| Artificial Intelligence (AI) and Search Engines (SEO) | | | | because the initial pool of data may no longer be |
| 05th August 2006 | | | | valid. They are slow and expensive, but are |
| Author: Thomas Guilles | | | | widely used: stock market prediction, electronic |
| By Jose Nuñez | | | | circuit design, traveling salesman problem, |
| For software to be considered intelligent, it must | | | | international equity strategies, etc. |
| have the ability of making decisions. By interacting | | | | Artificial Neural Networks (ANN): Used for pattern |
| with its environment through sensors and acting | | | | recognition, and classification tasks in particular. |
| inits environment through effectors. They interact | | | | They deal well with imprecise data, as they can |
| by searching, comparing, learning...There have | | | | classify as "not too sure". There's plenty of |
| been at least 7 types of intelligent agents | | | | different kinds of neural nets: Recurrent, Hopfield |
| identified: collaborative, interface, mobile, | | | | Network, Bolzman Mmachine, SVM, |
| information, reactive, hybrid, and smart. | | | | Self-Organizing-Maps among others. |
| The most common agents for SEOs are the | | | | Semantic Networks (SN): These show the |
| information ones. These are important because of | | | | relationships and concepts. First a graph is created |
| their power search capabilities. But each agent is | | | | containing nodes which represent concepts and |
| given a very small, well-defined purpose. And | | | | links used to represent reltionships between the |
| there are several running in parallel to each other, | | | | concepts. There are six types of Semantic |
| to achieve an overall task. A.I. uses different | | | | Networks: Categories, Definitional, Assertional, |
| methods to build up search agents. | | | | Implicational, Executable, Learning, Hybrid. |
| Genetic Algorithms (GA): These are used in | | | | Jose Nuñez is a Scientific SEO/SEM |
| random search procedures. They are really | | | | Specialist, PHD in Computer Engineering |
| efficient on problems that are very difficult to | | | | Technology. |
| solve. Using them on dynamic data sets is difficult, | | | | |