Wordfinding Problems and How to Overcome them Ultimately With the Help of a Computer

Our ultimate goal is to help authors to find an elusive word. Whenever we need a word, we look it up in the place where it is stored, the dictionary or the mental lexicon. The question is how do we manage to find the word, and how do we succeed to do this so quickly? While these are difficult questions, I believe to have some practical answers for them. Since it is unreasonable to perform search in the entire lexicon, I suggest to start by reducing this space (step-1) and to present then the remaining candidates in a clustered and labeled form, i.e. categorial tree (step-2). The goal of this second step is to support navigation. Search space is determined by considering words directly related to the input, i.e. direct neighbors (associations/co-occurrences). To this end many resources could be used. For example, one may consider an associative network like the Edinburgh Association Thesaurus (E.A.T.). As this will still yield too many hits, I suggest to cluster and label the outputs. This labeling is crucial for navigation, as we want users to find the target quickly, rather than drown them under a huge, unstructured list of words. Note, that in order to determine properly the initial search space (step-1), we must have already well understood the input [mouse1 / mouse2 (rodent/device)], as otherwise our list will contain a lot of noise, presenting ’cat, cheese’ together with ’computer, mouse pad’, which is not quite what we want, since some of these candidates are irrelevant, i.e. beyond the scope of the user’s goal.