Interesting article on the Generation5 " Application of Biological Metaphors for Identifying and Killing Spam
" (S. Evans). Is the topic? creation of an anti-spam filter Artificial Neural Networks (ANN, Artificial Neural Networks
). As introduccióny theoretical argument, the author makes an interesting analogy, the phenomenon of spam as a living ( Spam as a Living Organism), in constant evolution and adaptation " medium "(in this case the email and streaming media formats and through Internet), governed by its own laws of" natural fit. " Complete with C # source code, this article looks at how to Identify a Variety of spam using biological metaphors Such as neural networks. [ source] The article includes the source code of the filter, downloadable in a zip file, and developed in C #
language. Like all RNA , this code requires previous training to operate correctly, as explained in the text ( Training the Networks) . In the results section, we are informed that, in tests, has obtained a ruling just over 1000 emails analyzed. On another issues, not very far in any case, and in Generation5 , se
publico hace ya varias semanas otro artículo que me gustaría
destacar: " Intelligence
- Artificial and Otherwise " (Bruce L. Toy):
This paper presents a model of artificial intelligence, based
on an XML
memory structure, a logic architecture of divisible processor functions,
and real-time sensor integration. It incorporates learning, consciousness,
creativity, and growth in a framework that can be used to understand human
behavior, build an independent intelligent entity, or integrate a large-scale
computer network into an intelligent system.
fuente ] Using XML to implement the memory structure of the working model based on IA
features. Since the abstract of the article just quoted is sufficiently clear and expressive, I do not think we should add anything else about it except to recommend its reading.
0 comments:
Post a Comment