作者: Jiří Martínek , Ladislav Lenc , Pavel Král
DOI: 10.1007/978-3-030-01418-6_8
关键词:
摘要: This paper proposes a novel approach for multi-lingual multi-label document classification based on neural networks. We use popular convolutional networks this task with three different configurations. The first one uses static word2vec embeddings that are let as is, while the second initializes it and fine-tunes learning available data. last method randomly then they optimized to task. proposed is evaluated four languages, namely English, German, Spanish Italian from Reuters corpus. Experimental results show efficient best obtained F-measure reaches 84%.