Quotations-modules on online social networks
In one embodiment, a method includes accessing a plurality of communications, each communication being associated with a particular content item and including a text of the communication; extracting, for each of the communications, quotations from the text of the communication; determining, for each extracted quotation, partitions of the quotation; grouping the extracted quotations into clusters based on a respective degree of similarity among their respective partitions; calculating a cluster-score for each cluster based on a frequency of occurrence of partitions of quotations in the cluster in the communications; and generating a quotations-module comprising representative quotations, each representative quotation being a quotation from a cluster having a cluster-score greater than a threshold cluster-score.
1. A method comprising, by one or more computing devices:
accessing, by the one or more computing devices, a plurality of communications authored by one or more users of an online social network, each communication being associated with a particular content item and comprising a text of the communication;
extracting, for each of the plurality of communications, one or more quotations from the text of the communication;
determining, for each extracted quotation, one or more partitions of the quotation;
grouping, by the one or more computing devices, the extracted quotations into one or more clusters based on a respective degree of similarity among their respective one or more partitions;
calculating, by the one or more computing devices, a cluster-score for each cluster based on a frequency of occurrence of one or more partitions of one or more quotations in the cluster in communications associated with the particular content item;
calculating, by the one or more computing devices, a quotation-score for each quotation in a cluster having a cluster-score greater than a threshold cluster-score, wherein the quotation-score is calculated based on the following expression:
q+Σ i=0 n (α×similarity metric), wherein q is a count of occurrences of quotations that are exact matches and α is a respective count of occurrences of quotations that are not exact matches in a total set of unique extracted quotations defined from i=0 to i=n, and wherein the similarity metric is a measure of a degree of similarity between a quotation that is an exact match and the respective quotation that is not an exact match; and
generating, by the one or more computing devices, a quotations-module comprising one or more representative quotations, each representative quotation being a quotation from a cluster having a cluster-score greater than a threshold cluster-score, and each representative quotation having a quotation-score greater than a threshold quotation-score.
2. The method of claim 1 , further comprising:
accessing a social graph comprising a plurality of nodes and a plurality of edges connecting the nodes, each of the edges between two of the nodes representing a single degree of separation between them, the nodes comprising:
a first node corresponding to a first user associated with the online social network; and
a plurality of second nodes corresponding to a plurality of communications of the online social network.
3. The method of claim 1 , wherein the each communication is associated with the particular content item if it includes a link to an instance of the particular content item.
4. The method of claim 1 , wherein the extracting of the one or more quotations comprises:
comparing the text of the communication with text associated with the particular content item; and
extracting one or more text strings from the text of the communication that match one or more text strings associated with the particular content item.
5. The method of claim 1 , wherein the extracting of the one or more quotations comprises extracting one or more text strings demarcated by a quotation-indicator.
6. The method of claim 5 , wherein the quotation-indicator comprises a set of quotation marks.
7. The method of claim 1 , wherein determining the one or more partitions of each extracted quotation comprises partitioning the extracted quotation into one or more sentences.
8. The method of claim 1 , wherein determining the one or more partitions of each extracted quotation comprises partitioning the extracted quotation into one or more clauses.
9. The method of claim 1 , wherein the degree of similarity is determined based on a partition-similarity metric between the partitions, wherein the partition-similarity metric is based on a Levenshtein distance between the partitions.
10. The method of claim 1 , further comprising sending, to a client system of a first user of the online social network, the quotations-module for display to the first user.
11. The method of claim 10 , wherein the cluster-score is further based on demographic information associated with the first user.
12. The method of claim 10 , wherein the cluster-score is further based on a language associated with the user.
13. The method of claim 10 , wherein the cluster-score is further based on a user preference associated with the first user.
14. The method of claim 10 , wherein the cluster-score is further based on an affinity coefficient between the first user and an author of a communication from which the quotation was extracted.
15. The method of claim 1 , wherein the quotation-score is further based on one or more social signals associated with communications including the quotation.
16. The method of claim 1 , wherein the quotation-score is further based on a quotation quality of the quotation.
17. The method of claim 1 , wherein the quotation-score is further based on demographic information associated with the first user and demographic information associated with an author of a communication from which the quotation was extracted.
18. The method of claim 1 , wherein the similarity metric is a value between 0 and 1.
19. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
access a plurality of communications authored by one or more users of an online social network, each communication being associated with a particular content item and comprising a text of the communication;
extract, for each of the plurality of communications, one or more quotations from the text of the communication;
determine, for each extracted quotation, one or more partitions of the quotation;
group the extracted quotations into one or more clusters based on a respective degree of similarity among their respective one or more partitions;
calculate a cluster-score for each cluster based on a frequency of occurrence of one or more partitions of one or more quotations in the cluster in communications associated with the particular content item;
calculate a quotation-score for each quotation in a cluster having a cluster-score greater than a threshold cluster-score, wherein the quotation-score is calculated based on the following expression:
q+Σ i=0 n (α×similarity metric), wherein q is a count of occurrences of quotations that are exact matches and α is a respective count of occurrences of quotations that are not exact matches in a total set of unique extracted quotations defined from i=0 to i=n, and wherein the similarity metric is a measure of a degree of similarity between a quotation that is an exact match and the respective quotation that is not an exact match; and
generate a quotations-module comprising one or more representative quotations, each representative quotation being a quotation from a cluster having a cluster-score greater than a threshold cluster-score, and each representative quotation having a quotation-score greater than a threshold quotation-score.
20. A system comprising: one or more processors; and a non-transitory memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to:
access a plurality of communications authored by one or more users of an online social network, each communication being associated with a particular content item and comprising a text of the communication;
extract, for each of the plurality of communications, one or more quotations from the text of the communication;
determine, for each extracted quotation, one or more partitions of the quotation;
group the extracted quotations into one or more clusters based on a respective degree of similarity among their respective one or more partitions;
calculate a cluster-score for each cluster based on a frequency of occurrence of one or more partitions of one or more quotations in the cluster in communications associated with the particular content item;
calculate a quotation-score for each quotation in a cluster having a cluster-score greater than a threshold cluster-score, wherein the quotation-score is calculated based on the following expression:
q+Σ i=0 n (α×similarity metric), wherein q is a count of occurrences of quotations that are exact matches and α is a respective count of occurrences of quotations that are not exact matches in a total set of unique extracted quotations defined from i=0 to i=n, and wherein the similarity metric is a measure of a degree of similarity between a quotation that is an exact match and the respective quotation that is not an exact match; and
generate a quotations-module comprising one or more representative quotations, each representative quotation being a quotation from a cluster having a cluster-score greater than a threshold cluster-score, and each representative quotation having a quotation-score greater than a threshold quotation-score.