Abstract-Good practices in software project management are basic requirements for companies to stay in the market, because the effective project management leads to improvements in product quality and cost reduction. Fundamental measurements are the prediction of size, effort, resources, cost and time spent in the software development process. In this paper, predictive Artificial Neural Network (ANN) and Regression based models are investigated, aiming at establishing simple estimation methods alternatives. The results presented in this paper compare the performance of both methods and show that artificial neural networks are effective in effort estimation.

-A fuzzy decision system for helping air-traffic experts in controlling airplane velocities and in keeping an airplane flight within several constraints established to air lane sections is proposed in this paper. Automatic systems for airtraffic control are essential due to the ever increasing number of airplanes flying all over the world, the amount of environmental and airplane constraints and the necessity to guarantee the safety both for flights and for air-traffic control operators. The proposed system uses Mamdani direct inference method. Results show the effectiveness of the developed fuzzy system in controlling the airplane velocity to achieve the desired performance and encourage the adequacy of the system to include several different variables usually employed in airtraffic control.

Abstract. This paper describes a novel neural network based multiscale image restoration approach. The method uses a Multilayer Perceptron (MLP) trained with synthetic gray level images of artificially degraded co-centered circles. The main difference of the present approach to existing ones relies on the fact that the space relations are used and they are taken from different scales, which makes it possible for the neural network to establish space relations among the considered pixels in the image. This approach attempts at coming up with a simple method that leads to an optimum solution to the problem without the need to establish a priori knowledge of existing noise in the images. The multiscale data is acquired by considering different window sizes around a pixel. The performance of the proposed approach is close to existing restoration techniques but it was observed that the resulting images showed a slight increase in contrast and brightness. The proposed technique is also used as a preprocessing phase in a real-life classification problem of medical Magnetic Resonance Images (MRI) by using a fuzzy classification technique.

Goal driven Intelligent Agents and Fuzzy Reference Gain-Scheduling (FRGS) approach are described in this paper as interchangeable concepts that are able to deal with dynamic complex problems. It is advocated that the FRGS approach may be viewed as an autonomous goal-based agent, that is, a fuzzy logic based agent able to autonomously adapt itself to environmental changes introduced by external inputs. The concept of fuzzy systems and intelligent agent are employed in decision-making problems to lead to a certain degree of autonomy in decision support system. Although the FRGS method was originally proposed for control application, this approach was extended to decision-making tasks due to its ability of emulating human reasoning. This new agent approach uses the external input information also denominated reference (goal) as the driven mechanism to determine the behavior of the system in order to achieve the desired objectives (goal). Thus, the FRGS approach can be modeled in the framework of an adaptive and goal (also context or environment) driven agent.

-This work presents a network intrusion detection method, created to identify and classify illegitimate information in TCP/IP packet payload based on the Snort signature set that represents possible attacks to a network. For this development a type of neural network named Hamming Net was used. The choice of this network is based on the interest to investigate its adequacy to classify network events in real-time, due to is capability to learn faster than other neural network models, such as, multilayer perceptrons with backpropagation and Kohonen maps. A Hamming Net does not require exhaustive training to learn. TCP/IP packet payloads were used as input pattern to the Hamming Net and Snort signature as exemplar patterns. The challenges faced to model the input and exemplar data and the strategies adopted to capture and scan relevant data in TCP/IP packets and in Snort signatures are described in this paper. In addition, the application architecture, the processing stages and some test results are presented.

No abstract

In this paper a multi layer perceptron neural network is used to retrieve vertical atmospheric temperature profiles from satellite radiation data. The training set consists of data provided by the direct model characterized by the Radiative Transfer Equation (RTE) 2) Direct problem Equation (I) represents the direct problem, in which IA is the spectral radiance, A is the channel frequency; 3 is the layer to space atmospheric transmittance function; subscript s denotes surface [9]; and B is the Planck function which is a function of the temperature T and pressure p given by equation (2), where h is the Planck constant, c is the light speed, and kB is Boltzmann's constant.The meteorological process of the atmosphere requires information of the vertical structure of the temperature and the water vapor, which are indirectly provided by satellite radiation data due to the lack of radiosonde observational stations over the globe. These pieces of information are especially important for weather analyses and data assimilation in numerical weather forecasting models in meteorology.Satellite measured radiation is interpreted in terms of meteorological parameters, requiring the inversion of the Radiative Transfer Equation (RTE) that relates measurements of radiation at different spectral frequencies to the energy at different atmospheric regions. There is a degree of indetermination associated with the spectral resolution and the number of spectral channels.Noises in the measuring process imply in instabilities to this solution [3,4], which led to the development of different methodologies and models to improve satellite data processing. Due to the difficulties to obtain correct RTE solutions, several approaches and methods have been developed to infer information from satellite data [5][6][7].In this paper a multilayer perceptron artificial neural network (ANN) is designed to solve the inversion of remotely sensed data in a multidimensional function approximation approach. The ANN retrieved temperatures are compared to those obtained [1] and [2], who used B2(T) = 'tc2/25 [hehlk,2T -ii (2)The solution of equation (l) maybe approached by discretization using central finite differences (3)

No abstract