Use of Artificial Intelligence techniques to solve complex problems [closed]


In the face of the task of developing a complex application, I decided to make every time a human being interacts with the system, to record their decisions, apply weights and in the next suggestions use

After some research, I saw that what I want to do has already been named, RBC and Genetic Algorithms and are artificial intelligence techniques.

I know that I will have to do a lot of research and that there is probably not much material about it, but it would be very helpful if someone who has had experience developing a similar solution could share and indicate the path of the stones, regardless of the language used.

asked by anonymous 15.12.2015 / 20:20

2 answers


Good afternoon. As a TCC, I made software to support decision making for companies that work with technical maintenance (whether it's computers or not). My method for software decision-making was the calculation (with percentages) of a recurrence of a fact. For example:

Notebook HP - Modelo Pavilion DV6000
Defeito: Não liga
70% dos casos, problema de BGA
15% dos casos, pente de memória com defeito
10% dos casos, sistema operacional corrompido
5% dos casos, HD com defeito

I took into account the following variables: Type of Equipment, model of equipment and defect presented.

From this information I modeled a database that could store a knowledge base and that with each new open service order, the inference engine would aid in decision making. The same structure I repeated for the cause solution as follows:

Notebook HP - Modelo Pavilion DV6000
Defeito: Não liga
Causa: Sistema operacional corrompido
60% dos casos, formatação
30% dos casos, recuperação do sistema
10% dos casos, recuperação do boot do sistema

So, in short, what I did was: For each type and model of equipment with defect X, there may be N causes and each cause may have N solutions If necessary, put the model of the bank here.

Here's the template I've been working on to update: There is one more table named Events, which has the responsibility of storing when the event occurred, which further helps to filter out the probabilities of causes and remedies of the defect.

Hope this helps! Oss


Curiosity: I developed in php (without framework) and mysql. hehe

28.12.2015 / 17:58

This is not a good way to solve the timesheet issue.

This is a complex mathematical question and people will not provide good answers no matter how smart they are. Every artificial intelligence technique starts from a training, a database in which one has the best answers to a set of problems, from there the system learns to infer a new response close to the ideal for a similar new problem. Well I do not want to confuse you anymore, I'll tell you what your problem is.

This is an optimization problem, not artificial intelligence.

Defining with my words, every time we have a scarce resource that must be allocated, or a function that must be optimized for a goal obeying various constraints, we have an optimization problem.

Timetable is a classic problem of optimization, we have to allocate schedules for the work of people, say 24h, but we have several restrictions, for example, we have employees working 8 hours, 6 hours, we must have an interval of 1 hour between 8 o'clock, once a week, and several other legislative restrictions.

The algorithms then attempt to arrive at an optimized function that meets all constraints imposed, minimizing or maximizing the use of resources.

Another example would be cutting wood, maximizing reuse of the floor area and minimizing the cost. The constraints are the required cut shapes and their respective amount.

In fact, the time problem is exploited by some companies that provide consulting or software to facilitate modeling and resolution.

A former professor of mine gives classes in optimization, and has a company that specializes in making work scales using these techniques that I commented, the company is link

It even has software in which I did a temporary job, where people set up the work schedule and their restrictions, and the system resolves the best possible scale according to all the legislation, avoiding fines and minimizing the number of employees, in fact saving on corporate payroll.

I've had this business class and I find it harder than artificial intelligence, everything that has to decorate something I find more difficult :-), I did not find a good material to pass on but there is this one www.maxwell.vrac.puc that talks about the subject, good basically there is an algorithm called min-max to solve these problems, it does not have a high computational efficiency of the classic form, there are more optimized and complex variations and methods, but it's a good start. This teacher I spoke to was student of this alien individual here: Petr Iakovlevitch Ekel - link

This Petr teaches Mathematics Methods in the Master of Electrical Engineering of PUC and is simply the world's largest authority on the subject, the guy works for governments, optimizing the installation of electrical power meshes in countries, just to get an idea. ..

31.12.2015 / 01:18