Abstract: In one embodiment, a method includes retrieving comments associated with a content object on a social-networking system, determining a score for each of the comments, where the score is based on one or more signals associated with the comment, the signals are related to one or more audience interactions with the comment, and each of the audience interactions includes an interaction by a respective audience member with the comment. The method further includes ordering the comments based on the respective scores, and presenting the ordered comments to a target user. Each of the audience interactions may include one or more responses to the audience interaction by an author of the comment. The signals may be based on existence of an edge between a user node representing the audience member and a user node representing an author of the comment in a social graph on the social-networking system.

Abstract: In one embodiment, a method includes retrieving a set of comments associated with one or more content objects in a social-networking system, determining a score for each of the comments for a selected one of the content objects, where the score for each of the comments is based on signals associated with the comment, and the signals are related to a history of user interactions by a target user with the comments in the set, ordering the comments for the selected content object based on the respective scores, and presenting the ordered comments to the target user. The content objects may be presented in a newsfeed for the target user. The signals may be based on a history of user interactions by other users with the comments in the set, where the other users are connected to the target user in the social-networking system.

Abstract: In one embodiment, a method includes retrieving a plurality of comments associated with a content object on a social-networking system, determining a score for each of the comments, wherein the score is based on one or more signals associated with the comment, and where the signals are related to the identity of an author of the comment or content of the comment, ordering the comments based on the respective scores, and presenting one or more of the ordered comments to a target user. The method may further include excluding one or more of the comments based on a filtering condition. One of the signals may be based on how many times the comment has been (a) liked, (b) hidden, (c) marked as spam, or (d) replied to within a specified period of time. One of the signals may be based on a reputation of the author of the comment.

Abstract: In one embodiment, a method includes retrieving a plurality of comments associated with a content object, determining a score for each of the comments, where the score is based on one or more signals associated with the comment, and where the score is further based on a qualitative analysis of each of one or more impressions related to the comment, and wherein each of the impressions comprises an interaction by a historical user with the comment, ordering the comments based on the respective scores, presenting one or more of the ordered comments to a target user. The qualitative analysis may determine a category of interaction for the impression, the score may be based at least in part on the category of interaction, and the category of interaction may include user input related to the comment, such as a reply to the comment.

Abstract: In one embodiment, a method includes retrieving comments associated with a content object, determining one or more labels for each of the comments, where the labels for each comment are determined by a text classifier based on content associated with the comment, determining a score for each of the comments, where the score is based on one or more signals associated with the comment, and the signals are related to the labels associated with the comment. The method further includes ordering the comments based on the respective scores and presenting one or more of the ordered comments to a target user. A classifier algorithm may be applied to each comment to determine the one or more labels, where the classifier algorithm is trained to identify the one or more labels. Each signal may have a value based on a degree to which the label applies to the comment.

Abstract: In one embodiment, a method includes retrieving a plurality of comments associated with a content object on a social-networking system, determining a score for each of the comments, wherein the score is based on one or more signals associated with the comment, and where the signals are related to the identity of an author of the comment or content of the comment, ordering the comments based on the respective scores, and presenting one or more of the ordered comments to a target user. The method may further include excluding one or more of the comments based on a filtering condition. One of the signals may be based on how many times the comment has been (a) liked, (b) hidden, (c) marked as spam, or (d) replied to within a specified period of time. One of the signals may be based on a reputation of the author of the comment.

Abstract: In one embodiment, a method includes retrieving a plurality of comments associated with a content object, determining a score for each of the comments, where the score is based on one or more signals associated with the comment, and where the score is further based on a qualitative analysis of each of one or more impressions related to the comment, and wherein each of the impressions comprises an interaction by a historical user with the comment, ordering the comments based on the respective scores, presenting one or more of the ordered comments to a target user. The qualitative analysis may determine a category of interaction for the impression, the score may be based at least in part on the category of interaction, and the category of interaction may include user input related to the comment, such as a reply to the comment.

Abstract: In one embodiment, a method includes retrieving a set of comments associated with one or more content objects in a social-networking system, determining a score for each of the comments for a selected one of the content objects, where the score for each of the comments is based on signals associated with the comment, and the signals are related to a history of user interactions by a target user with the comments in the set, ordering the comments for the selected content object based on the respective scores, and presenting the ordered comments to the target user. The content objects may be presented in a newsfeed for the target user. The signals may be based on a history of user interactions by other users with the comments in the set, where the other users are connected to the target user in the social-networking system.

Abstract: In one embodiment, a method includes retrieving comments associated with a content object on a social-networking system, determining a score for each of the comments, where the score is based on one or more signals associated with the comment, the signals are related to one or more audience interactions with the comment, and each of the audience interactions includes an interaction by a respective audience member with the comment. The method further includes ordering the comments based on the respective scores, and presenting the ordered comments to a target user. Each of the audience interactions may include one or more responses to the audience interaction by an author of the comment. The signals may be based on existence of an edge between a user node representing the audience member and a user node representing an author of the comment in a social graph on the social-networking system.