Recommendation method fusing content awareness and feature similarity

A feature similarity and recommendation method technology, applied in special data processing applications, instruments, electrical digital data processing, etc., can solve problems such as cold start, unable to recommend new songs without historical records, and unknown content

Active Publication Date: 2021-11-05
湖南工商大学
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This type of method is unknown to the content of the song, and cannot recommend new songs without historical records, and there is a technical problem of cold ...

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
  • Recommendation method fusing content awareness and feature similarity
  • Recommendation method fusing content awareness and feature similarity
  • Recommendation method fusing content awareness and feature similarity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] like figure 1 As shown, a recommendation method that combines content awareness and feature similarity includes the following steps:

[0045] Step 1. Input the original audio into the music analysis library essentia, and the music analysis library essentia outputs features fun (interesting), passionate (excited), sad (depressed), happy (joyful), average loudness (average loudness), electronic (electronic ), timbre and pitch;

[0046] Among them, the music analysis library essentia is an open source C++ library for audio analysis and audio-based music information retrieval, which contains a large number of reusable algorithms that can implement audio input / output functions, standard digital signal processing blocks, statistical characteristics of data, and A large number of spectrum, time, pitch and advanced music descriptors, the library is included in Python, which belongs to the prior art, and will not be repeated here;

[0047] Step 2. Perform principal component a...

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

A recommendation method fusing content awareness and feature similarity relates to the cross technical field of recommendation systems and distributed computing, and comprises the following steps: firstly, decomposing playing frequency data by using a weighted matrix decomposition model: decomposing the playing frequency data into a product of a transpose matrix WT of a user preference matrix and an item (namely music) attribute matrix H; introducing content information into weighted matrix decomposition in consideration of the characteristics of a project (music): adopting a mode of rewriting a project attribute matrix H, specifically, rewriting the project attribute matrix H by adopting a content characteristic zi to obtain a rewritten project attribute matrix Hi; obtaining a prediction matrix A (WT * Hi) by adopting the product of a transpose matrix WT of the user preference matrix and the rewritten project attribute matrix Hi; performing cosine similarity calculation on the timbre and pitch of the item (music) to obtain a prediction matrix B; and fusing the obtained prediction matrix A and the prediction matrix B by using a fusion factor to obtain a recommended value F (mu, i), of which the specific expression is F (mu, i) = theta A (mu, i) + (1-theta) B (mu, i). According to the invention, the cold start problem in the recommendation process can be effectively solved.

Description

technical field [0001] The invention relates to the cross technical field of recommendation system and distributed computing, in particular to a recommendation method integrating content perception and feature similarity. Background technique [0002] The current mainstream music recommendation system is based on collaborative filtering, which predicts the user's interest based on the user's listening habits and similarity with other user profiles. This type of method is unknown to the content of the song, and cannot recommend new songs without historical records, and there is a technical problem of cold start. In order to solve the above technical problem, the present invention provides a recommendation method that combines content perception and feature similarity. Contents of the invention [0003] The purpose of the present invention is to provide a recommendation method that integrates content perception and feature similarity in order to solve the technical problems ...

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
IPC IPC(8): G06F16/68G06F16/635
CPCG06F16/686G06F16/635Y02D10/00
Inventor 徐雪松陈晓红李想胡春华梁伟
Owner 湖南工商大学
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