Specialized neural network and fuzzy logic algorithm in comparison for word frequency prediction in document filtering

For very large document collections or high volume streams of documents, finding relevant documents is a major information filtering problem. A major aid to information retrieval systems produces a word frequency measure estimated from some important parts of the document using neural network approaches. In this paper a fuzzy logic technique and as its simplified case a neural network algorithm will be proposed for this task. The comparison of these two and an alternative neural network algorithm will also be discussed.