Machine Learning and Systems Engineering

Most malfunctioning spacecraft require only a minor maintenance operation, but have to be retired due to the lack of so-called On-Orbit Servicing (OOS) opportunities. There is no maintenance and repair infrastructure for space systems. Occasionally, space shuttle based servicing missions are launched, but there are no routine procedures foreseen for the individual spacecraft. The unmanned approach is to utilize the explorative possibilities of robots to dock a servicer spacecraft onto a malfunctioning target spacecraft and execute complex OOS operations, controlled from ground. Most OOS demonstration missions aim at equipping the servicing spacecraft with a high degree of autonomy. However, not all spacecraft can be serviced autonomously. Equipping the human operator on ground with the possibility of instantaneous interaction with the servicer satellite is a very beneficial capability that complements autonomous operations. This work focuses on such teleoperated space systems with a strong emphasis on multimodal feedback, i.e. human spacecraft interaction is considered, which utilizes multiple human senses through which the operator can receive output from a technical device. This work proposes a new concept for free flyer control and shows the development of an according test environment.

Intelligent service robot is becoming one of the most interesting issues in the recent Robot research. The service robot monitors its surroundings, and provides a service to meet a user’s goal. The service often becomes too complex that one single robot may not handle efficiently. In other words, a group of robots may be needed to accomplish given task(s) by collaborating each other. We can define this activity as a robot grouping, and we need to study further to make better group(s) by considering their characteristics of the each robot. But, it is difficult and no formal methods to make such a specific group from the many heterogeneous robots that are different in their functions and structures. This paper describes an intelligent service robot framework that outlines a multi-layer structure, which is suitable to make a particular group of robots to solve given task by collaborating with other robots. Simulated experimentation for grouping from the generated several heterogeneous is done by utilizing

Entropy

algorithm. And the collaboration among the robots is done by the multi-level task planning mechanism.

We present two practical path planning algorithms based on Bezier curves for autonomous vehicles operating under waypoints and corridor constraints. Bezier curves have useful properties for the trajectory generation problem. This paper describes how the algorithms apply these properties to generate the reference trajectory for vehicles to satisfy the path constraints. Both algorithms generate the piecewise-Bezier-curves path such that the curves segments are joined smoothly with

C

2

constraint which leads to continuous curvature along the path. The degree of the curves are minimized to prevent them from being numerically unstable. Additionally, we discuss the constrained optimization problem that optimizes the resulting path for a user-defined cost function.

This chapter deals with the resource-constrained project scheduling problem that belongs to NP-hard optimisation problems. There are many different heuristic strategies how to shift activities in time when resource requirements exceed their available amounts. We propose a transformation of the problem to a sequence of simpler instances of (multi)knapsack problems that do not use traditionally predefined activity priorities and enable to maximise limited resources in all time intervals given by start or end of an activity and therefore to reduce the total time.

This chapter describes a development of data logger for indoor environment. Present work concentrates to environmental parameter (temperature and humidity) and more polluted contaminants (concentration level of CO and CO

2

). In this work four channels have been used for data logger and other four channels is open to external sensor module. The data collected will be stored in the EEPROM and output can be taken in note-pad in tabular corresponding to month/date/year using graphical user interface.

The inherent variability in climate processes results in significant impacts on renewable energy production. While a number of advancements have been made over the years, the accurate energy production estimates and the corresponding long-term variability at the full wind farm remains a big challenge. At the same time, long-term energy estimates and the variability are important for financial assessment of the wind farm projects. In this chapter, a machine learning approach to model wind energy output from the wind farm is presented. A multiobjective evolutionary optimization (MOEO) method has been applied for the optimization of an Artificial Intelligence learning methodology the “Support Vector Machines” (SVM). The optimum parameter search is conducted in an intelligent manner by narrowing the desired regions of interest that avoids getting struck in local optima. The National Center for Environmental Prediction (NCEP)’s global reanalysis gridded dataset has been employed in this study. The gridded dataset for this particular application consists of four points each consisting of five variables. A 40-years, 6-hourly energy prediction time series is built using the 40-years of reanalysis data (1968-present) after training against short-term observed farm data. This is useful in understanding the long-term energy production at the farm site. The results of MOEO-SVM for the prediction of wind energy are reported along with the multiobjective trade-off curves.

This paper discusses the idea of hybriding intelligent host-based and network-based stepping stone detections (SSD) in order to increase detection accuracy. Experiments to measure the True Positive Rate (TPR) and False Positive Rate (FPR) for both Intelligent-Network SSD (I-NSSD) and Intelligent-Host SSD (I-HSSD) are conducted. In order to overcome the weaknesses observed from each approach, a Hybrid Intelligent SSD (HI-SSD) is proposed. The advantages of applying both approaches are preserved. The experiment results show that HI-SSD not only increases the TPR but at the same time also decreases the FPR. High TPR means that accuracy of the SSD approach increases and this is the main objective of the creation of HI-SSD.

The adoption of open source software (OSS) by government has been a topic of interest in recent years. National, regional, and local government are using OSS in increasing numbers, yet the adoption rate is still very low. This study considers if it is possible from an organizational perspective for small to medium-sized cities to provide services and conduct business using only OSS. We examine characteristics of municipal government that may influence the adoption of OSS for the delivery of services and to conduct city business. Three characteristics are considered to develop an understanding of city behavior with respect to OSS: capability, discipline, and cultural affinity. Each of these general characteristics contributes to the successful adoption and deployment of OSS by cities. Our goal was to determine the organizational characteristics that promote the adoption of OSS. We conducted a survey to support this study resulting in 3,316 responses representing 1,286 cities in the Unites States and Canada. We found most cities do not have the requisite characteristics to successfully adopt OSS on a comprehensive scale and most cities not currently using OSS have no future plans for OSS.

Ant colony optimization (ACO), which has been based on the feeding behavior of ants, has a powerful solution searching ability. However, since processing must be repeated many times, the computation process also requires a very long time. In this chapter, we discuss a new ACO algorithm that incorporates adaptive greedy mechanism to shorten the processing time. The proposed algorithm switches two selection techniques adaptively according to generation. In addition, the new pheromone update rules are introduced in order to control the balance of the intensification and diversification. Experiments using benchmark data prove the validity of the proposed algorithm.

A simple two-neuron model of a discrete Hopfield neural network is considered. The local stability is analyzed with the associated characteristic model. In order to study the dynamic behavior, the Pitchfork bifurcation is examined. In the case of two neurons, one necessary condition for yielding the Pitchfork bifurcation is found. In addition, the stability and direction of the Pitchfork bifurcation are determined by applying the normal form theory and the center manifold theorem.

Grammatical evolution (GE) is one of the newest among computational methods (Ryan et al. 1998; O’Neill and Ryan 2001). Basically, it is a tool used to automatically generate Backus-Naur-Form (BNF) computer programmes. The method’s evolution mechanism may be based on a standard genetic algorithm (GA). GE is very often used to solve the problem of a symbolic regression, determining a module’s own parameters (as it is also the case of other optimization problems) as well as the module structure itself. A Sum Square Error (SSE) method is usually used as the testing criterion. In this paper, however, we will present the original method, which uses a Sum epsilon Tube Error (STE) optimizing criterion. In addition, we will draw a possible parallel between the SSE and STE criteria describing the statistical properties of this new and promising minimizing method.

Soft sensor are used to infer the critical process variables that are otherwise difficult, if not impossible, to measure in broad range of engineering fields. Adaptive Neuro-Fuzzy Inference System (ANFIS) has been employed to develop successful ANFIS based inferential model that represents the dynamics of the targeted system. In addition to the structure of the model, the quality of the training as well as of the testing data also plays a crucial role in determining the performance of the soft sensor. This paper investigates the impact of data quality on the performance of an ANFIS based inferential model that is designed to estimate the average air temperature in distributed heating systems. The results of the two experiments are reported. The results show that the performance of ANFIS based sensor models is sensitive to the quality of data. The paper also discusses how to reduce the sensitivity by an improved mathematical algorithm.

In this paper, we present significant advances of the novel meccano technique for simultaneously constructing adaptive tetrahedral meshes of 3-D complex solids and their volume parametrization. Specifically, we will consider a solid whose boundary is a surface of genus zero. In this particular case, the automatic procedure is defined by a surface triangulation of the solid, a simple meccano composed by one cube and a tolerance that fixes the desired approximation of the solid surface. The main idea is based on an automatic mapping from the cube faces to the solid surface, a 3-D local refinement algorithm and a simultaneous mesh untangling and smoothing procedure. Although the initial surface triangulation can be a poor quality mesh, the meccano technique constructs high quality surface and volume adaptive meshes. Several examples show the efficiency of the proposed technique. Future possibilities of the meccano method for meshing a complex solid, whose boundary is a surface of genus greater than zero, are commented.

In this work a buck converter model for multi-domain simulations is proposed and compared with a state-of-the-art buck converter model. In the proposed model no switching events are calculated. By avoiding the computation of the switching events in power electronic models the processing time of multi-domain simulations can be decreased significantly. The proposed model calculates any operation point of the buck converter in continuous inductor current conduction mode (CICM) while considering the conduction losses and switching losses. It is possible to utilize the proposed modeling approach also for other dc-to-dc converter topologies. Laboratory test results for the validation of the proposed model are included.

The effect of the tool electrode wear on the accuracy is very important problem in the Rotating Electrical Discharge Machining (REDM). Two mathematical models of REDM are presented: the first one considers machining with the face of the end tool electrode and the second one considers EDM with the lateral side of the electrode. The software for computer simulation of EDM machining with the side and face of the electrodes has been developed. This simulation model for NC contouring EDM using rotating electrode may also be applied for tool electrode path optimization. The experimental results confirm the validity of the proposed mathematical models and the simulation software.

This article presents a new method for system identification based on dynamic neural networks using prior knowledge. A discrete chart is derived from a given signal flow chart. This discrete chart is implemented in a dynamic neural network model. The weights of the model correspond to physical parameters of the real system. Nonlinear parts of the signal flow chart are represented by nonlinear subparts of the neural network. An optimization algorithm trains the weights of the dynamic neural network model. The proposed identification approach is tested with a nonlinear two mass system.

Many robotics problems do not take the dynamics of the actuators into account in the formulation of the control solutions. The fallacy is in assuming that forces/torques can be instantaneously and accurately generated. In practice, actuator dynamics may be unknown. This paper presents a Model Reference Adaptive Controller (MRAC) for the actuators of a biped robot that mimics a human walking motion. The MRAC self-adjusts so that the actuators produce the desired torques. Lyapunov stability criterion and a rate of convergence analysis is provided. The control scheme for the biped robot is simulated on a sagittal plane to verify the MRAC scheme for the actuators. Next, the paper shows how a neural network (NN) can learn to generate its own walking gaits using successful runs from the adaptive control scheme. In this case, the NN learns to estimate and anticipate the reference commands for the gaits.

Steady state and dynamic vehicle models are derived for analyses of requirements on motors and batteries for battery electric vehicles. Vehicle level performance requirements such as driving range, maximum cruise speed, maximum gradeability, and maximum acceleration are used to analyze the requirements on motors and batteries including motor power and torque, battery weight and specific energy. MATLAB simulation tools are developed to allow validation of these vehicle level performance requirements for a given set of motor/battery.

We present mathematical modeling and numerical simulations, using the finite volume method, of a coaxial particle-laden airflow entering an expansion in a vertical pipe. An Eulerian approach is used for the gas (air) phase, which is modeled by the unsteady Favre-averaged Navier–Stokes equations. A Lagrangian model is used for the dispersed (particles) phase. The results of the simulations using three implementations of the

k

−ε turbulence model (standard, renormalization group – RNG, and realizable) are compared with measured axial profiles of the mean gas-phase velocities. The standard model achieved the best overall performance. The realizable model was unable to satisfactorily predict the radial velocity; it is also the most computationally-expensive model. The simulations using the RNG model predicted extra recirculation zones.

In this chapter, the design of robust stabilization and output tracking performance of multi-input multi-output processes with input saturations and unknown disturbance are considered. The proposed control technique is the robust anti-windup generalized predictive control (RAGPC) scheme for multivariable processes. The proposed control scheme embodies both the optimal attributes of generalized predictive control and the robust performance feature of operator-based theoretic approach. As a result, a strongly robust stable feedback control system with disturbance rejection feature and good tracking performance is achieved.

A lot of research on scheduling manufacturing systems with priority rules has been done. Most studies, however, concentrate on simplified scenarios considering only one type of resource, usually machines. In this study priority rules are applied to a more realistic scenario, in which machines and operators are dual-constrained and have a re-entrant process flow. Interdependencies of priority rules are analyzed by long-term simulation. Strength and weaknesses of various priority rule combinations are determined at different utilization levels. Further insights are gained by additionally solving static instances optimally by using a mixed integer linear program (MILP) of the production system and comparing the results with those of the priority rules.

A hybrid fault diagnosis method is proposed in this paper which is based on analytical and fuzzy logic theory. Analytical redundancy is employed by using statistical analysis. Fuzzy logic is then used to maximize the signal- to-threshold ratio of the residual and to detect different faults. The method was successfully demonstrated experimentally on hydraulic actuated system test rig. Real data and simulation results have shown that the sensitivity of the residual to the faults is maximized, while that to the unknown input is minimized. The decision of whether ‘a fault has occurred or not?’ is upgraded to ‘what is the severity of that fault?’ at the output. Simulation results show that fuzzy logic is more sensitive and informative regarding the fault condition, and less sensitive to uncertainties and disturbances.

A multigrid finite volume method has been developed to accelerate the solution process for simulating complex biological transport phenomena, involving convection, diffusion, and reaction. The method has been applied to a computational model which includes media flow in a cell-lined cylindrical vessel and fibroblast growth factor-2 (FGF-2) within the fluid capable of binding to receptors and heparan sulfate proteoglycans (HSPGs) on the cell surface. The differential equations were discretized by the finite volume method and solved with the multigrid V-cycle algorithm. Our work indicates that the multigrid finite volume method may allow users to investigate complex biological systems with less CPU time.

Mineral processes are multi-variable, power-intensive and strongly coupled with large delay and nonlinearities. The properties of controllability, observability and theory of minimal realization for linear systems are well understood and have been very useful in analyzing such systems. This paper deals with analogous questions for nonlinear systems with application to mineral processing. A method that can control and provide accurate prediction of optimum milling condition and power consumption, water and chemical additive requirement is developed for mineral plants operation. A fuzzy mining algorithm is proposed for extracting implicit generalized knowledge on grading process performance as qualitative values. It integrates fuzzy-set concepts and generalized data mining technologies to achieve this purpose. Using a generalized similarity transformation for the error dynamics, simulation results show that under boundedness condition the proposed approach guarantees the global exponential convergence of the error estimation. Although the nominal performance of the process is improved, the robust stability still is not guaranteed to fully avoid the mill plugging.

In this chapter we present the development of a combined cycle power plant simulator for operator’s training. This simulator has been designed and developed by the Simulation Department of the Instituto de Investigaciones Eléctricas. This is one of the several technological developments carried out by the Simulation Department, and it is part of a full scale simulators group that belongs to the Comisión Federal de Electricidad that offers the electrical service in the whole country. The simulator is currently under testing and evaluation by the final user, before entering on service at the National Center for Operator’s Training and Qualification. Tendencies of these development and impact within the operators’ scope as well as some results and future works are also presented.

This paper deals with the problem of texture-features-extraction in digital mammograms using non-Shannon measures of entropy. Texture-features-extraction is normally achieved using statistical texture-analysis method based on gray-level histogram moments. Entropy is important texture feature to measure the randomness of intensity distribution in a digital image. Generally, Shannon’s measure of entropy is employed in various feature-descriptors implemented so far. These feature-descriptors are used for the purpose of making a distinction between normal and abnormal regions in mammograms. As non-Shannon entropies have a higher dynamic range than Shannon’s entropy covering much wider range of scattering conditions, they are more useful in estimating scatter density and regularity. Based on these considerations, an attempt is made to develop a new type of feature-descriptor using non-Shannon’s measures of entropy for classifying normal and abnormal mammograms. Experiments are conducted on images of mini-MIAS (Mammogram Image Analysis Society) database to examine its effectiveness. The results of this study are quite promising for extending the work towards the development of a complete Computer Aided Diagnosis (CAD) system for early detection of breast cancer.

Direction-of-arrival (DOA) estimation is an important algorithm in array processing. Most traditional DOA estimation methods focus on narrow-band sources. A new wideband DOA estimation approach based on arbitrary group delay is proposed in this paper. In the proposed algorithm, echo gain at each direction is calculated based on digital group delay compensation. Digital group delay can realize time delay precisely, and has no restriction on delay step. The proposed method is able to operate arbitrary waveform and is suitable for linear frequency modulation signals, nonlinear frequency modulation signals, and even carrier free signals. Simulations with Gaussian function are carried out to show the effectiveness of the proposed algorithm.

In this paper, we propose a “Spatial speaker”, an enhancement of a Java FreeTTS speech synthesizer with the additional function of spatial positioning of both the speaker and the listener. Our module enables the reading of an arbitrary text from a file or webpage to the user from a fixed or changing position in space through normal stereo headphones. Our solution combines the following modules: FreeTTS speech synthesizer, a custom made speech processing unit, MIT Media Lab HRTF library, JOAL positioning library and Creative X-Fi sound card. The paper gives an overview of the design of the “Spatial Speaker” and proposes three different practical applications of such a system for visually impaired and blind computer users. Some preliminary results of user studies confirmed the system’s usability and showed its great potential also in other types of applications and auditory interfaces. The entire system is developed as a single Java class which can be imported and used in any Java application.

This chapter presents a novel approach for automatic annotation and content based video retrieval by making use of the features extracted during the process of detecting commercial boundaries in a recorded Television (TV) program. In this approach, commercial boundaries are primarily detected using audio and the detected boundaries are validated and enhanced using splash screen of a program in the video domain. Detected splash screen of a program at the commercial boundaries is used for automatic annotation of recorded video which helps in fast content based video retrieval. The performance and validity of our approach is demonstrated using the videos recorded from different Indian Television broadcasts.

In the recent years, there has been an increasing demand for streaming high quality multimedia content over wireless networks. Demand for triple play services (voice, video, data) by a number of simultaneously communicating users often results in lack of acceptable multimedia quality. In addition, the large transmission distance and the limited battery power of the hand-held wireless device serves as a major bottleneck over transmitting video directly from the network operator to the end-users. There has been no successful mechanism developed till date that would provide live video streaming over wireless networks. In this paper,

clusterDAM

– a novel cluster-based architecture is proposed for

delivering adaptive multimedia

content over two-hop wireless networks. An intermediate relay node serves as a proxy-client-server between the base station and the mobile users which ensures that the multimedia delivery is adapted in real-time according to the channel conditions and the quality of the multimedia content. An extensive simulation-based analysis with different kinds of network traffic and protocols indicate that the cluster-DAM architecture is significantly superior to the traditional single-hop design, not only in terms of the perceived video quality, but also in terms of the loss rate and the average bit rate.

Many different phenomena, arising from scientific, technical or social areas, can be modeled by a system depending on a certain number

n

of random Boolean variables. The so-called complex stochastic Boolean systems (CSBSs) are characterized by the ordering between the occurrence probabilities Pr{

u

} of the 2

n

associated binary strings of length

n

, i.e.,

u

=(

u

1

,

…

,

u

n

) ∈ {0,1}

n

. The intrinsic order defined on {0,1}

n

provides us with a simple positional criterion for ordering the binary

n

-tuple probabilities without computing them, simply looking at the relative positions of their 0s and 1s. For every given binary

n

-tuple

u

, this paper presents two simple formulas – based on the positions of the 1-bits (0-bits, respectively) in

u

– for counting (and also for rapidly generating, if desired) all the binary

n

-tuples

v

whose occurrence probabilities Pr{

v

} are always less than or equal to (greater than or equal to, respectively) Pr{

u

}. Then, from these formulas, we determine the closed interval covering all possible values of the rank (position) of

u

in the list of all binary

n

-tuples arranged by decreasing order of their occurrence probabilities. Further, the length of this so-called ranking interval for

u

, also provides the number of binary

n

-tuples

v

incomparable by intrinsic order with

u

. Results are illustrated with the intrinsic order graph, i.e., the Hasse diagram of the partial intrinsic order.

The recent years, phase classification has frequently been discussed as a method to guide scheduling, compiler optimizations and program simulations. In this chapter, I introduce a new classification method called

Setvectors

. I show that the new method outperforms classification accuracy of state-of-the-art methods by approximately 6–25%, while it has about the same computational complexity as the fastest known methods. Additionally, I introduce a new method called

PoEC

(Percentage of Equal Clustering) to objectively compare phase classification techniques.

The efficiency of a cryptographic algorithm is based on its time taken for encryption and decryption and the way it produces different cipher text from a plain text. The RSA, the widely used public key algorithm and other public key algorithms may not guarantee that the cipher text is fully secured. As an alternative approach to handling ASCII value of the text in the cryptosystem, a magic square of order 16 equivalent to ASCII set generated from two different approaches is thought of in this work. It attempts to enhance the efficiency by providing add-on security to the cryptosystem. Here, encryption/decryption is based on numerals generated by magic square rather than ASCII values. This proposed work provides another layer of security to any public key algorithms. Since, this model is acting as a wrapper to a public-key algorithm, it ensures that the security is enhanced. Further, this approach is experimented in a simulated environment with 2, 4, and 8 processor model using Maui scheduler which is based on back filling philosophy.

With the emergence of grid environments featuring dynamic resources and varying user profiles, there is an increasing need to develop reliable tools that can effectively coordinate the requirements of an application with available computing resources. The ability to predict the behavior of complex aggregated systems under dynamically changing workloads is particularly desirable, leading to effective resource usage and optimization of networked systems. To explore these issues, some economic/market based models have been introduced in the literature, where users, external schedulers, and local schedulers negotiate to optimize their objectives. In this chapter, we have proposed a mathematical model for Grid resource allocation with aims to minimize the cost of the grid users when multiple grid users are trying to utilize the grid system at the same time.

IPv6 adds many improvements to IPv4 in areas such as address space, built-in security, quality of service, routing and network auto-configuration. IPv6 nodes use the Neighbor Discovery (ND) protocol to discover other nodes on the link, to determine their link-layer addresses, to find routers, to detect duplicate address, and to maintain reachability information about the paths to active neighbors. ND is vulnerable to various attacks when it is not secured. The original specifications of ND called for the use of IPsec as a security mechanism to protect ND messages. However, its use is impractical due to the very large number of manually configured security associations needed for protecting ND. For this reason, the Secure Neighbor Discovery Protocol (SEND) was proposed. In this work, we present Easy-SEND, an open source implementation of SEND that can be used in production environment or as a didactic application for the teaching and learning of the SEND protocol. Easy-SEND is easy to install and use, and it has an event logger that can help network administrators to troubleshoot problems or students in their studies. It also includes a tool to generate and verify Cryptographically Generated Addresses (CGA) that are used with SEND.

Nowadays, there are a wide variety of network benchmark tools, giving researchers and network administrators many options to work with. However, this variety tends to hinder the selection process of the appropriate tool. Furthermore, sometimes users are forced to try several tools in order to find one that calculates a desired gauge, so they have to learn how to manipulate different tools and how to interpret the obtained results. This research offers a compilation of network benchmark tools currently used, with the purpose of guiding the selection of one tool over the others, by outlining their main features, strengths and weaknesses.

This chapter discusses the continuous effort to explore stock price and trend prediction from finance perspective as well as from the integration of three major research areas namely data mining, artificial intelligence and decision support. These areas have been explored to design a hybrid stock price prediction model with relevant techniques into the stock price analysis and prediction activities.

A primary challenge in mobile ad hoc network scenario is the dynamic differentiation of provided levels of Quality of Service (QoS) depending on client characteristics and current resource availability. In this context the paper focuses on characterizing the average end-to-end delay and maximum achievable per-node throughput for In-vehicle ad hoc multimedia network with stationary and mobile nodes. This work considers an approximation for the expected packet delay based on the assumption that the traffic at each link acts as an independent M/D/1 queue. The chapter establishes the network model, throughput model and delay model for ad hoc multimedia network. It further presents analysis and evaluation of the delay model based on an experimental multimedia communication scenario.

In most of the public-key cryptosystems like RSA, ElGamal, etc.; modular exponentiation plays a vital role for performing encryption/decryption operations. In other public-key cryptosystems like ECC, scalar point multiplication, kP where k is an arbitrary integer in the range 1 < k < ord(P) and P is a point in the elliptic curve is the central operation. In cryptographic algorithms, exponent is always an integer and can be performed faster than the traditional square and multiply method by iteratively reducing the small gain may be made if the numbers of multiplications are organized properly. For that some integer can be represented in the form of sum of squares and based on the sum of squares larger exponent e can be reduced into smaller one. Then, the addition chain is used to minimize the number of multiplications in the smaller exponent to speed up the operations. Similarly, in the case of ECC to speed up kP fuzzy modular arithmetic is considered.

Most of the methods that generate decision trees for a specific problem use examples of data instances in the decision tree generation process. This chapter proposes a method called “

RBDT-1

” - rule based decision tree - for learning a decision tree from a set of decision rules that cover the data instances rather than from the data instances themselves.

RBDT-1

method uses a set of declarative rules as an input for generating a decision tree. The method’s goal is to create on-demand a short and accurate decision tree from a stable or dynamically changing set of rules. We conduct a comparative study of

RBDT-1

with existing decision tree methods based on different problems. The outcome of the study shows that in terms of tree complexity (number of nodes and leaves in the decision tree)

RBDT-1

compares favorably to

AQDT-1, AQDT-2

which are methods that create decision trees from rules.

RBDT-1

compares favorably also to

ID3

which is a famous method that generates decision trees from data examples. Experiments show that the classification accuracies of the different decision trees produced by the different methods under comparison are equal.

Regenerative medicine is the application of tissue, sciences, engineering, computations, related biological, and biochemical principles that restore the structure and function of damaged tissues and organs. This new field encompasses many novel approaches to treatment of disease and restoration of biological function. Scientists are one-step closer to create a gene therapy/stem cell combination to combat genetic diseases. This research may lead to not only curing the disease, but also repairing the damage left behind. However, the development of gene therapy vectors with sufficient targeting ability, efficiency, and safety must be achieved before gene therapy can be routinely used in human. Delivery systems based on both viruses and non-viral systems are being explored, characterized, and used for in vitro and in vivo gene delivery. Although advances in gene transfer technology have been made, an ideal system has not yet been constructed. The development of scalable computer systems constitutes one-step toward understanding dynamics and potential of this process. Therefore, the primary goal of this work is to develop a computer model that will support investigations of both viral and non-viral methods for gene transfer on regenerative tissues including genetically modified stem cells. Different simulation scenarios were implemented, and their results were encouraging compared to ex-vivo experiments, where, the error rate lies in the range of acceptable values in this domain of application

We made our own detrended fluctuation analysis (DFA) program. We applied it to the cardio-vascular system for checking characteristics of the heartbeat of various individuals. Healthy subjects showed a normal scaling exponent, which is near 1.0. This is in agreement with the original report by Peng et al. published in 1990s. We found that the healthy exponents span from 0.9 to 1.19 in our temporary guideline based on our own experiments. In the present study, we investigated the persons who have an extra-systole heartbeat, so called as premature ventricular contractions (PVCs), and revealed that their arrhythmic heartbeat exhibited a low scaling exponent approaching to 0.7 with no exceptions. In addition, alternans, which is the heartbeats in period-2 rhythms, and which is also called as the harbinger of death, exhibited a low scaling exponent like the PVCs. We may conclude that if it would be possible to make a device that equips a DFA program, it might be useful to check the heart condition, and contribute not only in statistical physics but also in biomedical engineering and clinical practice fields; especially for health check. The device is applicable for people who are spending an ordinary life, before they get seriously heart sick.

We are concerned by imminent future problems caused by biological dangers, here we think of a way to solve them. One of them is analyzing endemic models, for this we make a study supported by Computer Algebra Systems (CAS) and Mechanized Reasoning (MR). Also we show the advantages of the use of “CAS” and “MR” to obtain in that case, an epidemic threshold theorem. We prove a previously obtained theorem for S

n

IR endemic model. Moreover using “CAS+MR” we obtain a new epidemic threshold theorem for the S

n

I

m

R epidemic model and for the staged progressive SI

m

R model. Finally we discuss the relevance of the theorems and some future applications.

Main objective of this pilot study was to present a method to convenient monitoring of detailed ambulatory movements in daily life, by use of a portable measurement device employing single tri-axial accelerometer. In addition, the purpose of this review article is to provide researchers with a guide to understanding some commonly-used accelerometers in physical activity assessment. Specially, we implemented a small-size wearable data storing system in real time that we used Micro SD-Memory card for convenient and long period habitual physical activity monitoring during daily life. Activity recognition on these features was performed using Fuzzy c means classification algorithm recognized standing, sitting, lying, walking and running with 99.5% accuracy. This study was pilot test for our developed system’s feasibilities. Further application of the present technique may be helpful in the health promotion of both young and elderly.

In this paper, we propose a simulation model to study the protein folding problem. We describe the main properties of proteins and describe the protein folding problem according to the existing approaches. Then, we propose to simulate the folding process when a protein is represented by an amino acid interaction network. This is a graph whose vertices are the proteins amino acids and whose edges are the interactions between them. We propose a genetic algorithm of reconstructing the graph of interactions between secondary structure elements which describe the structural motifs. The performance of our algorithms is validated experimentally.

This research analyses and discusses the use of Multiobjective fitness function to evolve Finite State Automata. Such automata can describe system’s behavior mathematically in an efficient manner. However system’s behavior must highly depend on its input-output specifications. Genetic Programming is used, and the fitness function is built to guide the evolutionary process in two different cases. First case: Single point fitness function is used where the only focus is on the correctness of the evolved automata. Second case: multiobjective fitness function is used since every real-world problem involves simultaneous optimization of several incommensurable and often competing objectives. Multiobjective optimization is defined as a problem of finding a Finite State Automata which satisfies: parsimony, efficiency, and correctness. It has been presented that for large and complex problems it is necessary to divide them into sub problem(s) and simultaneously breed both sub-program(s) and a calling program.