George Boole was the first to describe a formal language for logic reasoning in 1847. The next milestone in artificial intelligence history was in 1936, when Alan M. Turing described the Turing-machine. Warren McCulloch and Walter Pitts created the model of artificial neurons in 1943, and it was in 1944 when J. Neumann and O. Morgenstern determined the theory of decision, which provided a complete and formal frame for specifying the preferences of agents. In 1949 Donald Hebb presented a value changing rule for the connections of the artificial neurons that provide the chance of learning, and Marvin Minsky and Dean Edmonds created the first neural computer in 1951. Artificial intelligence (AI) was born in the summer of 1956, when John McCarthy first defined the term. It was the first time the subject caught the attention of researchers, and it was discussed at a conference at Dartmouth. The next year, the first general problem solver was tested, and one year later, McCarty?regarded as the father of AI?announced the LISP language for creating AI software. Lisp, which stands for list processing, is still used regularly today. Herbert Simon in 1965 stated: “Machines will be capable, within twenty years, of doing any work a man can do.” However, years later scientists realized that creating an algorithm that can do anything a human can do is nearly impossible. Nowadays, AI has a new meaning: creating intelligent agents to help us do our work faster and easier (Russel & Norvig, 2005; McDaniel, 1994; Shirai & Tsujii, 1982; Mitchell, 1996; Schreiber, 1999). Perceptrons was a demonstration of the limits of simple neural networks published by Marvin Minsky and Seymour Papert in 1968. In 1970, the first International Joint Conference on Artificial Intelligence was held in Washington, DC. PROLOG, a new language for generating AI systems, was created by Alain Colmerauer in 1972. In 1983, Johnson Laird, Paul Rosenbloom, and Allen Newell completed CMU dissertations on SOAR.
In 1950 Alan Turing suggested a definition for deciding whether software is intelligent or not. In his theory the software’s intelligent behavior can be measured like a human intellectual efficiency. The software is intelligent when a human being does not know if he or she is chatting with the software or with another human. That test was called the Turing test, and here is how it works: if the software passes the test, it is called intelligent software—also called intelligent agent—which percepts the environment with sensors and acting with effectors.
The term embodied conversational agent (ECA) (Cassel, 2007; Huget, 2003; Cassel, Sullivan, Prevost, & Churchill, 2000) is used for special software or hardware as an extension of an intelligent agent, not just because these are able to communicate with the user via natural language, but also for their emotion system (Benkő & Sik Lányi, 2007). There are many emotional models (Ruebenstrunk, 1998) for creating an embodied conversational agent, including:
Virtual reality (VR) can be used for designing and testing an ECA because the developing process can be easier and cheaper with VR technology (Ortiz, Oyarzun, Carretero, & Nestor, 2006; Takacs & Kiss, 2003). An avatar is a spatial creature that usually symbolizes or simulates a human being in exterior and in behavior also. The next article describes the VTR, the modeled emotions, and the avatars that were created for a virtual therapy room.
Key Terms in this Chapter
Turing Test: Test created by Alan M. Turing, who said that a machine can think. The test analyzes if a person in an isolated room can decide exactly who he or she is chatting with—a human being or an AI agent. The agent is in the other room and the chatting is in written form.
Backward Reasoning: Searching from the initial state to the final state in, for example, action planning.
Robot: A device that performs programmed operations or that operates by remote control. A robot senses external feedback derived from ongoing operations and reacts to sensed data by modifying its actions accordingly.
Neural Network: A computation system containing a set of connected elements to solve arithmetic problems. The basis of the neural network computation is to analyze how the brain works and simulate it.
Fuzzy Logic: Techniques for reasoning under uncertainty. It is capable of working with concepts such as ‘thin’, ‘fat’, ‘long’, and ‘short’, if there is no exact data for supporting the decision.
Knowledge Engineering: Related to mathematical logic, and building, maintaining, and developing knowledge-based systems.
Searching: A searching problem contains the initial state, the operators, the final state, and the cost of searching. The searching is efficient if it finds an optimal solution if a solution exists and it can be reached in a short time. Searching methods include: blind search (breadth-first, depth-first, depth-limited, iterative deepening, iterative broadening, uniform-cost) and heuristic search (hill-climbing, best-first, A, A*, IDA*, SMA*, simulated annealing).
Bayesian Network: A mathematic model in graphic form that represents a set of variables and their probabilistic independencies. It can be used, for example, to calculate the probability of a patient having a specific disease.
Forward Reasoning: Searching from the initial state to the final state in, for example, action planning.
Strong AI: The main goal of strong AI is to create an AI agent that can think and have a mind.
Expert System: A computer program that contains subject-specific knowledge of human experts. Such systems are used for giving advice. Also known as a knowledge-based system.
Intelligent Agent: An agent is an entity with the capability to observe and act in an environment. It is intelligent if it interacts like a human being. It can be a robot or a software system, depending on the environment.
Knowledge Representation: Describes how information can be stored efficiently. It uses state space and the information is represented with a graph. States are represented with nodes and actions with arcs.
Genetic Algorithm: A method of evolutionary computation for problem solving. There are states also called sequences and a set of possibility final states. Methods of mutation are used on genetic sequences to achieve better sequences.
Artificial Intelligence: The capability of a device to perform functions that are normally associated with human intelligence, such as reasoning, learning, and self-improvement.
General Problem Solver: Uses means-ends-analysis heuristic for solving formalized symbolic problems. A GPS computer program solves simple problems that can be formalized such as the Towers of Hanoi.
Weak AI: Refers to software to study or solve problems and reasoning tasks that do not need the full range of human cognitive abilities.
Heuristic: General advice that is usually efficient but sometimes cannot be used; also it is a validate function that adds a number to the state of the problem.