Figure 2

Fragment of the transition network to detect resource names. This is an extract of the transition network used to identify resource names from the surrogates. According to the inputs (labels over the arcs), the transition network passes from state 0 to other states until it reaches a final state that determines the identification of a new resource. Different tags are used as labels: punctuation marks and words within quotes represent literals---i.e. marks or words that are in the surrogates---, and abbreviations in capital letters that represent morphological tags. For instance, NP is a proper noun, CC represents a conjunction, NN refers to common noun, DT symbolizes a determiner, JJ denotes an adjective, RB is a general adverb, OF and IN represent prepositions, PDT denotes a determiner, VB represents a verb, VBG refers to a verb in gerund form (ending in -ing) or SYM is a symbol or formula.