A trial for data retrieval using conceptual fuzzy sets

We describe trial applications of fuzzy sets to data retrieval. The objectives are to test their ability to achieve conceptual matching between retrieved objects and the user's intention and to connect real data with symbolic notations. The algorithm proposed retrieves data that conceptually fit the meanings of the entered keyword. An algorithm is described that uses fuzzy sets to handle word ambiguity (the main cause of vagueness in the meaning of a word). It is based on conceptual fuzzy sets (CFSs), which represent the meaning of words by chaining other related words. Two trial applications of this algorithm to data retrieval are described. First, an application to image retrieval shows variation of data retrieval with conceptual matching and transformation of numeric values into symbols. Next, an application to the agent recommending a TV program shows the method that lets CFSs fit to the sense of a user by Hebbian learning.