Wyniki wyszukiwania - BazTech

1

Decision making support system for managing advertisers by ad fraud detection

Gabryel Marcin, Scherer Magdalena M., Sułkowski Łukasz, Damaševičius Robertas

Journal of Artificial Intelligence and Soft Computing Research

|

2021

|

Vol. 11, No. 4

331--339

EN

Efficient lead management allows substantially enhancing online channel marketing programs. In the paper, we classify website traffic into human- and bot-origin ones. We use feedforward neural networks with embedding layers. Moreover, we use one-hot encoding for categorical data. The data of mouse clicks come from seven large retail stores and the data of lead classification from three financial institutions. The data are collected by a JavaScript code embedded into HTML pages. The three proposed models achieved relatively high accuracy in detecting artificially generated traffic.

2

Novel hybrid approaches based on evolutionary strategy for streamfow forecasting in the Chellif River, Algeria

Zakhrouf Mousaab, Bouchelkia Hamid, Stamboul Madani, Kim Sungwon

Acta Geophysica

|

2020

|

Vol. 68, no. 1

167--180

EN

In this study, the feedforward neural networks (FFNNs) were proposed to forecast the multi-day-ahead streamfow. The parameters of FFNNs model were optimized utilizing genetic algorithm (GA). Moreover, discrete wavelet transform was utilized to enhance the accuracy of FFNNs model’s forecasting. Therefore, the wavelet-based feedforward neural networks (WFFNNs-GA) model was developed for the multi-day-ahead streamfow forecasting based on three evolutionary strategies [i.e., multi-input multi-output (MIMO), multi-input single-output (MISO), and multi-input several multi-output (MISMO)]. In addition, the developed models were evaluated utilizing fve diferent statistical indices including root mean squared error, signal-to-noise ratio, correlation coefcient, Nash–Sutclife efciency, and peak fow criteria. Results provided that the statistical values of WFFNNs-GA model based on MISMO evolutionary strategy were superior to those of WFFNNs-GA model based on MISO and MIMO evolutionary strategies for the multi-day-ahead streamfow forecasting. Results indicated that the performance of WFFNNs-GA model based on MISMO evolutionary strategy provided the best accuracy. Results also explained that the hybrid model suggested better performance compared with stand-alone model based on the corresponding evolutionary strategies. Therefore, the hybrid model can be an efcient and robust implement to forecast the multi-day-ahead streamfow in the Chellif River, Algeria.

3

Convergence Analysis of Multilayer Feedforward Networks Trained with Penalty Terms: A review

Wang J., Yang G., Liu S., Zurada J. M.

Journal of Applied Computer Science Methods

|

2015

|

Vol. 7 No. 2

89--103

EN

Gradient descent method is one of the popular methods to train feedforward neural networks. Batch and incremental modes are the two most common methods to practically implement the gradient-based training for such networks. Furthermore, since generalization is an important property and quality criterion of a trained network, pruning algorithms with the addition of regularization terms have been widely used as an efficient way to achieve good generalization. In this paper, we review the convergence property and other performance aspects of recently researched training approaches based on different penalization terms. In addition, we show the smoothing approximation tricks when the penalty term is non-differentiable at origin.

4

Generalization of raster images containing patterns featuring stochastic repetitiveness

Rataj A.

Machine Graphics and Vision

|

2007

|

Vol. 16, No. 1/2

191-201

EN

The issue of reconstruction of missing or unreliable parts of an image is one of the basic problems in image processing. For example, there are a number of methods for texture generation on the basis of a small sample. This paper presents a method that 'bottlenecks' an image processing feedforward neural network so that only some basic traits of the image are preserved. These basic traits are in turn used to generalize the image, thus filtering out any unusual parts of the image. The ability of neural networks and several other learning machines to generalize is based on the premise of smoothness of the generalizing function. Thus, in order to detect advanced patterns that exhibit complex traits like repetitiveness, instead of training these machines directly with raw data, transforms of the patterns like the Fast Fourier Transform are sometimes performed. In this paper it is shown, that a simple feedforward neural network, without any pre-processing of the training data, using the described 'bottleneck' architecture, can properly predict a stochastically repetitive pattern in a raster image.