Automatic text summarization has long been studied and used. The growth in the amount of information on the web results in more demands for automatic methods for text summarization. Designing a system to produce human-quality summaries is difficult and therefore many researchers have focused on sentence or paragraph extraction, which is a kind of summarization. In this paper we introduce a new method to make such extracts. Genetic Algorithm-based sentence selection is used to make a summary, and once the summary is created, it is evaluated using a fitness function. The fitness function is based on three following factors: Readability Factor, Cohesion Factor, and Topic-Relation Factor. In the paper we introduce these factors, and discusses the Genetic Algorithm with the specific fitness function. Evaluation results are also shown and discussed in the paper.

No abstract

Emotions are the subject of study in many research areas, including psychology, physiology and artificial intelligence. However, there is not yet a general computational definition for emotions and many of the related works in artificial intelligence are based on uncertain assumptions about the origin, features, and applications of emotions. Therefore, we direct our studies toward achieving a better understanding of emotions and generating a computational model that could be a base for further research in the field. As a first step, we present the results of our study into the possible role of emotions in mental resource management. To do so, we will introduce a model of emotions based on managing mental resources. We have applied it to a general model of mind and the whole model has been implemented in an artificial life simulation environment called Zamin to evaluate its ability to produce emotional behavior and to make better decisions. The results, as discussed in this article, show that a resource management mechanism could be the source of emotion generation in the mind.

No abstract

Artificial Immune System algorithms use antibodies which fully specify the solution of an optimization, learning, or pattern recognition problem. By being restricted to fully specified antibodies, an AIS algorithm can not make use of schemata or classes of partial solutions. This paper presents a symbiotic artificial immune system (SymbAIS) algorithm which is an extension of CLONALG algorithm. It uses partially specified antibodies and gradually builds up building blocks of suitable sub-antibodies. The algorithm is compared with CLONALG on multimodal function optimization and combinatorial optimization problems and it is shown that it can solve problems that CLONALG is unable to solve.