Homepage2Vec: Language-Agnostic Website Embedding and Classification
This addresses the need for better multilingual website classification tools, though it is incremental as it builds on existing natural language representation progress.
The authors tackled the problem of limited multilingual and embedding support in website classification by releasing a dataset of over two million websites in 92 languages and introducing Homepage2Vec, a language-agnostic model that achieves a macro-averaged F1-score of 0.90 for classification.
Currently, publicly available models for website classification do not offer an embedding method and have limited support for languages beyond English. We release a dataset of more than two million category-labeled websites in 92 languages collected from Curlie, the largest multilingual human-edited Web directory. The dataset contains 14 website categories aligned across languages. Alongside it, we introduce Homepage2Vec, a machine-learned pre-trained model for classifying and embedding websites based on their homepage in a language-agnostic way. Homepage2Vec, thanks to its feature set (textual content, metadata tags, and visual attributes) and recent progress in natural language representation, is language-independent by design and generates embedding-based representations. We show that Homepage2Vec correctly classifies websites with a macro-averaged F1-score of 0.90, with stable performance across low- as well as high-resource languages. Feature analysis shows that a small subset of efficiently computable features suffices to achieve high performance even with limited computational resources. We make publicly available the curated Curlie dataset aligned across languages, the pre-trained Homepage2Vec model, and libraries