with a product, each review containing content expressing opinions of the product, wherein the plurality of keywords are selected from the content based on a relevance score of tuples of words in the content, wherein the multiple tuples include various permutations of the same root words, and wherein the relevance score is based on the spacing between two words in each tuple in the multiple tuples;
selecting a subset of reviews from the set of reviews, each review in the subset of reviews containing a keyword from the plurality of keywords;
generating a keyword rating value based on the content belonging to the subset of reviews; and
presenting a summary of the set of reviews, the summary being configured to graphically display the keyword and the corresponding keyword rating value along with another keyword and another corresponding keyword rating value simultaneously.
16. The computer program product of claim 15 , wherein the content includes a reviewer-specified rating value rating the product and wherein generating the keyword rating value further comprises calculating the average or median of the reviewer-specified rating value belonging to each review in the subset of reviews.
17. The computer program product of claim 15 , wherein the tuples of words are segments of the content that do not contain a word from a predefined stop list of words.
18. The computer program product of claim 15 , wherein the relevance score indicates a degree of relevance of the keyword to the set of reviews.
19. The computer program product of claim 15 , wherein a user to whom the summary is presented is provided an opportunity to vote on a usefulness of the summary.
20. The computer program product of claim 15 , wherein the set of reviews is hierarchically categorized by the plurality of keywords to result in at least each of the one or more keywords being associated with a subset of the set of reviews.
21. The computer program product of claim 15 , wherein at least one of the plurality of keywords comprises a link which when selected directs to a set of one or more other keywords hierarchically categorized under that keyword.