| 基于情感主题的音乐分类研究 |
| |
| 引用本文： | 张宏,阮泽楠.基于情感主题的音乐分类研究[J].教育技术导刊,2019,18(7):15-18. |
| |
| 作者姓名： | 张宏 阮泽楠 |
| |
| 作者单位： | 浙江理工大学 经济管理学院，浙江 杭州 310018 |
| |
| 基金项目： | 国家社会科学基金项目（15BSH107） |
| |
| 摘 要： | 为确定歌词隐含的情感主题对音乐分类的作用，在传统主题模型中融入情感、语义元素，定义基于情感主题的音乐分类标准并进行音乐分类。结合文本情感词典、Word2vec词向量空间，将主题模型的基础主题进一步归类为情感主题，并通过爬取网易云音乐歌曲信息进行模型训练及测试。实验证明，该模型具有较好的分类效果，对音乐情感分类平均准确率达到80%。
|
| 关 键 词： | 音乐分类 情感分析 主题模型 Word2vec LDA |
| 收稿时间： | 2018-11-28 |
Music Classification Research Based on Emotion Topic |
| |
| ZHANG Hong,RUAN Ze-nan.Music Classification Research Based on Emotion Topic[J].Introduction of Educational Technology,2019,18(7):15-18. |
| |
| Authors: | ZHANG Hong RUAN Ze-nan |
| |
| Institution: | School of Economics and Management， Zhejiang Sci-tech University， Hangzhou 310018， China |
| |
| Abstract: | In order to identify the role of the emotional topic implicit in the lyrics on music classification， this paper incorporates emotion and semantic elements into the traditional topic model to define music classification criteria based on emotional topic and classify music. Combining the text sentiment dictionary and the Word2vec， the basic topics in the topic model are further classified into some emotional topics， and the model is trained and tested by crawling the song information of Netease cloud music. The experiment proves that the model has a good effect， and the average accuracy rate of music emotion classification reaches 80%. |
| |
| Keywords: | music classification sentiment analysis topic model Word2vec LDA |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
