Document Classification by Computing an Echo in a Very Simple Neural Network

Abstract

In this paper we present a new classification system called ECHO. This system is based on a principle of echo and applied to document classification. It computes the score of a document for a class by combining a bottom-up and a topdown propagation of activation in a very simple neural network. This system bridges a gap between Machine Learning methods and Information Retrieval since the bottom-up and the top-down propagations can be seen as the measures of the specificity and exhaustivity which underlie the models of relevance used in Information Retrieval. The system has been tested on the Reuters 21578 collection and in the context of an international challenge on large scale hierarchical text classification with corpus extracted from Dmoz and Wikipedia. Its comparison with other classification systems has shown its efficiency.

Document Classification Using Artificial Neural Network

“…There are various literatures found in the field of text classification like [13] proposed neural network approach by representing text with its ASCII format. [4] Represents problem as information selection in which the process is based on spreading activation methods. Merkl and Rauber [11] article argue in favour of establishing a hierarchical organization of the document space based on unsupervised neural network using hierarchical feature map for text archive organization, a type of self organizing map.…”

Probabilistic Cascading for Large Scale Hierarchical Classification

“…Many researchers approach hierarchical classification problems [2,8] using flat classification, i.e. ignoring the hierarchy.…”