Introduction
Computer scientists have in recent years come up with ways of solving problems that occur in the real life situations by use of computers. Artificial intelligence is the field that computer scientists have concentrated much on so as to achieve intelligent agent software. It is a study whereby computer systems are developed which can act and think like human beings. In this domain computer scientists have narrowed this domain to come up with knowledge based systems.
The basic structure of a knowledge based, system is that it has a knowledge base which store the instructions that guide the system. This knowledge base is composed of facts in real life situations.When a knowledge based system is faced with a problem or encounter a problem, what it does is that it compares the input with what is in the knowledge base and deduces a conclusion from these facts and show them to the user through the interface (Gilbert 1997).
Also there must be an inference engine whereby the system compares the problem and the facts.This component of the system can be compared with the brain of a human being which helps the system to come up with a concrete decision. There are minor components such as knowledge acquisition facility. This component again helps the whole system to continue gaining more knowledge meaning that it can become wiser. It also means that it can adapt to new situations by storing new cases hence acquiring knowledge from the environment. These basic components are the same ones that develop an agent although agents are developed for a specific domain.
Autonomous agents can therefore be defined as systems confined in an environment (Boden 1994). These systems can also be a component of that environment and it acts and senses that environment over time so as to accomplish its goals and react or respond to stimuli. These agents therefore use perception to identify events in the environment or to detect these events and hence respond to them. Perception also helps to identify all the objects around it. Such agents must therefore communicate so as to exchange information with each other.
Examples of Agent human interactions includes: interface agents that assist users to accomplish their goals efficiently and effectively when using a computer system. They are autonomous in the fact that they customize themselves according to the requirements and goals of the user. At any one time these interface agents keep waiting for the inputs from the user so that it can customize itself accordingly (Boden 1994). Another agent human interaction is the emotional human intelligent system. It is developed with the main aim of understanding other people or agents and respond accordingly.It should therefore have capability of self-awareness.
In this case study a system that detects the status of the baby, that is, if the baby is awake, wet or hungry and it has an interface implemented with agent human interaction is considered.
Effects
In some cases this system can act independent or autonomous.By this it means that the system is using the facts or knowledge in the database to make conclusions. It must therefore have an interface that is highly sensitive to its environment. It will achieve this by perceiving its environment.That is applicable only when the baby is awake because such instructions such as “if the baby is crying means the baby is awake” are in the knowledge base. The agent human interaction of this system will be developed in such a way that it can customize itself according to the behavior of the baby.
Application
Agent human interactions can be applied to diverse systems such as the one above-for monitoring the baby. What is essentially done is only to customize such a system to fit into a specific domain. A personal agent can also be developed using Agent human interaction.It is developed in a manner that it can do conversation with one another just like human being (Maes 1995). Other applications include in music, military and in games. In games, it is used to maximize its moves and minimize the moves of the opponent. It can therefore detect the moves of the opponent and come up with concrete decisions on how to make intelligent moves.
Problem
The main problem with the system that monitors the baby is how to alter the knowledge in the database so that it conforms to the babies with disability. For example if the baby cannot make movements or the baby cannot produce any sound, it has to be developed with other facts in the database. Also the baby may be speaking a different language from the one programmed in the agent human interaction. This poses another problem also.
Solutions
The solution is to come up with agent human interaction software that integrates many possibilities. By this it means that it should consider all the possibilities in a manner that it will not be only limited to those children that are not disabled. It should also provide away that will allow the user to modify the facts or the rules in the database to make much more flexible.
Both solutions are crucial to this system because they will allow the user to experience more satisfaction regardless of the situation. Both of them make the system more robust in the sense that all the situations will be catered for.
The strength of the system is that it is in a position to perceive the status of the baby and deduce the correct action to take. However this is only possible if all the cases, facts or the rules have been coded well in the knowledge base. It has also the capacity to customize its interface according to the goals of the user. If the user is allowed to change the domain knowledge it provides much more satisfaction because of the flexibility involved. There are also several weaknesses such as a case where the system will not understand more than one language or a case whereby it will not understand local languages or mother tongue. This is the main reason it is suggested that a solution based on the possibility of changing the domain knowledge be incorporated.
Conclusion
Agent human interaction so far has been a success but a lot need to be done so as to overcome the challenges at hand. These challenges include inaccessible techniques for implementing agents and of course lack good techniques to implement natural language processing and perception which are very crucial. It is estimated that in the future, agents will be developed that will be able to interact naturally with a normal human being including socializing. It is also expected that as the power of the computer hardware continue to increase, future agents will surpass human brain. The current agents are developed to make regular work easier to human beings and also work in harsh conditions where human beings cannot ( Maes 1995). To achieve all these, more artificial intelligent techniques need to be explored and also implement more complex behaviors.
References
Boden, M. (1994) Agents and Creativity, Communications of the ACM 37,7, 117-121.
Gilbert, D. (1997) Intelligent Agents: The Right Information at the Right Time, IBM Corporation.
Maes, P. (1995) Agents that reduce work and information overload. In Communications of the ACM, volume 38.