The Frame Problem in Text Analysis

The aim of this research is to evaluate the extent to which Thesaurus allow us to modify a researcher's knowledge frame. A well calculated Thesaurus has the power to overwrite existent knowledge frames, or habitual "heuristics" for the humanities, if word occurrence data are rigorously manipulated by the algorithms of statistical linguistics. However, even if all targets and means of data gathering and analyzing are readily available, there remains behind various interpretations of subjects a sort of Frame question of how we partition off the texts and documents to avoid arbitrary text segmentation. Frames are needed before we can gather and interpret the data for a word occurrence computation. If the problem of text segmentation remains unresolved, any experiment in quantitative text analysis will be still far from being realized. We need a sort of “TextTiling” methodology enabling an objective segmentation based on objective criteria.