Have you heard or seen  some cool computer robots or programs that can really act cool you just want to fall in love with them?, yeah, l do most of the time( on youtube lol). Some of these programs or robots are fascinating to me but those l really like to talk about are the ones called expert systems. Expert systems are programs written using the knowledge of an expert in a a particular field.

For example, a gynaecologist is an expert doctor in treating women for special needs. I don’t want to go into what they do because that’s not the point here lol. If you want to know what they do, better go study that or peek a boo(hide in their office all day watching them work lol). So there are two groups of people who come together to create an expert system. The expert in the field and a knowledge Engineer. A knowledge Engineer is a cool computer science programmer(Artificial intelligent programmer), who can study or interact with the expert and gain details about how he does his job, how he comes about conclusions, how he figure what prescription, how he figure out this and that.

The knowledge Engineer will then write a program using rules detailing all the actions that an expert performs to make a decision. These rules are stored in what is called the Knowledge Base. The knowledge Base is specific to a particular case. That is to say, if you are building an expert system for medical case, that knowledge base is useless to a construction work knowledge base. In any case, the expert in that field is called the domain expert and the computer guy is the same Knowledge Engineer in every case.

The Engineer also builds a Fact database that will contain any specific detail about a particular case. The Fact database and the Knowledge base are  literally connected to one of  the already built Expert System Shells that exist in the world of Computer science. The expert system shells are tool kits that the Engineer can use to manipulate the knowledge base and the fact database to provide meaningful outcome to the user. The expert shell has a knowledge base editor that the engineer can use to edit or add information and allow updates to the system.

The expert system shells also have an inference Engine that can use the Rules and Facts to reach a conclusion for any type of query related to this field. So for example, if the Expert system is in the medical fied, and there are rules stored in the knowledge base telling how to detect that some has cancer. The patient comes to the hospital, then the doctor will ask questions about how he or she feels and then, those symptoms are entered into the Fact database for this particular patient. When the doctor finished and ask the system if the patient has cancer?  The inference engine will take those facts and look at the rules in the knowledge base to see if any rule matches any fact in the database. When a match is found for any rule, then the inference engine can perform either forward chaining on backward chaining, which are ways of finding a conclusion from a given query in the expert system shell. if it’s able to prove using the facts and rules that the patient has cancer, it will say cancer is proved. if the system get stacked on the way and couldn’t continue the prove because something is missing about the symptoms to proof cancer, it will ask the doctor, is the patient having this or that feeling?. Once the doctor provide that answer, it will continue trying to find the conclusion. If the system asks a question or proves something and the doctor wants to know why it proves or asks him that question, the doctor can also ask the system questions. Like, why did you prove cancer? There is another component of the expert system shell called the  Explanation system that will handle the explanation base on the rules and the facts that are already in the system or lacking in the system.

These systems can help a lot because it’s like repository of smart knowledge. Some examples of the Tool kits or the expert system shells  Eclipse, CLIPs(C Langauge Integrated Production System),OPS5,ART and JESS.

An example of an expert system is MYCIN developed at Stanford University in 1984. It’s used in the medical field and can assist a doctor to prescribe antimicrobial drugs for blood infections.

These systems are cool but do you wonder if one day, they started getting knowledge on their own and take over the world? lol

Haha, that’s not going to happen. They are bunch of idiots, if humans don’t feed them knowledge, they can’t do anything, so don’t be scared. But when a crazy person programmed a robot to go about killing everybody, then we have a problem. watch out !!!! lol

Leave a Reply

Your email address will not be published. Required fields are marked *

Name *