Reptile-based music tagging method and system

A labeling and music technology, applied in the field of crawler-based music labeling method and system, to achieve good usability, improve usability, and improve utilization

Active Publication Date: 2019-01-29
SOUTH CHINA UNIV OF TECH
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in music labeling, there is no comprehensive, efficient and high-quality technical solution for labeling music library

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
  • Reptile-based music tagging method and system
  • Reptile-based music tagging method and system
  • Reptile-based music tagging method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] See Figure 1 to Figure 4 , figure 1 It is a flow chart of the reptile-based music labeling method disclosed in the first embodiment and the corresponding steps. Such as Figure 1 to Figure 4 As shown, the method includes the following steps:

[0063] S1. Perform preprocessing for crawler music label records from different music websites;

[0064] S2, converting the song list tag into a song tag;

[0065] Wherein said step S2 in song list label transformation step specifically is:

[0066] S21. For each subcategory label, count its frequency of occurrence in all song list label records;

[0067] S22. For each song, count the subcategory tags it contains and the corresponding frequencies;

[0068] S23. For each song, calculate the credible value of each song list tag;

[0069] S24. Screen out the song list tag with a higher credibility value of the song as the song tag of the song;

[0070] S25. Determine whether there are other songs, if so, execute step S23, ot...

Embodiment 2

[0109] There is a kind of crawler-based music labeling system corresponding to the above-mentioned method, and the system includes:

[0110] The data preprocessing module is used to preprocess the crawler music label records from different music websites to make them have a unified format; including removing the irrelevant characters, unifying the connectors of multiple singers into plus signs, etc.

[0111] The label conversion module is used to convert the song list label into a song label through the calculation of the song list label credibility value. Specifically: the song list label of a song is converted into the song label, and the label conversion module is used to count the frequency of occurrence of each of the subcategory labels in all the song list label records, In addition, for a song, count the subcategory tags it contains and the frequency corresponding to the tag, calculate the credible value of each song list tag that the song contains, and filter out the o...

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 crawler-based music labelling method and a crawler-based music labelling system. According to the method and the system, the availability of music label records is improved by pre-processing the music label records on the basis of crawling the music label records of a plurality of music websites by a crawler; meanwhile, a uniform format of the music label records from different music websites is guaranteed; moreover, song menu labels are converted into song labels by calculating credit values of the song menu labels; fuzzy matching is performed on the song names in the music label records crawled by the crawler and the song names in an unlabeled song bank by utilizing an edit distance algorithm, and a corresponding relation is established, so that the utilization rate of the crawler music label records is improved; more music labels are labelled on the song bank to the maximal extent on the premise of guaranteeing label accuracy; meanwhile, a set of complete music label system is established according to the music labels of the music websites, and labels are fused on the basis of the labels of the music label system.

Description

technical field [0001] The invention belongs to the technical field of data acquisition and reliability calculation, and more specifically relates to a crawler-based music labeling method and system. Background technique [0002] With the continuous expansion of the digital music market, the characteristic annotation of digital music has become a hot spot in the research at this stage, and the music label is the most important characteristic annotation. Music tags have rich semantics, but the premise of this tagging method is that there is an active music community with a large number of highly sticky users; on the other hand, music tags are random and discrete, which is easy to form a "cold start" (Cold Start), the newly released music will be difficult to be retrieved by users due to the lack of labels, resulting in "Popularity Bias", affected by the first labeling of users, the user's choice will follow the previous popular trend, thus making Most users only listen to a ...

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): G06F16/9535G06F16/635
CPCG06F16/635G06F16/9535
Inventor 郭泽豪王振宇戴瑾如
Owner SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products