Abstract. The ionospheric critical frequency, foE2, is forecast 1 hour in advance by using artificial neural networks. The value of foE2 at the time instant k of the day is designated byf(k). The inputs used for the neural network are the time of day; the day of year; season information; past observations offoF2; the first difference A l(k ) = f(k)-f(k-1); the second difference A 2 (k) = A 1 (k) -A l(k -1); the relative difference RA (k) = A l(k)/f(k); geomagnetic indices Kp, ap, Dst, sunspot number, and solar 10.7-cm radio flux; and the solar wind magnetic field components By and B•. This paper gives a new method, and it is the first application of neural networks for modeling both temporal and spatial dependencies. In order to understand the physical characteristics of the process and determine how important a particular input is, a test which shows the relative significance of inputs to the neural networks is performed at the output. The performance of a neural network is measured by considering errors. For the errors to be more meaningful, training and test times and times for comparison with other results are selected from the same solar activity period. Among the various neural network structures, the best configuration is found to be the one with one hidden layer with five hidden neurons, giving an absolute overall error of 5.88%, or 0.432 MHz. IntroductionSolar activity has an impact on magnetospheric dynamics, which in turn influences the plasma density in the ionosphere. It is important to quantify the plasma density variability for both scientific and practical reasons. One of the practical methods for determining the ionospheric variability' is to quantify the electron density or the ionospheric critical frequencyfo The objective is to determine the short-term details of the ionospheric variability, and 1 hour is a typical time period for obtaining the details of the medium reasonably well from the point of view of a telecom-1497

No abstract

No abstract

Abstract:In this paper, an improved version of smart program recommendation system based on Hybrid Broadcast Broadband Television (HbbTV) technology is proposed. The learning part which was based on the artificial neural network (ANN) has been enhanced by incorporating the genetic algorithm. Instead of assigning all users to the same ANN, clustering is introduced by utilizing preferred genre information obtained explicitly. The number of clusters is found automatically. Gathering the user data and presenting the television program recommendations to the user are realized by the HbbTV technology. The proposed system has been tested by the data from 248 people and has given successful results.

In this paper, an earlier method proposed by the authors to make smart recommendations utilizing artificial intelligence and the latest technologies developed for the television area is expanded further using controlled clustering with genetic algorithms (CCGA). For this purpose, genetic algorithms (GAs), artificial neural networks (ANNs), and hybrid broadcast broadband television (HbbTV) are combined to get the users' television viewing habits and to create profiles. Then television programs are recommended to the users based on that profiling. The data gathered by the developed HbbTV application for previous studies are reused in this study. These data are employed to cluster users. The number of clusters is found by CCGA, a method proposed in this paper. For each cluster formed by CCGA, a separate ANN is designed to learn the viewing habits of the users of the corresponding cluster. The weight matrices are initialized also by GA. The recommendations produced using the proposed model are then presented by the same HbbTV application developed by the authors. Clustering with GAs gives better results when compared to the well-known K-means clustering algorithm.