Systems and methods for providing a comment-centered news reader

Patent metadata

Field	Value
Publication number	US-11516159-B2
Application number	US-201515578203-A
Country	US
Kind code	B2
Filing date	May 29, 2015
Priority date	May 29, 2015
Publication date	Nov 29, 2022
Grant date	Nov 29, 2022

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.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
receiving, at a computing device, a content item produced by a source system, the content item being accessible to other computing devices via a network;
receiving, at the computing device, a first comment associated with the content item and a second comment associated with the content item, the first comment and the second comment produced by one or more of the other computing devices;
predicting, at the computing device, a first subsection of the content item to link to the first comment based on a first semantic similarity between the content item and the first comment, wherein predicting the first subsection comprises:
comparing pairs comprising the first comment and each sentence of the content item; and
predicting the sentence of the content item to link with the first comment based at least on the comparison of the pairs;

predicting, at the computing device, a second subsection of the content item to link to the second comment based on a second semantic similarity between the content item and the second comment;
linking the first comment with the first subsection based on the first semantic similarity;
linking the second comment with the second subsection based on the second semantic similarity;
when the first subsection of the content item is presented in a first section of a display device, presenting the first comment in a second section of the of the display device; and
when the second subsection of the content item is presented in the first section of the display device, presenting the second comment in the second section of the display device.

2. The computer-implemented method of claim 1 , wherein comparing the identified pairs comprises:
identifying lexical-level features for the first comment and the sentence of each pair;
identifying entity-level features for the first comment a and the sentence of each pair;
identifying topic vectors for the first comment and the sentence of each pair;
creating feature vectors based at least on the identified lexical-level features, entity-level features and topic vectors; and
comparing the created feature vectors.

3. The computer-implemented method of claim 2 , wherein comparing the created feature vectors comprises:
analyzing the created feature vectors using a classifier trained based on a sample set of comments and sentences from a plurality of differently sourced content items.

4. The computer-implemented method of claim 3 , wherein the classifier comprises:
a plurality of feature vectors created from:
lexical-level features for the first comment and the sentence of each pair;
entity-level features for the first comment and the sentence of each pair;
topic vectors for the first comment and the sentence of each pair; and
feature vectors based at least on the identified lexical-level features,

wherein each of the plurality of feature vectors of the classifier comprises a weight value based on manually assigned link value information.

5. The computer-implemented method of claim 4 , further comprising generating the classifier using a support vector machine.

6. The computer-implemented method of claim 1 , further comprising:
receiving a request for the content item from at least one of the other computing devices; and
transmitting information associated with the first subsection of the content item to the at least one of the other computing devices initiating the request for the content item.

7. The computer-implemented method of claim 1 , further comprising:
receiving a request for the content item from at least one of the other computing devices;
transmitting the requested content item to the at least one of the other computing devices; and
transmitting information associated with the first subsection to the at least one of the other computing devices initiating the request for the content item.

8. The computer-implemented method of claim 7 , wherein the requested content item comprises a website including an article.

9. The computer-implemented method of claim 1 , wherein presenting the first comment in the second section of the of the display device comprises scrolling the first comment horizontally or vertically when the size of the first comment exceeds a threshold.

10. A computing device, comprising:
a processor;
a display device; and
a computer-readable storage medium in communication with the processor, the computer-readable storage medium having computer-executable instructions stored thereupon that, when executed by the processor, perform operations, comprising:
receiving information associated with a content item produced by a source system, the content item being accessible to other computing devices via a network;
receiving a plurality of comments associated with the content item, the plurality of comments produced by one or more of the other computing devices;
based on the information and the plurality of comments, predicting at least two subsections of the content item to link to at least two of the plurality of comments based, at least in part, on semantic similarity between the content item and the plurality of comments, wherein predicting the at least two subsections comprises:
comparing pairs comprising at least one of the plurality of comments and each sentence of the content item; and
predicting the sentence of the content item to link with the at least one of the plurality of comments based at least on the comparison of the pairs;

linking the at least two of the plurality of comments with one or more of the at least two predicted subsections based on the semantic similarity;
presenting a first one of the at least two of the plurality of comments in a first section of the display device while presenting a first one of the at least two predicted subsections of the content item in a second section of the display device; and
scrolling the content item to present a second one of the at least two predicted subsections of the content item in the second section of the display device; and
presenting a second one of the at least two of the plurality of comments in the first section of the display device while presenting the second one of the at least two predicted subsections of the content item in the second section of the display device.

11. The computing device of claim 10 , wherein comparing the identified pairs comprises:
identifying lexical-level features for the at least one of the plurality of comments and the sentence of each pair;
identifying entity-level featuresfor the at least one of the plurality of comments and the sentence of each pair;
identifying topic vectors for the at least one of the plurality of comments and the sentence of each pair;
creating feature vectors based at least on the identified lexical-level features, entity-level features and topic vectors; and
comparing the created feature vectors.

12. The computing device of claim 11 , wherein comparing the created feature vectors comprises:
analyzing the created feature vectors using a classifier trained based on a sample set of comments and sentences form a plurality of differently sourced content items.

13. The computing device of claim 12 , wherein the classifier comprises:
a plurality of feature vectors created from:
lexical-level features for the at least one of the plurality of comments and the sentence of each pair;
entity-level features for the at least one of the plurality of comments and the sentence of each pair; topic vectors for the comment and the sentence of each pair; and
feature vectors based at least on the identified lexical-level features,

wherein each o

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Wei Furu
• Zhou Ming
• Liu Yang
• Cao Ziqiang
• Huang Shaohan
• Dong Li
• Cui Lei

Classifications

• G06F40/131

  Fragmentation of text files, e.g. creating reusable text-blocks; Linking to fragments, e.g. using XInclude; Namespaces · CPC title

• G06F40/295

  Named entity recognition · CPC title

• G06F40/134

  Hyperlinking · CPC title

• G06F40/289

  Phrasal analysis, e.g. finite state techniques or chunking · CPC title

• G06F40/284

  Lexical analysis, e.g. tokenisation or collocates · CPC title