![]() | Only 14 pages are availabe for public view |
Abstract The interface agent is a computer program that employs machine-learning techniques to assist a user in dealing with a particular application by using a computer, learning new ”short cut” methods and suggesting a better ways for doing the user tasks. End- user programming and /or knowledge engineering for knowledge acquisition are used to learn interface agent but these learning systems require a large amount of work from the knowledge engineer. Furthermore, the knowledge of the agent is fixed and cannot be customized to the habits of individual users. This work presents a framework of multi-agent system, which consists of adaptive interface agent, feature extraction and set of agents trained by using neural network that collaborates with adaptive interface agent to detect the femur fractures in x-ray images automatically? Then, the proposed system decreases the workload of radiologist and improves the accuracy of diagnose their. The adaptive interface agent provides a semi-intelligent system learnt by the ”Customizer Dialog” from the user’s interests, goals and general preferences. A major problem with the learning approach is that the agent requires a sufficient number of examples before it takes accurate predictions. During this period, the user |