融合情感的異構(gòu)圖神經(jīng)網(wǎng)絡(luò)音樂會話推薦算法

打開文本圖片集

本文引用格式：，.融合情感的異構(gòu)圖神經(jīng)網(wǎng)絡(luò)音樂會話推薦算法[J].自動化與信息工程，2025，46（3）：9-16

LU Zhenye， DU Yuxiao. Emotion-enhanced heterogeneous graph neural network for music session-based recommendation algorithm[J]. Automation& Information Engineering，2025，46（3）：9-16.

關(guān)鍵詞：會話推薦；異構(gòu)圖神經(jīng)網(wǎng)絡(luò)；音樂情感；匿名用戶推薦中圖分類號：TP391.3 文獻(xiàn)標(biāo)志碼：A 文章編號：1674-2605（2025）03-0002-08DOI： 10.12475/aie.20250302 開放獲取

Emotion-enhanced Heterogeneous Graph Neural Network for Music Session-based Recommendation Algorithm

LU Zhenye DU Yuxiao （School of Automation， Guangdong University of Technology， Guangzhou 510oo6， China）

Abstract： To addressthe limitations ofcurrent music session-based recommendation methods foranonymous or new users such a simplisticrecommendations based solelyon short-term sesions and neglect ofemotional factors influencinguserchoices this studyproposes anemotion-enhanced heterogeneous graph neuralnetwork for music sesson-basedrecommendationalgorithm. Thealgorithmconstructsasession-basedrecommendationsystemusinghistoricaldatafromallusersandcurentsessionsviagraph neural networks，integratingmusical emotional semantics toprovide more acuraterecommendationsforanonymous/nwusers. ExperimentalresultsontheNowplayingdatasetdemonstrate thatcomparedtothesuboptimal GNN-basedesionrecommedation method， the proposed algorithm achieves a 2.1% improvement in P@20 and a 6.8% increase in MRR @20 ， effectively enhancing recommendation performance.

Keywords： sesion-based recommendation; heterogeneous graph neural network; music emotion; anonymous user recommendation

0 引言

信息技術(shù)的飛速發(fā)展，如5G、智能手機(jī)、云服務(wù)等，為音樂傳播帶來了前所未有的機(jī)遇。（剩余10234字）