Data for: A Multimodal Generative and Fusion Framework for Recognizing Faculty Homepages

This dataset includes image feautres, layout features, textual content features, and target labels. 2. The image features are 4-dimensional vectors, in which the elements include the number of zero-face images, the number of one-face images, the number of multiple-face images, and the total number of images. 3.

The layout features are 300-dimensional vectors, in which every element calculates the total number of a HTML leaf tag. 4. The textual content features are 400-dimensional vectors. Every vector is a padded/truncated word sequence from a textual content. 5.