Integrating rush orders into existent schedules for a complex job shop problem

This paper investigates the problem of inserting new rush orders into a current schedule of a real world job shop floor. Effective rescheduling methods must achieve reasonable levels of performance, measured according to a certain cost function, while preserving the stability of the shop floor, i.e. introducing as few changes as possible to the current schedule. This paper proposes new and effective match-up strategies which modify only a part of the schedule in order to accommodate the arriving jobs. The proposed strategies are compared with other rescheduling methods such as “right shift” and “insertion in the end”, which are optimal with respect to stability but poor with respect to performance, and with “total rescheduling” which is optimal with respect to performance but poor with respect to stability. Our results and statistical analysis reveal that the match-up strategies are comparable to the “right shift” and “insertion in the end” with respect to stability and as good as “total rescheduling” with respect to performance.     

Landmine detection and classification with complex-valued hybrid neural network using scattering parameters dataset

Neural networks have been applied to landmine detection from data generated by different kinds of sensors. Real-valued neural networks have been used for detecting landmines from scattering parameters measured by ground penetrating radar (GPR) after disregarding phase information. This paper presents results using complex-valued neural networks, capable of phase-sensitive detection followed by classification. A two-layer hybrid neural network structure incorporating both supervised and unsupervised learning is proposed to detect and then classify the types of landmines. Tests are also reported on a benchmark data.     

Estimation of landmine characteristics in sandy desert using neural networks

Many places in the world are heavily contaminated with landmines, which cause that many resources are not utilized. This makes landmine detection and removal challenges for research. To guarantee reliable landmine sensing system, deep analysis and many test cases are required. The proposed concept is based on application of 1 kPa external constant pressure (lower than the landmine activation pressure) to the sand surface. The resultant contact pressure distribution is dependent on the imbedded object characteristics (type and depth). Then neural networks (NN) are trained to find the inverse solution of the sand–landmine problem. In other words, when the contact pressure is known, NN can estimate the imbedded object type and depth. In this work, using finite element modeling, the existence of landmines in sand is modeled and analyzed. The resultant contact pressure distribution for five objects (1—anti-tank, 2—anti-personnel, 3—can with diameter and height of 200 mm, 4—spherical rock with 200 mm diameter, and 5—sand without any object) in sand at different depths is used in training NN. Three NN are developed to estimate the landmine characteristics. The first one is perceptron type which classifies the introduced objects in sand. The other two feed-forward NN (FFNN) are developed to estimate the depth of two landmine types. The NN detection rates of anti-tank and anti-personnel landmines are 100 and 67 % in training, and 95 and 70 % in validation, respectively. As test cases, the detection rates of the NN in case of landmine inclination angles (0°–30°) are studied. The results show same detection rates as those at no inclination. A random noise 10 % of the average signal does not affect NN detection rates, which are the same as 95 and 70 % as in validation for anti-tank and anti-personnel, respectively, while with 20 % noise detection rates are decreases to 90 and 50 % for anti-tank and anti-personnel, respectively.

     

Complexity of identification of linear systems with rational transfer functions

We study the complexity of worst-case time-domain identification of linear time-invariant systems using model sets consisting of degree-n rational models with poles in a fixed region of the complex plane. For specific noise level δ and tolerance levels τ, the number of required output samples and the total sampling time should be as small as possible. In discrete time, using known fractional covers for certain polynomial spaces (with the same norm), we show that the complexity isO(n ²) for theH ^∞ norm,O(n) for the ℓ² norm, and exponential inn for the ℓ¹ norm, for each δ and τ. We also show that these bounds are tight. For the continuous-time case we prove analogous results, and show that the input signals may be compactly supported step functions with equally spaced nodes. We show, however, that the internodal spacing must approach 0 asn increases.     

A Comparison of State-of-the-Art Classification Techniques with Application to Cytogenetics

Several state-of-the-art techniques – a neural network, Bayesian neural network, support vector machine and naive Bayesian classifier – are experimentally evaluated in discriminating fluorescence   in situ hybridisation (FISH) signals. Highly-accurate classification of valid signals and artifacts of several cytogenetic probes (colours) is required for detecting abnormalities in FISH images. More than 3100 FISH signals are classified by each of the techniques into colour and as real or artifact with accuracies of around 98% and 88%, respectively. The results of the comparison also show a trade-off between simplicity represented by the naive Bayesian classifier, and high classification performance represented by the other techniques.     

Learning and relocation capabilities of a CAM-brain machine

Two techniques to enhance the capabilities of a CAM-brain machine are proposed. The first is a learning capability that is realized by providing a “decay register” in each neuron cell. The second is a neural network relocation capability that makes it possible to compact the evolved neural network and make room for further evolution. Both techniques operate in an extrinsic manner and are considered supplementary to the intrinsic evolutionary capability of a CAM-brain. This work was presented, in part, at the Third International Symposium on Artificial Life and Robotics, Oita, Japan, January 19–21, 1998     

Non-derivable itemsets for fast outlier detection in large high-dimensional categorical data

Detecting outliers in a dataset is an important data mining task with many applications, such as detection of credit card fraud or network intrusions. Traditional methods assume numerical data and compute pair-wise distances among points. Recently, outlier detection methods were proposed for categorical and mixed-attribute data using the concept of Frequent Itemsets (FIs). These methods face challenges when dealing with large high-dimensional data, where the number of generated FIs can be extremely large. To address this issue, we propose several outlier detection schemes inspired by the well-known condensed representation of FIs, Non-Derivable Itemsets (NDIs). Specifically, we contrast a method based on frequent NDIs, FNDI-OD, and a method based on the negative border of NDIs, NBNDI-OD, with their previously proposed FI-based counterparts. We also explore outlier detection based on Non-Almost Derivable Itemsets (NADIs), which approximate the NDIs in the data given a δ parameter. Our proposed methods use a far smaller collection of sets than the FI collection in order to compute an anomaly score for each data point. Experiments on real-life data show that, as expected, methods based on NDIs and NADIs offer substantial advantages in terms of speed and scalability over FI-based Outlier Detection method. What is significant is that NDI-based methods exhibit similar or better detection accuracy compared to the FI-based methods, which supports our claim that the NDI representation is especially well-suited for the task of detecting outliers. At the same time, the NDI approximation scheme, NADIs is shown to exhibit similar accuracy to the NDI-based method for various δ values and further runtime performance gains. Finally, we offer an in-depth discussion and experimentation regarding the trade-offs of the proposed algorithms and the choice of parameter values.     

Dealing with missing values in neural network-based diagnostic systems

Backpropagation neural networks have been applied to prediction and classification problems in many real world situations. However, a drawback of this type of neural network is that it requires a full set of input data, and real world data is seldom complete. We have investigated two ways of dealing with incomplete data — network reduction using multiple neural network classifiers, and value substitution using estimated values from predictor networks — and compared their performance with an induction method. On a thyroid disease database collected in a clinical situation, we found that the network reduction method was superior. We conclude that network reduction can be a useful method for dealing with missing values in diagnostic systems based on backpropagation neural networks.     

Practical real-time intrusion detection using machine learning approaches

The growing prevalence of network attacks is a well-known problem which can impact the availability, confidentiality, and integrity of critical information for both individuals and enterprises. In this paper, we propose a real-time intrusion detection approach using a supervised machine learning technique. Our approach is simple and efficient, and can be used with many machine learning techniques. We applied different well-known machine learning techniques to evaluate the performance of our IDS approach. Our experimental results show that the Decision Tree technique can outperform the other techniques. Therefore, we further developed a real-time intrusion detection system (RT-IDS) using the Decision Tree technique to classify on-line network data as normal or attack data. We also identified 12 essential features of network data which are relevant to detecting network attacks using the information gain as our feature selection criterions. Our RT-IDS can distinguish normal network activities from main attack types (Probe and Denial of Service (DoS)) with a detection rate higher than 98% within 2 s. We also developed a new post-processing procedure to reduce the false-alarm rate as well as increase the reliability and detection accuracy of the intrusion detection system.     

Monitoring the performance of the paper making process

This paper presents two approaches for monitoring the performance of the paper making process. The first uses a neural network vibration-based condition monitoring system for providing advance warning of press felt problems. The characteristics of this system include spectral analysis through acceleration enveloping, a peak detection algorithm as a feature extractor and a neural network classifier. The second approach makes use of multivariate statistical techniques. This involves the generation of a multivariate statistical representation of nominal process behaviour. Process malfunctions are then identified by deviations from the developed model. Both approaches are successfully applied to an industrial paper machine.     

基于拟牛顿算法神经网络的入侵检测系统的研究

近年来,神经网络技术在入侵检测中得到了广泛应用,其中最具代表性的是BP神经网络,但其本身所具有的局部极小性质限制了检测性能的提高。利用人工神经网络可以解决当前其他入侵检测方法中所遇到许多问题,有望成为异常检测中统计方法的替代品,是研制具有学习和适应能力的入侵检测系统重要手段之一。通过抽取部分混合实例以及典型攻击实例进行模式训练、测试后,在BP神经网络优化算法进行对比研究的基础上,利用拟牛顿算法对传统BP算法进行改进,从而提高入侵检测系统的收敛度,检测率。实验分析可得,在一定的训练方法基础上,基于拟牛顿算法优化神经网络和其他几种算法相比,在针对多种攻击类型上检测率有不同程度的提高。     

Database Integration Using Neural Networks: Implementation and Experiences

Applications in a wide variety of industries require access to multiple heterogeneous distributed databases. One step in heterogeneous database integration is semantic integration: identifying corresponding attributes in different databases that represent the same real world concept. The rules of semantic integration can not be ‘pre-programmed’ since the information to be accessed is heterogeneous and attribute correspondences could be fuzzy. Manually comparing all possible pairs of attributes is an unreasonably large task. We have applied artificial neural networks (ANNs) to this problem. Metadata describing attributes is automatically extracted from a database to represent their ‘signatures’. The metadata is used to train neural networks to find similar patterns of metadata describing corresponding attributes from other databases. In our system, the rules to determine corresponding attributes are discovered through machine learning. This paper describes how we applied neural network techniques in a database integration problem and how we represent an attribute with its metadata as discriminators. This paper focuses on our experiments on effectiveness of neural networks and each discriminator. We also discuss difficulties of using neural networks for this problem and our wish list for the Machine Learning community. Received 18 February 1999 / Revised 22 April 1999 / Accepted in revised form 20 November 1999     

A fuzzy vector valued KNN-algorithm for automatic outlier detection

The K nearest neighbors approach is a viable technique in time series analysis when dealing with ill-conditioned and possibly chaotic processes. Such problems are frequently encountered in, e.g., finance and production economics. More often than not, the observed processes are distorted by nonnormal disturbances, incomplete measurements, etc. that undermine the identification, estimation and performance of multivariate techniques. If outliers can be duly recognized, many crisp statistical techniques may perform adequately as such. Geno-mathematical programming provides a connection between statistical time series theory and fuzzy regression models that may be utilized e.g., in the detection of outliers. In this paper we propose a fuzzy distance measure for detecting outliers via geno-mathematical parametrization. Fuzzy KNN is connected as a linkable library to the genetic hybrid algorithm (GHA) of the author, in order to facilitate the determination of the LR-type fuzzy number for automatic outlier detection in time series data. We demonstrate that GHA[Fuzzy KNN] provides a platform for automatically detecting outliers in both simulated and real world data.     

Metamorphic worm that carries its own morphing engine

Metamorphic malware changes its internal structure across generations, but its functionality remains unchanged. Well-designed metamorphic malware will evade signature detection. Recent research has revealed techniques based on hidden Markov models (HMMs) for detecting many types of metamorphic malware, as well as techniques for evading such detection. A worm is a type of malware that actively spreads across a network to other host systems. In this project we design and implement a prototype metamorphic worm that carries its own morphing engine. This is challenging, since the morphing engine itself must be morphed across replications, which imposes restrictions on the structure of the worm. Our design employs previously developed techniques to evade detection. We provide test results to confirm that this worm effectively evades signature and HMM-based detection, and we consider possible detection strategies. This worm provides a concrete example that should prove useful for additional metamorphic detection research.     

A survey of land mine detection technology

This paper describes the state of the art in land mine detection technology and algorithms. Landmine detection is a growing concern due to the danger of buried landmines to people's lives, economic growth and development. Most of the injured people have no connection with the reason why the mines were placed. There are 50–100 million landmines in more than 80 countries around the world. Deactivation is estimated at 100 000 mines per year, against the nearly 2 million mines laid annually. In this paper we describe and analyse sensor technology available including state‐of‐the‐art technology such as ground penetrating radar (GPR), electromagnetic induction (EMI) and nuclear quadrupole resonance (NQR) among others. Robotics, data processing and algorithms are mentioned, considering support vectors, sensor fusion, neural networks, etc. Finally, we establish conclusions highlighting the need to improve not only the way images are acquired, but the way this information is processed and compared.     

Rushing out detection system for safe driving using foveated image processing

We propose a detection system for a dangerous situation while driving using foveated image processing and a neural network (NN). This system detects the situation of a man rushing out who must be avoided while driving. A foveated image is transformed by log polar mapping (LPM) from an image acquired from a camera mounted on a car. LPM imitates the primate visual system, and it is possible not only to obtain both a high central resolution and a wide field of view, but also to significantly reduce processing image data. This transformed image is suited to the detection of an object that moves toward the camera. To detect this object, we calculate flowow vectors on time-scaled images, and process them by a NN that outputs a warning signal when a dangerous situation is detected. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

An evolutionary approach for achieving scalability with general regression neural networks

In this paper, we present an approach to overcome the scalability issues associated with instance-based learners. Our system uses evolutionary computational techniques to determine the minimal set of training instances needed to achieve good classification accuracy with an instance-based learner. In this way, instance-based learners need not store all the training data available but instead store only those instances that are required for the desired accuracy. Additionally, we explore the utility of evolving the optimal feature set used by the learner for a given problem. In this way, we attempt to deal with the so-called “curse of dimensionality” associated with computational learning systems. To these ends, we introduce the Evolutionary General Regression Neural Network. This design uses an estimation of distribution algorithm to generate both the optimal training set as well as the optimal feature set for a general regression neural network. We compare its performance against a standard general regression neural network and an optimized support vector machine across four benchmark classification problems.     

Covariance-guided landmine detection and discrimination using ground-penetrating radar data

Ground penetrating Radar (GPR) can detect and deliver the response signal from any buried kind of object like plastic or metallic landmines, stones, and wood sticks. It delivers three kinds of data: Ascan, Bscan, and Cscan. However, it cannot discriminate between landmines and inoffensive objects or ‘clutter.’ One-class classification is an alternative to detect landmines, especially, as landmines features data are unbalanced. In this article, we investigate the effectiveness of the Covariance-guided One-Class Support Vector Machine (COSVM) to detect, discriminate, and locate landmines efficiently. In fact, compared to existing one-class classifiers, the COSVM has the advantage of emphasizing low variance directions. Moreover, we will compare the one-class classification to multiclass classification to tease out the advantage of the former over the latter as data are unbalanced. Our method consists of extracting Ascan GPR data. Extracted features are used as an input for COSVM to discriminate between landmines and clutter. We provide an extensive evaluation of our detection method compared to other methods based on relevant state of the art one-class and multiclass classifiers, on the well-known MACADAM database. Our experimental results show clearly the superiority of using COSVM in landmine detection and localization.     

Predicting box-office success of motion pictures with neural networks

Predicting box-office receipts of a particular motion picture has intrigued many scholars and industry leaders as a difficult and challenging problem. In this study, the use of neural networks in predicting the financial performance of a movie at the box-office before its theatrical release is explored. In our model, the forecasting problem is converted into a classification problem-rather than forecasting the point estimate of box-office receipts, a movie based on its box-office receipts in one of nine categories is classified, ranging from a ‘flop’ to a ‘blockbuster.’ Because our model is designed to predict the expected revenue range of a movie before its theatrical release, it can be used as a powerful decision aid by studios, distributors, and exhibitors. Our prediction results is presented using two performance measures: average percent success rate of classifying a movie's success exactly, or within one class of its actual performance. Comparison of our neural network to models proposed in the recent literature as well as other statistical techniques using a 10-fold cross validation methodology shows that the neural networks do a much better job of predicting in this setting.     

Customer churn prediction by hybrid neural networks

As churn management is a major task for companies to retain valuable customers, the ability to predict customer churn is necessary. In literature, neural networks have shown their applicability to churn prediction. On the other hand, hybrid data mining techniques by combining two or more techniques have been proved to provide better performances than many single techniques over a number of different domain problems. This paper considers two hybrid models by combining two different neural network techniques for churn prediction, which are back-propagation artificial neural networks (ANN) and self-organizing maps (SOM). The hybrid models are ANN combined with ANN (ANN + ANN) and SOM combined with ANN (SOM + ANN). In particular, the first technique of the two hybrid models performs the data reduction task by filtering out unrepresentative training data. Then, the outputs as representative data are used to create the prediction model based on the second technique. To evaluate the performance of these models, three different kinds of testing sets are considered. They are the general testing set and two fuzzy testing sets based on the filtered out data by the first technique of the two hybrid models, i.e. ANN and SOM, respectively. The experimental results show that the two hybrid models outperform the single neural network baseline model in terms of prediction accuracy and Types I and II errors over the three kinds of testing sets. In addition, the ANN + ANN hybrid model significantly performs better than the SOM + ANN hybrid model and the ANN baseline model.