Unlock instant, AI-driven research and patent intelligence for your innovation.

Query error-correction method and device in video searching

A technology of video search and error correction method, which is applied in the field of video search, can solve the problems of large error correction, a large amount of manual work, and the hysteresis of the range of action of offline methods, so as to achieve the effect of improving accuracy

Active Publication Date: 2017-10-27
BEIJING QIYI CENTURY SCI & TECH CO LTD
View PDF5 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These relationships are either defined manually or come from mining user logs. Although some wrong inputs can be covered, this kind of method requires a lot of manual work on the one hand, and on the other hand, the scope of this offline method is hysteresis: The user may only be found and put into the error correction dictionary if the user has entered
However, due to the diversity of user input, many error modes are completely unpredictable
However, the existing online methods have too simple error correction criteria for query words, and the probability of error correction is relatively high.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Query error-correction method and device in video searching
  • Query error-correction method and device in video searching
  • Query error-correction method and device in video searching

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] This embodiment provides a query word error correction method in video search, please refer to figure 1 , the method includes:

[0049] Step S11: receiving the query word input by the user;

[0050] Step S12: Perform word deletion processing on the query words to obtain a word string set;

[0051] The rules for word deletion processing are set in advance. The word deletion processing can be to delete each word of the query word to obtain a string set, or to delete multiple words of the query word to obtain a word string set, or to delete each word of the query word to obtain a string set. Each word is deleted separately and multiple words are deleted respectively to obtain a string set.

[0052] For example, the query word (query) input by the user is "Legend of Heroes of Shooting Mink". Delete each word of the query word to obtain the string set as follows: delete the first word of the query word to get "The Legend of Diao Heroes"; delete the second word of the quer...

Embodiment 2

[0088] This embodiment provides a query word error correction device in video search, please refer to Figure 4 , the device consists of:

[0089] A receiving unit 11, configured to receive a query word input by a user;

[0090]The processing unit 12 is used to perform word-deletion processing on the query word to obtain a word string set;

[0091] The rules for word deletion processing are set in advance. The word deletion processing can be to delete each word of the query word to obtain a string set, or to delete multiple words of the query word to obtain a word string set, or to delete each word of the query word to obtain a string set. Each word is deleted separately and multiple words are deleted respectively to obtain a string set.

[0092] The matching unit 13 is used to use each word in the word string set and the query word to match the error correction candidate word dictionary obtained in advance to obtain the error correction candidate word;

[0093] The error c...

Embodiment 3

[0105] This embodiment provides another query word error correction device in video search, please refer to Figure 6 , the device is relatively Figure 4 Disclosed devices also include:

[0106] The dictionary generation unit 15 is used to receive the words input by the user, use the words as error correction candidates, perform word deletion processing on the words, obtain a set of word strings after word deletion, and establish the words and words , the corresponding relationship between the word string sets after the word deletion process.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a query error-correction method and device of in video searching. Character deletion processing is carried out on a query to obtain a character string set after character deletion; and each word in the character string set and the query are utilized to carry out matching on error-correction candidate words in a pre-obtained error-correction candidate word dictionary. Typical error modes such as word addition, word reduction, a word mistake and disorder are covered through such a simple strategy, and an accuracy rate of online error correction of the query is increased.

Description

technical field [0001] The present invention relates to the technical field of video search, and more specifically, to a query word error correction method and device in video search. Background technique [0002] Finding the user's wrong input and correcting it to a correct input that meets the user's real intention is a key step for the search engine to preprocess the query. Search engines need to recall related documents (doc) for a specific query. In practical applications, errors contained in the query, such as typos, word additions, disordered order, etc., may cause the search engine to recall a large number of docs that do not match the user's true intentions. When the query is corrected, on the one hand, the user's true intention is restored, which is the embodiment of the intelligence of the search engine; on the other hand, it is an important supplement to the retrieval and sorting algorithms, reducing the difficulty of retrieval. Statistics show that more than 1...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/30
CPCG06F16/374G06F16/73G06F16/90344
Inventor 孙超博
Owner BEIJING QIYI CENTURY SCI & TECH CO LTD