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

A hybrid word segmentation tagging method and system based on integrated video meta information

A meta-information tag and meta-information technology, applied in special data processing applications, instruments, electrical digital data processing, etc., can solve problems such as waste, high user requirements, unfavorable video website promotion, etc., and achieve the effect of reducing complexity

Active Publication Date: 2016-03-30
ALIBABA (CHINA) CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

And this is obviously too demanding for the user
In other words, the search engine in the video website of the prior art has led to a threshold requirement for the user, which is not conducive to the promotion of the video website, and to attract more users from all walks of life, and also makes the user waste a lot of time. to search for the video you are interested in

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
  • A hybrid word segmentation tagging method and system based on integrated video meta information
  • A hybrid word segmentation tagging method and system based on integrated video meta information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] see figure 1 , discloses a hybrid word segmentation tagging method for comprehensive video meta-information according to the present invention, comprising the following steps:

[0051] The video meta information collection step S110 is to collect multi-dimensional video meta information related to the video itself, and the video meta information includes, but is not limited to, the attributes of the video itself and / or the business attributes of the website. For example, video meta information can include:

[0052] 1. Whether the video is charged;

[0053] 2. Whether the video is exclusive to members;

[0054] 3. Which advertiser sponsors the video;

[0055] 4. Which activities of the website are related to the video;

[0056] 5. The shooting year, region, director, actors, etc. of the video;

[0057] 6. The category of the video, the style of the video, etc.;

[0058] 7. The recent popularity and popularity of the video;

[0059] 8,…….

[0060] The video meta i...

Embodiment 2

[0084] This embodiment lists a specific video search process.

[0085] In the existing video websites, some users like to search for "paid movies", so the present invention selects the video meta-information whether the movie is charged or not.

[0086] Therefore, when segmenting the basic video information for the engine, whether the video is charged or not and the video category information inherent in the video meta information are obtained from the video meta information library. If a video named "Jack the Giant Slayer" is marked as paid, and the category value of this video represents a movie, then the video meta information of "paid movie" is collected, and then the video meta information is compared with the user query Corresponding to the keywords, a set of video meta information tags containing at least the video meta information tag words of "paid movie" and "paid cinema" is obtained. Then the video title and video meta information tags are word-cut, and the followi...

Embodiment 3

[0092] A hybrid word segmentation tagging system that synthesizes video meta information, including the following units:

[0093] The video meta information collection unit 210 collects video meta information related to the video itself in a multi-dimensional manner, and the video meta information includes the attributes of the video itself and / or the business attributes of the website;

[0094] The video meta information tag generation unit 220 analyzes the semantic relationship between the video meta information and the keywords queried by the user, and generates video meta information tag words according to the video meta information;

[0095] The hybrid word segmentation and tagging unit 230 that fuses video meta information tags, including

[0096] The basic text information segmentation word sub-unit, segments the basic text information related to the video, and the output of the obtained word segmentation is an output sequence;

[0097] The video meta information tag w...

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 discloses a mixed word segmentation and labeling method and system for general video meta-information. The method comprises the steps as follows: a video meta-information collection step, wherein the video meta-information related to videos is collected in a multidimensional manner, and the video meta-information comprises video attributes and / or website business attributes; a video meta-information label generation step, wherein semantic relations between the video meta-information and keywords queried by a user are analyzed, and video meta-information label words are generated according to the video meta-information; and a mixed word segmentation and labeling step integrated with video meta-information labels, wherein basic character information related to the videos is segmented, obtained word segmentation outputs are output sequences, word segmentation is performed on multiple groups of the video meta-information labels respectively, and the outputs are labeled as different output sequences. According to the method and the system, the video meta-information is translated into text labels, the lower weight is given, and the text labels are added into a search engine, so that the search experience of the user is improved in a reverse hitting manner, and the problem of priority reversal is solved.

Description

technical field [0001] The present application relates to a keyword segmenting tagging method for a search engine, in particular, the present invention relates to a hybrid tagging method and a system for combining video meta-information of a video. Background technique [0002] Traditional search engines look for content containing keywords in massive information based on keywords entered by users. Whether the query results are relevant depends on a series of factors, including whether the keywords entered by the user are in the target document, whether the keywords entered by the user can filter out a small enough number of documents to prevent the target document from being submerged in a large number of hit results, and whether the user can Think of what keywords its target document might contain. Therefore, the engine based on keyword hit requires users to understand the working principle of search engines, otherwise it is easy for users to find the content they want. ...

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 Patents(China)
IPC IPC(8): G06F17/30
CPCG06F16/783G06F16/7867
Inventor 刘伟姚键潘柏宇卢述奇
Owner ALIBABA (CHINA) CO LTD