Rentian Huang

A novel hybrid artificial immune inspired approach for online break-in fraud detection

Abstract This paper proposes a new hybrid model for online fraud detection of the Video-on-Demand System, which is aimed to improve the current Risk Management Pipeline (RMP) by adding Artificial Immune System (AIS) based fraud detection for logging data. The AIS based model combines two artificial immune system algorithms with behavior based intrusion detection using Classification and Regression trees (CART). Immune inspired algorithms include the improved version of negative selection called Conserved Self Pattern Recognition Algorithm (CSPRA) and a recently established algorithm inspired by Danger Theory (DT) called Dendritic Cells Algorithm (DCA). The hybrid method based on stacking-bagging demonstrates higher detection rate lower false alarm, and handles high dimensional data set better when compared to the results achieved using only CSPRA, DCA, and CART.

On the use of innate and adaptive parts of artificial immune systems for online fraud detection

This paper describes a hybrid model for online fraud detection of the Video-on-Demand System as an E-commence application, which combines algorithms from the main two distinct viewpoints of the self, non-self theory and danger theory. Our artificial immune based algorithm includes the improved version of negative selection called Conserved Self Pattern Recognition Algorithm (CSPRA) and a recently established algorithm inspired by Danger Theory (DT) called Dendritic Cells Algorithm (DCA). The experimental results based on our Video-on-Demand case study demonstrate that the hybrid approach has a higher detection rate and lower false alarm when compared with the results achieved by only using CSPRA or DCA as individual algorithms.

Licence Plate Character Recognition Based on Support Vector Machines with Clonal Selection and Fish Swarm Algorithms

This paper proposes a new hybrid approach in Licence Plate Character Recognition (LPCR) based on Support Vector Machines (SVMs) with Clonal Selection and Fish Swarm Algorithms. The Artificial Immune Technique is used through Clonal Selection Algorithm (CSA) to dynamically select the best training data set for SVMs throughout training. The Artificial Fish Swarm Algorithm (AFSA) is for parameters optimization which including C, and for SVMs. This method has been applied in a car park monitoring system with comparison with Back Propagation Neural Networks (BPNN) and standard SVMs. The experimental results show that CSA helped SVMs reduce the size of training dataset and training time; with the parameters optimization by AFSA. Our new hybrid method has a favorable performance in terms of being more accurate and robust.

Licence Plate Character Recognition Using Artificial Immune Technique

This paper proposes the application of Artificial Immune Technique in Licence Plate Character Recognition (LPCR). The use of Clonal Selection Algorithm (CSA) is composed of two main stages: (1) dynamic training samples; and (2) a choice of the best antibodies based on the three main clonal operations of cloning, clonal mutation and clonal selection. Once memory cells are established it will output the classification results using Fuzzy K-Nearest Neighbor (KNN) approach. The performance of CSA is compared to the Back Propagation Neural Networks (BPNN) in solving a LPCR problem. The experimental results show that the Artificial Immune Technique has a favorable performance in terms of being more accurate and robust.

An Affective Interface for Conveying User Feedback

Capturing and understanding feedback received from customers in an efficient manner plays a critical role inareas such as Customer relationship management (CRM).Measures of customers’ attitude towards a product or serviceare traditionally obtained using questionnaires formatted as xpointLikert scales, with results typically displayed numericallyand sometimes presented in graphic forms. Current ‘affective’CRM solutions have focused on customer emotion detection by recognising non-verbal cues and physiological signals from customers (Input related research). However, these methods, although technologically advanced are only at a conceptual level and lack the viability and flexibility required by current CRM solutions. This paper presents a system prototype for Non-Verbal Affective User Feedback based on output related affective research. The system converts multi-criteria based estimates of a customer’s attitude towards a product or service,obtained using traditional CRM surveys, into a non-verbalrepresentation in the form of virtual facial expressions, oremoticons. Real facial expressions provide a feedback in aconcise pictorial form whose emotional content is readilyrecognisable. We focus on enabling the system to choose thoseemoticons that, with regards to their emotional content, would be unambiguously legible by users. Our approach is intended to improve the efficiency of the representation mode of CRM survey data making it visually comprehensible ‘at-a-glance’.

Towards an artificial immune system for online fraud detection

Fraud is one of the largest growing problems experienced by many organizations as well as affecting the general public. Over the past decade the use of global communications and the Internet for conducting business has increased in popularity, which has been facing the fraud threat. This paper proposes an immune inspired adaptive online fraud detection system to counter this threat. This proposed system has two layers: the innate layer that implements the idea of Dendritic Cell Analogy (DCA), and the adaptive layer that implements the Dynamic Clonal Selection Algorithm (DCSA) and the Receptor Density Algorithm (RDA). The experimental results demonstrate that our proposed hybrid approach combining innate and adaptive layers of immune system achieves the highest detection rate and the lowest false alarm rate compared with the DCA, DCSA, and RDA algorithms for Video-on-Demand system.

Electronic Fraud Detection for Video-on-Demand System Using Hybrid Immunology-Inspired Algorithms

This paper proposes an improved version of current electronic fraud detection system by using logging data sets for Video-on-Demand system. Our approach is focused on applying Artificial Immune System based fraud detection algorithm for logging data information and accounting and billing purposes. Our hybrid approach combines algorithms from innate and adaptive parts of immune system, inspired by the Self non-self theory and the Danger theory. Our research proved the possibility of combining these to perform E-fraud detection. The experimental results demonstrated that hybrid approach has higher detection rate, lower false alarm when compared with the performances achieved by traditional classification algorithms such as Decision Tree, Support Vector Machines, and Radial Basis Function Neural Networks. Our approach also outperforms AIS approaches that use Dendritic Cell Algorithm, Conserved Self Pattern Recognition Algorithm, and Clonal Selection Algorithm individually.

Application of Support Vector Machines in Financial Literacy Modelling

Financial literacy modelling is a very complicated process, which influenced by many factors such as demographics, languages, income levels, culture, age, and sex. This paper proposes a new model based on support vector machines (SVMs) to measure financial literacy of youth in the Australian society with respect to their financial knowledge of credit cards, loans and superannuation. In order to examine the feasibility of SVM, we compared it with a multi-layer back-propagation (BP) artificial neural network (ANN) model. The experiment shows that SVMs outperform the neural network model in that SVMs results show promising results and capabilities for modelling financial literacy in an efficient and robust approach. The results of training and validation have shown that the SVMs model has higher accuracy compared with the algorithm of BP ANN model. Thus SVMs can be considered as a new financial literacy modelling technique.

The Application of Artificial Immune Systems for the Prediction of Premature Delivery

One of the most challenging tasks currently facing the healthcare community is the identification of premature labour. Premature birth occurs when the baby is born before completion of the 37-week gestation period. The incomplete understanding of the physiology of the uterus and parturition means that premature labour prediction is a difficult task. The reason for this may be that the initial symptoms of preterm labour occur commonly in normal pregnancies. There is some misclassification in regard to recognizing full-term and preterm labour; approximately 20% of women who are identified as reaching full-term labour actually deliver prematurely. This paper explores the applicability of Artificial Immune System (AIS) technique as a new methodology to classify term and preterm records. Our AIS approach shows better results when compared with Neural Network, Decision Tree, and Support Vector Machines, achieving more than 92% accuracy overall.

On the Use of Intelligent Systems for the Modelling of Financial Literacy Parameters

Research has shown that more young people lack good financial literacy and make poor financial decisions. Financial literacy is not only important for individuals, but also for families, financial institutions, and the entire economy. In this paper, artificial neural networks (ANNs) and support vector machines (SVMs) are used as tools to model the financial literacy levels of young university students across Australia and three Western European countries. The goal was to ascertain the students’ level of financial knowledge in relation to the use of credit card and loan facilities based on a number of input parameters such as age, gender and educational level. Sensitivity analysis is applied to determine the relative contribution of each input parameter to the overall financial literacy model. The experiments show that ANNs and SVMs exhibit promising results and capabilities for effectively modeling financial literacy. Our findings indicate that the main determinants of young people’s level of financial literacy include educational level, length of employment, age, and credit card status – in terms of the use of credit card facilities, and gender, living status and credit card status – in terms of the use of loan facilities.