Method and apparatus for presenting information

Patent metadata

Field	Value
Publication number	US-11651015-B2
Application number	US-201916670814-A
Country	US
Kind code	B2
Filing date	Oct 31, 2019
Priority date	Jan 8, 2019
Publication date	May 16, 2023
Grant date	May 16, 2023

Abstract

Embodiments of the present disclosure provide a method and apparatus for presenting information. The method may include: acquiring target release information and a comment information set associated with the target release information; and generating, for comment information in the comment information set, usefulness probabilities and predicted comment scores of the comment information based on the comment information and the target release information. The method may further include: presenting, based on obtained usefulness probability set and predicted comment score set, the comment information in the comment information set.

First claim

What is claimed is:

1. A method for presenting information, comprising:
acquiring target release information and a comment information set associated with the target release information;
generating, for comment information in the comment information set, usefulness probabilities and predicted comment scores of the comment information based on the comment information and the target release information; and
presenting, based on obtained usefulness probability set and predicted comment score set, the comment information in the comment information set,
wherein the generating usefulness probabilities and predicted comment scores of the comment information based on the comment information and the target release information comprises:
extracting word vector codes and character vector codes from the comment information, and generating initial comment codes based on the extracted word vector codes and the character vector codes;
extracting word vector codes and character vector codes from the target release information, and generating initial release codes based on the extracted word vector codes and the character vector codes;
inputting the initial comment codes into a first bidirectional long and short-term memory recurrent neural network to obtain comment codes;
inputting the initial release codes into a second bidirectional long and short-term memory recurrent neural network to obtain release codes;
generating attention mechanism codes based on the release codes and the comment codes;

inputting the attention mechanism codes into a logistic regression model to obtain the usefulness probabilities; and
inputting the attention mechanism codes into a linear regression model to obtain the predicted comment scores.

2. The method according to claim 1 , wherein the presenting, based on the obtained usefulness probability set and predicted comment score set, the comment information in the comment information set comprises:
presenting the comment information in the comment information set in a descending order of corresponding predicted comment scores.

3. The method according to claim 1 , wherein the presenting, based on the obtained usefulness probability set and predicted comment score set, the comment information in the comment information set comprises:
hiding or folding the comment information corresponding to usefulness probabilities smaller than a preset threshold.

4. An apparatus for presenting information, comprising:
at least one processor; and
a memory storing instructions, wherein the instructions when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising:
acquiring target release information and a comment information set associated with the target release information;
generating, for comment information in the comment information set, usefulness probabilities and predicted comment scores of the comment information based on the comment information and the target release information; and
presenting, based on obtained usefulness probability set and predicted comment score set, the comment information in the comment information set,
wherein the generating usefulness probabilities and predicted comment scores of the comment information based on the comment information and the target release information comprises:
extracting word vector codes and character vector codes from the comment information, and generating initial comment codes based on the extracted word vector codes and the character vector codes;
extracting word vector codes and character vector codes from the target release information, and generating initial release codes based on the extracted word vector codes and the character vector codes;
inputting the initial comment codes into a first bidirectional long and short-term memory recurrent neural network to obtain comment codes;
inputting the initial release codes into a second bidirectional long and short-term memory recurrent neural network to obtain release codes; generating attention mechanism codes based on the release codes and the comment codes;
inputting the attention mechanism codes into a logistic regression model to obtain the usefulness probabilities; and
inputting the attention mechanism codes into a linear regression model to obtain the predicted comment scores.

5. The apparatus according to claim 4 , wherein the presenting, based on the obtained usefulness probability set and predicted comment score set, the comment information in the comment information set comprises:
presenting the comment information in the comment information set in a descending order of corresponding predicted comment scores.

6. The apparatus according to claim 4 , wherein the presenting, based on the obtained usefulness probability set and predicted comment score set, the comment information in the comment information set comprises:
hiding of folding the comment information corresponding to usefulness probabilities smaller than a preset threshold.

7. A non-transitory computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, causes the processor to perform operations, the operations comprising:
acquiring target release information and a comment information set associated with the target release information;
generating, for comment information in the comment information set, usefulness probabilities and predicted comment scores of the comment information based on the comment information and the target release information; and
presenting, based on obtained usefulness probability set and predicted comment score set, the comment information in the comment information set,
wherein the generating usefulness probabilities and predicted comment scores of the comment information based on the comment information and the target release information comprises:
extracting word vector codes and character vector codes from the comment information, and generating initial comment codes based on the extracted word vector codes and the character vector codes;
extracting word vector codes and character vector codes from the target release information, and generating initial release codes based on the extracted word vector codes and the character vector codes;
inputting the initial comment codes into a first bidirectional long and short-term memory recurrent neural network to obtain comment codes;
inputting the initial release codes into a second bidirectional long and short-term memory recurrent neural network to obtain release codes;
generating attention mechanism codes based on the release codes and the comment codes;

inputting the attention mechanism codes into a logistic regression model to obtain the usefulness probabilities; and
inputting the attention mechanism codes into a linear regression model to obtain the predicted comment scores.

8. The non-transitory computer readable storage medium according to claim 7 , wherein the presenting, based on the obtained usefulness probability set and predicted comment score set, the comment information in the comment information set comprises:
presenting the comment information in the comment information set in a descending order of corresponding predicted comment scores.

9. The non-transitory computer readable storage medium according to claim 7 , wherein the presenting, based on the obtained usefulness probability set and predicted comment score set, the comment information in the comment information set comprises:
hiding or folding the comment information corresponding to usefulness probabilities smaller than a preset threshold by hiding or folding.

Assignees

• Baidu online network technology beijing co ltd

Inventors

• Fan Miao
• Guo Lin
• Feng Chao
• Sun Mingming
• Li Ping
• Wang Haifeng

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06F16/335 Primary

  Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536) · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title