WebRequests

Abstract

A dataset of labeled requests assembled from several public datasets, namely, Malicious-URLs, PKDD, and CSIC 2010 (also included). To merge the datasets, only the URI of each web request was used. To construct a feature vector to train the networks, each URI was tokenized in unigrams following a bag-of-words approach. For each URI, the values of the unigrams were computed using term frequency–inverse document frequency (TF–IDF). Each URI was represented by an l1-normalized vector composed of the 500 most frequent tokens across the entire dataset.