M. Alamgir Hossain

An adaptive ensemble classifier for mining concept drifting data streams

It is challenging to use traditional data mining techniques to deal with real-time data stream classifications. Existing mining classifiers need to be updated frequently to adapt to the changes in data streams. To address this issue, in this paper we propose an adaptive ensemble approach for classification and novel class detection in concept drifting data streams. The proposed approach uses traditional mining classifiers and updates the ensemble model automatically so that it represents the most recent concepts in data streams. For novel class detection we consider the idea that data points belonging to the same class should be closer to each other and should be far apart from the data points belonging to other classes. If a data point is well separated from the existing data clusters, it is identified as a novel class instance. We tested the performance of this proposed stream classification model against that of existing mining algorithms using real benchmark datasets from UCI (University of California, Irvine) machine learning repository. The experimental results prove that our approach shows great flexibility and robustness in novel class detection in concept drifting and outperforms traditional classification models in challenging real-life data stream applications.

Test Data Generation from UML State Machine Diagrams using GAs

Automatic test data generation helps testers to validate software against user requirements more easily. Test data can be generated from many sources; for example, experience of testers, source program, or software specification. Selecting a proper test data set is a decision making task. Testers have to decide what test data that they should use, and a heuristic technique is needed to solve this problem automatically. In this paper, we propose a framework for generating test data from software specifications. The selected specification is Unified Modeling Language (UML) state machine diagram. UML state machine diagram describes a system in term of state which can be changed when there is an action occurring in the system. The generated test data is a sequence of these actions. These sequences of action help testers to know how they should test the system. The quality of generated test data is measured by the number of transitions which is fired using the test data. The more transitions test data can fire, the better quality of test data is. The number of coverage transitions is also used as a feedback for a heuristic search for a better test set. Genetic algorithms (GAs) are selected for searching the best test data. Our experimental results show that the proposed GA-based approach can work well for generating test data for some types of UML state machine diagrams.

Intelligent Traffic Control Decision Support System

When non-recurrent road traffic congestion happens, the operator of the traffic control centre has to select the most appropriate traffic control measure or combination of measures in a short time to manage the traffic network. This is a complex task, which requires expert knowledge, much experience and fast reaction. There are a large number of factors related to a traffic state as well as a large number of possible control measures that need to be considered during the decision making process. The identification of suitable control measures for a given non-recurrent traffic congestion can be tough even for experienced operators. Therefore, simulation models are used in many cases. However, simulating different traffic scenarios for a number of control measures in a complicated situation is very time-consuming. In this paper we propose an intelligent traffic control decision support system (ITC-DSS) to assist the human operator of the traffic control centre to manage online the current traffic state. The proposed system combines three soft-computing approaches, namely fuzzy logic, neural network, and genetic algorithm. These approaches form a fuzzy-neural network tool with self-organization algorithm for initializing the membership functions, a GA algorithm for identifying fuzzy rules, and the back-propagation neural network algorithm for fine tuning the system parameters. The proposed system has been tested for a case-study of a small section of the ring-road around Riyadh city. The results obtained for the case study are promising and show that the proposed approach can provide an effective support for online traffic control.

Decision support for coordinated road traffic control actions

Selection of the most appropriate traffic control actions to solve non-recurrent traffic congestion is a complex task, which requires significant expert knowledge and experience. Also, the application of a control action for solving a local traffic problem could create traffic congestion at different locations in the network because of the strong interrelations between traffic situations at different locations of a road network. Therefore, coordination of control strategies is required to make sure that all available control actions serve the same objective. In this paper, an Intelligent Traffic Control System (ITCS) based on a coordinated-agent approach is proposed to assist the human operator of a road traffic control centre to manage the current traffic state. In the proposed system, the network is divided into sub-networks, each of which has its own associated agent. The agent of the sub-network with an incident reacts with other affected agents in order to select the optimal traffic control action, so that a globally acceptable solution is found. The agent uses an effective way of calculating the control action fitness locally and globally. The capability of the proposed ITCS has been tested for a case study of a part of the traffic network in the Riyadh city of Saudi Arabia. The obtained results show its ability to identify the optimal global control action. Highlights? Identification/ranking of coordinated road traffic control actions from global view. ? Divided road sub-network with its own local DSS with fuzzy neural network engine. ? Prediction of fitness of local control action from sub-network boundary conditions. ? User adaptable factors in calculation of global performance of control actions. ? Analysis of different traffic scenarios of a case study to demonstrate DSS ability.

An intelligent multi-agent approach for road traffic management systems

Due to the strong interrelations between traffic situations at different locations of a road network the traffic control actions applied for solving a local traffic problem can create another traffic congestion at a different location in the network. This can result the average travel time on the network level, even after the application of the control actions, to be the same or worse. Therefore, coordinative control strategies are required to make sure that all available control actions serve the same objective. In this paper, an intelligent decision support system based on multi-agent approach is proposed to assist the human operator of the road traffic control centre to manage the current traffic state. In the proposed system, the total network is divided in sub-networks, each of which has its own evaluation agent. In the proposed system the agent will be able to react with other (affected) agents through a high level agent called coordinator to find the optimal global traffic control action using an intelligent traffic control. The capability of the proposed multi-agent-based decision support system was tested for a case study of a part of the traffic network in the Riyadh city of Saudi Arabia. The obtained results show the ability of the proposed multi-agent-based system to identify the optimal global control action.

Real Time Identification of Road Traffic Control Measures

The operator of a traffic control centre has to select the most appropriate traffic control action or combination of actions in a short time to manage the traffic network when non-recurrent road traffic congestion happens. This is a complex task, which requires expert knowledge, much experience and fast reaction. There are a large number of factors related to a traffic state as well as a large number of possible control actions that need to be considered during the decision making process. The identification of suitable control actions for a given non-recurrent traffic congestion can be tough even for experienced operators. Therefore, simulation models are used in many cases. However, simulating different traffic actions for a number of control measures in a complicated situation is very time-consuming. This chapter presents an intelligent method for the real-time identification of road traffic actions which assists the human operator of the traffic control centre in managing the current traffic state. The proposed system combines three soft-computing approaches, namely fuzzy logic, neural networks, and genetic algorithms. The system employs a fuzzy-neural network tool with self-organization algorithm for initializing the membership functions, a genetic algorithm (GA) for identifying fuzzy rules, and the back-propagation neural network algorithm for fine tuning the system parameters. The proposed system has been tested for a case-study of a small section of the ring-road around Riyadh city in Saudi Arabia. The results obtained for the case study are promising and demonstrate that the proposed approach can provide an effective support for real-time traffic control.

GA-based learning for rule identification in fuzzy neural networks

GA-based approach within a three stages-learning for Fuzzy Neural Network systems.GA to identify relevant rules in a promising way from all possible fuzzy rules.Performance comparison with other 19 approaches reported in the literatures. Employing an effective learning process is a critical topic in designing a fuzzy neural network, especially when expert knowledge is not available. This paper presents a genetic algorithm (GA) based learning approach for a specific type of fuzzy neural network. The proposed learning approach consists of three stages. In the first stage the membership functions of both input and output variables are initialized by determining their centers and widths using a self-organizing algorithm. The second stage employs the proposed GA based learning algorithm to identify the fuzzy rules while the final stage tunes the derived structure and parameters using a back-propagation learning algorithm. The capabilities of the proposed GA-based learning approach are evaluated using a well-examined benchmark example and its effectiveness is analyzed by means of a comparative study with other approaches. The usefulness of the proposed GA-based learning approach is also illustrated in a practical case study where it is used to predict the performance of road traffic control actions. Results from the benchmarking exercise and case study effectively demonstrate the ability of the proposed three stages learning approach to identify relevant fuzzy rules from a training data set with a higher prediction accuracy than alternative approaches.

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks

Identification of fuzzy rules is an important issue in designing of a fuzzy neural network (FNN). However, there is no systematic design procedure at present. In this paper we present a genetic algorithm (GA) based learning algorithm to make use of the known membership function to identify the fuzzy rules form a large set of all possible rules. The proposed learning algorithm initially considers all possible rules then uses the training data and the fitness function to perform rule- selection. The proposed GA based learning algorithm has been tested with two different sets of training data. The results obtained from the experiments are promising and demonstrate that the proposed GA based learning algorithm can provide a reliable mechanism for fuzzy rule selection.

Intelligent Facial Action and emotion recognition for humanoid robots

This research focuses on the development of a realtime intelligent facial emotion recognition system for a humanoid robot. In our system, Facial Action Coding System is used to guide the automatic analysis of emotional facial behaviours. The work includes both an upper and a lower facial Action Units (AU) analyser. The upper facial analyser is able to recognise six AUs including Inner and Outer Brow Raiser, Upper Lid Raiser etc, while the lower facial analyser is able to detect eleven AUs including Upper Lip Raiser, Lip Corner Puller, Chin Raiser, etc. Both of the upper and lower analysers are implemented using feedforward Neural Networks (NN). The work also further decodes six basic emotions from the recognised AUs. Two types of facial emotion recognisers are implemented, NN-based and multi-class Support Vector Machine (SVM) based. The NN-based facial emotion recogniser with the above recognised AUs as inputs performs robustly and efficiently. The Multi-class SVM with the radial basis function kernel enables the robot to outperform the NN-based emotion recogniser in real-time posed facial emotion detection tasks for diverse testing subjects.

Model reference adaptive scheme for multi-drug infusion for blood pressure control.

Using multiple interacting drugs to control both the mean arterial pressure (MAP) and cardiac output (CO) of patients with different sensitivity to drugs is a challenging task which this paper attempts to address. A multivariable model reference adaptive control (MRAC) algorithm is developed using a two-input, two-output patient model. The control objective is to maintain the homodynamic variables MAP and CO at the normal values by simultaneously administering two drugs; sodium nitroprusside (SNP) and dopamine (DPM). Computer simulations were carried out to investigate the performance of this controller. The results show that the proposed adaptive scheme is robust with respect to disturbances and variations in model parameters.