The ability to understand and process websites is crucial across various domains. It lays the foundation for machine understanding of websites. Specifically, website embedding proves invaluable when monitoring local government websites within the context of digital transformation. In this paper, we present a comparison of different state-of-the-art website embedding methods and their capability of creating a reasonable website embedding for our specific task. The models consist of visual, mixed, and textual-based embedding methods. We compare the models with a baseline model which embeds the header section of a website. We measure the performance of the models using zero-shot and transfer learning. We evaluate the performance of the models on three different datasets. Additionally to the embedding scoring, we evaluate the classification performance on these datasets. From the zero-shot models Homepage2Vec with visual, a combination of visual and textual embedding, performs best in general over all datasets. When applying transfer learning, TF-IDF & FNN, a text based model, outperforms the others in both cluster scoring as well as precision and F1-score in the classification task. However, time is an important factor when it comes to processing large data quantities. Thus, when additionally considering the time needed, our baseline model is a good alternative, being 1.88 times faster with a maximum decrease of 10 % in the F1-score.

Webpage segmentation is a crucial task in web analysis, enabling improvements in information retrieval, user experience, and automated web understanding.However, existing segmentation datasets often lack both comprehensive visual and textual segmentation, as well as classification systems that capture the functional and qualitative aspects of webpages.In this paper, we introduce a novel webpage segmentation dataset that addresses these gaps by providing both visual and textual segmentations, alongside two classification frameworks.The first framework defines a nominal classification of segments based on their functional roles, such as main content, header, footer, and navigation bar.The second introduces an ordinal classification assessing the digital maturity of webpage segments, offering a structured evaluation of their design evolution and complexity.By integrating both classification schemes into a single dataset, our approach enables a more holistic analysis of webpage structures.Furthermore, given the rapid evolution of web design conventions, content structures, and technological trends, our dataset is designed to reflect contemporary webpage characteristics, ensuring its relevance for modern applications in web analysis and machine learning.Additionally, we provide first results on visual segmentation, demonstrating the effectiveness of our dataset in practical applications.