Teaching Neural Networks to Detect the Authors of Texts Using Lexical Descriptors

This paper proposes a means of using an artificial neural network todistinguish the authors of paragraphs. Once the network has beentrained, its hidden layer activations are recorded as a representation ofthe average number of words and average characters of words in aparagraphs of an author. This stored information can then be used toidentify the texts written by authors. This computational task is solvedby dividing it into a number of computationally simple tasks and thencombining the solutions to those tasks. Computational simplicity isachieved by distributing the learning task among a number of experts,which in turn divides the input space into a set of subspaces. Thecombination of these experts is said to constitute a committee machine.Basically, it fuses knowledge acquired by experts to arrive at an overalldecision that is supposedly superior to that attainable by anyone of themacting alone. By this, we succeeded to distinguish the paragraphsauthored by Ivo Andri , from the ones authored by Mehmed Me aSelimovi .