scholarly journals Decision Making Support System for Managing Advertisers By Ad Fraud Detection

2021 ◽  
Vol 11 (4) ◽  
pp. 331-339
Author(s):  
Marcin Gabryel ◽  
Magdalena M. Scherer ◽  
Łukasz Sułkowski ◽  
Robertas Damaševičius
Keyword(s):  
Neural Networks ◽  
Decision Making ◽  
Financial Institutions ◽  
Support System ◽  
Categorical Data ◽  
Fraud Detection ◽  
Feedforward Neural Networks ◽  
High Accuracy ◽  
Online Channel ◽  
Lead Management

Abstract 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.

scholarly journals Neural Networks in Credit Risk Classification of Companies in the Construction Sector

2018 ◽  
Vol 2 (2) ◽  
pp. 63-77 ◽  
Author(s):  
Aleksandra Wójcicka
Keyword(s):  
Neural Networks ◽  
Decision Making ◽  
Credit Risk ◽  
Financial Institutions ◽  
Real World ◽  
Decision Making Process ◽  
Construction Sector ◽  
Real World Data ◽  
Input Variables

The financial sector (banks, financial institutions, etc.) is the sector most exposed to financial and credit risk, as one of the basic objectives of banks' activity (as a specific enterprise) is granting credit and loans. Because credit risk is one of the problems constantly faced by banks, identification of potential good and bad customers is an extremely important task. This paper investigates the use of different structures of neural networks to support the preliminary credit risk decision-making process. The results are compared among the models and juxtaposed with real-world data. Moreover, different sets and subsets of entry data are analyzed to find the best input variables (financial ratios).

scholarly journals Auto-Associative Neural Networks to Improve the Accuracy of Estimation Models

2010 ◽  
pp. 66-81 ◽  
Author(s):  
Salvatore A. Sarcia ◽  
Giovanni Cantone ◽  
Victor R. Basili
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Software Engineering ◽  
Feedforward Neural Networks ◽  
High Accuracy ◽  
Reduction Technique ◽  
Primary Reason ◽  
Model Improvement ◽  
Auto Associative Neural Network ◽  
Estimation Models

Prediction of software engineering variables with high accuracy is still an open problem. The primary reason for the lack of high accuracy in prediction might be because most models are linear in the parameters and so are not sufficiently flexible and suffer from redundancy. In this chapter, we focus on improving regression models by decreasing their redundancy and increasing their parsimony, i.e., we turn the model into a model with fewer variables than the former. We present an empirical auto-associative neural network-based strategy for model improvement, which implements a reduction technique called Curvilinear component analysis. The contribution of this chapter is to show how multi-layer feedforward neural networks can be a useful and practical mechanism for improving software engineering estimation models.

scholarly journals Functional connectome fingerprinting using shallow feedforward neural networks

2021 ◽  
Vol 118 (15) ◽  
pp. e2021852118
Author(s):  
Gokce Sarar ◽  
Bhaskar Rao ◽  
Thomas Liu
Keyword(s):  
Neural Networks ◽  
Recurrent Neural Networks ◽  
Resting State ◽  
State Of The Art ◽  
Feedforward Neural Networks ◽  
High Accuracy ◽  
Correlation Matrices ◽  
Functional Connectome ◽  
Temporal Features ◽  
Data Points

Although individual subjects can be identified with high accuracy using correlation matrices computed from resting-state functional MRI (rsfMRI) data, the performance significantly degrades as the scan duration is decreased. Recurrent neural networks can achieve high accuracy with short-duration (72 s) data segments but are designed to use temporal features not present in the correlation matrices. Here we show that shallow feedforward neural networks that rely solely on the information in rsfMRI correlation matrices can achieve state-of-the-art identification accuracies (≥99.5%) with data segments as short as 20 s and across a range of input data size combinations when the total number of data points (number of regions × number of time points) is on the order of 10,000.

scholarly journals Functional Connectome Fingerprinting Using Shallow Feedforward Neural Networks

2020 ◽  
Author(s):  
Gokce Sarar ◽  
Bhaskar Rao ◽  
Thomas Liu
Keyword(s):  
Neural Networks ◽  
State Of The Art ◽  
Feedforward Neural Networks ◽  
High Accuracy ◽  
Functional Magnetic Resonance ◽  
Resonance Imaging ◽  
Correlation Matrices ◽  
Functional Connectome ◽  
Temporal Features ◽  
Data Points

Although individual subjects can be identified with high accuracy using correlation matrices computed from resting-state functional magnetic resonance imaging (rsfMRI) data, the performance significantly degrades as the scan duration is decreased. Recurrent neural networks can achieve high accuracy with short duration (72s) data segments but are designed to use temporal features not present in the correlation matrices. Here we show that shallow feedforward neural networks that rely solely on the information in rsfMRI correlation matrices can achieve state-of-the-art identification accuracies (≥ 99.5%) with data segments as short as 20s and across a range of input data-size combinations when the total number of data points (# regions × # time points) is on the order of 10,000.

Estimating Pigment Concentrations from Spectral Images Using an Encoder‐Decoder Neural Network

2020 ◽  
Vol 64 (3) ◽  
pp. 30502-1-30502-15
Author(s):  
Kensuke Fukumoto ◽  
Norimichi Tsumura ◽  
Roy Berns
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Absorption Coefficient ◽  
Spectral Data ◽  
High Accuracy ◽  
Pigment Concentration ◽  
Scattering Coefficient ◽  
A Value ◽  
Input And Output ◽  
Pigment Concentrations

Abstract A method is proposed to estimate the concentration of pigments mixed in a painting, using the encoder‐decoder model of neural networks. The model is trained to output a value that is the same as its input, and its middle output extracts a certain feature as compressed information about the input. In this instance, the input and output are spectral data of a painting. The model is trained with pigment concentration as the middle output. A dataset containing the scattering coefficient and absorption coefficient of each of 19 pigments was used. The Kubelka‐Munk theory was applied to the coefficients to obtain many patterns of synthetic spectral data, which were used for training. The proposed method was tested using spectral images of 33 paintings, which showed that the method estimates, with high accuracy, the concentrations that have a similar spectrum of the target pigments.

Sign in / Sign up

Export Citation Format

Share Document