Systems and methods for providing a comment-centered news reader

Abstract

Methods and systems for linking comments to portions of content items. An example computing device receives information associated with a content item produced by a source system, the content item being accessible to other the computing devices via a network and receives a comment associated with the content item, the comment produced by one of the other computing devices. In response to receiving the information and the comment, the computing device predicts a subsection of the content item to link to the received comment based at least on details associated with the content item and the comment, then makes information associated with the predicted subsection of the content item available to other computing devices requesting access to the content item.

1 .- 20 . (canceled) 
     
     
         21 . A client device comprising:
 a processor;   memory storing instructions that, when executed, cause the client device to perform operations comprising:
 receiving a content item; 
 detecting a comment associated with the content item; 
 comparing the comment to each sentence of the content item; 
 identifying a sentence of the content item to link to the comment based on a semantic similarity between the sentence and the comment; and 
 linking the sentence to the comment such that the comment is presented in a first section of a display area when the sentence is presented in a second section of the display area. 
   
     
     
         22 . The client device of  claim 21 , wherein:
 the first section is a content presentation portion of the client device; and   the second section is a comment presentation portion of the client device, the first section and the second section being separate sections of the client device.   
     
     
         23 . The client device of  claim 21 , wherein the client device further comprises a third section that is a comment entry portion, the third section being separate from the first section and the second section. 
     
     
         24 . The client device of  claim 23 , wherein detecting the comment comprises receiving the comment into the third section while the content item is displayed in the first section. 
     
     
         25 . The client device of  claim 23 , the operations further comprising:
 navigating the content item to display the sentence in the first section;   in response to navigating the content item to display the sentence, displaying the comment in the second section;   navigating the content item such that the sentence is no longer displayed in the first section; and   in response to navigating the content item such that the sentence is no longer displayed, removing the comment from the second section.   
     
     
         26 . The client device of  claim 21 , wherein:
 the content item is accessible to other client devices; and   detecting the comment comprises detecting that the comment has been submitted by one of the other client devices.   
     
     
         27 . The client device of  claim 21 , wherein comparing the comment to each sentence of the content item comprises:
 for each comment/sentence pair:
 identifying at least one of words, word stems, function words, or parts of speech for the comment/sentence pair; 
 identifying name entities for the comment/sentence pair; and 
 determining a topic vector for the comment/sentence pair. 
   
     
     
         28 . The client device of  claim 27 , wherein comparing the comment to each sentence of the content item further comprises:
 for each comment/sentence pair:
 extracting features from the topic vector; and 
 creating one or more feature vectors for the comment/sentence pair. 
   
     
     
         29 . The client device of  claim 28 , wherein identifying the sentence of the content item to link to the comment comprises:
 comparing a feature vector for the comment to a feature vector for each sentence; and   identifying a feature vector for the sentence is a best match for feature vector for the comment based on semantic similarities between the feature vector for the sentence and the feature vector for the comment.   
     
     
         30 . The client device of  claim 21 , wherein:
 when the comment is presented in the first section, the sentence is highlighted or otherwise emphasized in the second section.   
     
     
         31 . The client device of  claim 21 , wherein:
 when the comment exceeds a size of the first section, the comment is scrolled at least one of horizontally or vertically to enable an entirety of the comment to be viewed.   
     
     
         32 . A server device comprising:
 a processor;   memory storing instructions that, when executed, cause the server device to perform operations comprising:
 detecting a comment associated with a content item that is accessible to one or more client devices; 
 comparing the comment to each sentence of the content item; and 
 linking a sentence of the content item to the comment based on a semantic similarity between the sentence and the comment, wherein the linking causes the comment to be presented in a first section of a display area of the one or more client devices when the sentence is presented in a second section of the display area. 
   
     
     
         33 . The server device of  claim 32 , wherein comparing the comment to each sentence of the content item comprises:
 identifying at least one of lexical-level features, entity-level, or topic vectors for first comment and each sentence;   creating feature vectors based on the at least one of the lexical-level features, the entity-level, or the topic vectors; and   comparing the feature vectors.   
     
     
         34 . The server device of  claim 33 , wherein creating the feature vectors comprises:
 analyzing the feature vectors using a classifier trained based on a sample set of comments and sentences from a plurality of content items.   
     
     
         35 . The server device of  claim 34 , wherein each of the feature vectors comprises a weight value based on link value information. 
     
     
         36 . The server device of  claim 33 , wherein the lexical-level features include a cosine similarity between the sentence and the comment. 
     
     
         37 . The server device of  claim 32 , wherein detecting the comment comprises:
 receiving, by a linking module of the server device, the comment and information identifying a subsection of the content item that was a target of the comment, the comment and information being received from the one or more client devices.   
     
     
         38 . The server device of  claim 37 , wherein the linking module provides to the one or more client devices an indication that the sentence is linked to the content item. 
     
     
         39 . The server device of  claim 37 , wherein the linking module stores pairs of linked sentences and comments for future use. 
     
     
         40 . A method comprising:
 detecting a comment has been added to a content item;   comparing the comment to each sentence of the content item;   identifying a section of the content item to link to the comment based on a semantic similarity between one or more sentences and the comment; and   linking the section to the comment such that the comment is presented in a first portion of a display area when at least a portion of the section is presented in a second section of the display area.