Computational Intelligence Methods

The rapid development of internet, cloud computing and sensor networks lead to develop and deploy the Internet of Things (IoT) which is a hot topic for the researchers. It has started to be used in various areas. Thus, agriculture is one of the most popular IoT research area. In agriculture environment, farming platform area is being a huge open structure and farmers must protect the crops from extreme weather conditions namely; wind speed/direction, precipitation, air temperature, solar radiations, and relative humidity etc. These extreme weather conditions effect crops and farms very significantly. But with the benefits of Internet of Things technologies, an agriculture business become more easy and efficient despite extreme weather conditions. This paper provides a model of smart agriculture environment using neural networks that helps the farmers to make more accurate predictions for the future according to weather conditions. This paper proposed a time-delay radial basis function (TDRBF) network approach to model temporal and sequential relationship between the various weather condition sensor readings from the agricultural environment. The performance of the acquired network model was analysed statistically and presented in this paper. As a result, the results of the neural network model show that it could be used to predict the desired weather condition sensor readings beforehand in order to increase the productivity in agricultural environment and also it is possible that by using such an intelligent learning system could provide a life-long learning for the changing weather conditions in the farming area over the years.

The rapid development of internet, cloud computing and sensor networks lead to develop and deploy the Internet of Things (IoT) which is a hot topic for the researchers. It has started to be used in various areas. Thus, agriculture is one of the most popular IoT research area. In agriculture environment, farming platform area is being a huge open structure and farmers must protect the crops from extreme weather conditions namely; wind speed/direction, precipitation, air temperature, solar radiations, and relative humidity etc. These extreme weather conditions effect crops and farms very significantly. But with the benefits of Internet of Things technologies, an agriculture business become more easy and efficient despite extreme weather conditions. This paper provides a model of smart agriculture environment using neural networks that helps the farmers to make more accurate predictions for the future according to weather conditions. This paper proposed a time-delay radial basis functi...

Because characterization of ink pens on documents represents an important forensic discipline, the need of legal systems for faster, in-situ and more accurate technique that identify and differentiate questioned documents involved in criminal or legal matters is of great interest. Laser-induced breakdown spectroscopy (LIBS) was experimentally proved to be a promising technique emerging in the field of forensic applications, specifically, questioned documents despite being from the same source or different source. However, the call for newly improved data analysis is unlimited and the need to know if the used ink has the same principal components or not. In the present work, the application of PCA has been demonstrated as statistics-based spectral analysis of the output resultant LIBS spectra for the forensic questioned documents analysis. The LIBS spectra were acquired from IR laser induced plasmas of black ink on regular document paper of ten brands of black gel inks commercially available in the Egyptian market. The idea behind principal component analysis is based on the loadings for the first three principal components. The obtained results have been shown to characterize ink pens used in the analysis. However, the first principal component encompasses the dominant characterization for the used pens. The results of this study indicate that Laser Induced Breakdown Spectroscopy (LIBS) of ink pens aided with PCA technique can promote the LIBS capability to faster, error-free and automatic elemental identification and differentiation for in-situ forensic purposes, specifically, questioned documents.

The scope of this study is the evaluation of early detection methods for cavitation phenomena in centrifugal irrigation pumps by analyzing the produced vibration and sound signals from a low-cost sensor and data acquisition system and comparing several computational methods. Vibration data was acquired using the embedded accelerometer sensor of a smartphone device. Sound signals were obtained using the embedded microphone of the same commercial smartphone. The analysis was based on comparing the signals in different operating conditions with reference to the best efficiency operating point of the pump. In the case of vibrations, data was acquired for all three directional axes. The signals were processed by computational methods to extract the relative features in the frequency domain and use them to train an artificial neural network to be able to identify the different pump operating conditions while the cavitation phenomenon evolves. Three different classification algorithms were...

The scope of this study is the evaluation of early detection methods for cavitation phenomena in centrifugal irrigation pumps by analyzing the produced vibration and sound signals from a low-cost sensor and data acquisition system and comparing several computational methods. Vibration data was acquired using the embedded accelerometer sensor of a smartphone device. Sound signals were obtained using the embedded microphone of the same commercial smartphone. The analysis was based on comparing the signals in different operating conditions with reference to the best efficiency operating point of the pump. In the case of vibrations, data was acquired for all three directional axes. The signals were processed by computational methods to extract the relative features in the frequency domain and use them to train an artificial neural network to be able to identify the different pump operating conditions while the cavitation phenomenon evolves. Three different classification algorithms were...

In this study, a modern adaptive signal processing technique called variational mode decomposition (VMD) has been used for power quality (PQ) events detection. Numerous single, as well as multiple PQ events, are simulated according to IEEE std. 1159-2009 and laboratory experimental signals are collected and passed through the VMD algorithm. VMD decomposes the signal into different modes and from these modes, different features have been extracted. To reduce the dimension of the feature set Fischer linear discriminant analysis (FDA) has been used. As a new contribution to the literature, VMD aided FDAbased feature selection with reduced kernel extreme learning machine technique has been used for detection and classification of multiple PQ disturbances. The performance of the proposed combined technique shows higher classification accuracy while classifying multiple PQ disturbances and the results are comparable with many existing methods.

Images in biomedical articles are often referenced for clinical decision support, educational purposes, and medical research. Authors-marked annotations such as text labels and symbols overlaid on these images are used to highlight regions of interest which are then referenced in the caption text or figure citations in the articles. Detecting and recognizing such symbols is valuable for improving biomedical information retrieval. In this research, image processing and computational intelligence methods are integrated for object segmentation and discrimination and applied to the problem of detecting arrows on these images. Evolving Artificial Neural Networks (EANNs) and Evolving Artificial Neural Network Ensembles (EANNEs) computational intelligence-based algorithms are developed to recognize overlays, specifically arrows, in medical images. For these discrimination techniques, EANNs use particle swarm optimization and genetic algorithm for artificial neural network (ANN) training, a...

Classification of normal sinus rhythm (NSR), paroxysmal atrial fibrillation (PAF), and persistent atrial fibrillation (AF) is crucial in order to diagnose and effectively plan treatment for patients. Current classification models were primarily developed by electrocardiogram (ECG) signal databases, which may be unsuitable for local patients. Therefore, this research collected ECG signals from 60 local Thai patients (age 52.53 ± 23.92) to create a classification model. The coefficient of variance (CV), the median absolute deviation (MAD), and the root mean square of the successive differences (RMSSD) are ordinary feature variables of RR irregularities used by existing models. The square of average variation (SAV) is a newly proposed feature that extracts from the irregularity of RR intervals. All variables were found to be statistically different using ANOVA tests and Tukey’s method with a p-value less than 0.05. The methods of artificial neural network (ANN) and adaptive neuro-fuzzy...

Civil and military structures are susceptible and vulnerable to damage due to the environmental and operational conditions. Therefore, the implementation of technology to provide robust solutions in damage identification (by using signals acquired directly from the structure) is a requirement to reduce operational and maintenance costs. In this sense, the use of sensors permanently attached to the structures has demonstrated a great versatility and benefit since the inspection system can be automated. This automation is carried out with signal processing tasks with the aim of a pattern recognition analysis. This work presents the detailed description of a structural health monitoring (SHM) system based on the use of a piezoelectric (PZT) active system. The SHM system includes: (i) the use of a piezoelectric sensor network to excite the structure and collect the measured dynamic response, in several actuation phases; (ii) data organization; (iii) advanced signal processing techniques to define the feature vectors; and finally; (iv) the nearest neighbor algorithm as a machine learning approach to classify different kinds of damage. A description of the experimental setup, the experimental validation and a discussion of the results from two different structures are included and analyzed.

In this research, there are three models of Artificial Neural Network (ANN), multi-layer feed-forward (MLFF), generalized regression neural network (GRNN), and radial basis (RB), are used for the prediction of the octane number of bio-gasoline. The models provide a solution for any ...

Biometrics is based on the analysis of biometric features and is usually used for person recognition. In this article we analyze biometric features related to touch screen gestures. The proposed approach to person identification has a great application potential due to the popularity of mobile devices, so research in this area will be an important direction of technological development. The aims of this research are to design and build a data acquisition system to collect surveys resulting from the execution of single finger gestures on a mobile device screen and to propose data preprocessing procedure and to indicate the best classification method for person recognition based on these surveys. The experiment with fifty persons, nine gestures and ten repetitions was performed, so 4 500 surveys acquired and analyzed within the developed system. Depending on the classification method the accuracy of the task of gesture-based person recognition was in the range from 96.60% to 99.29%. All the patterns, which achieved the highest performance in terms of the typical performance measures consist of a single connected movement of a finger over the touch screen. The patterns which include multiple disconnected segments and/or taps are less successful in the task of recognizing people.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The managers’ strategic behavior of commercial real estate in a competitive environment is discussed. The means of this research are methods of mathematical and experimental game theory. Illustrates an example of a decision-theoretic game problem with two players engaged in the management of competing shopping centers. The strategies of these players are determined by the costs of the development. Since the number of visitors is constantly and changing the quality of shopping centers, managers actually pull visitors from each other, the game-theoretical formulation of the problem of management of shopping centers corresponds to the zero-sum game. The model example shows that with the help of mathematical modeling it is possible to determine the theoretical expected behavior of agents. The model of the software module created specifically for experimental research is described. The experiments are designed to compare the theoretically predicted behavior with the real actions of...

The air quality of the living area influences human health to a certain extent. Therefore, it is particularly important to detect the quality of indoor air. However, traditional detection methods mainly depend on chemical analysis, which has long been criticized for its high time cost. In this research, a rapid air detection method for the indoor environment using laser-induced breakdown spectroscopy (LIBS) and machine learning was proposed. Four common scenes were simulated, including burning carbon, burning incense, spraying perfume and hot shower which often led to indoor air quality changes. Two steps of spectral measurements and algorithm analysis were used in the experiment. Moreover, the proposed method was found to be effective in distinguishing different kinds of aerosols and presenting sensitivity to the air compositions. In this paper, the signal was isolated by the forest, so the singular values were filtered out. Meanwhile, the spectra of different scenarios were analyz...

The recognition of the head movements is a challenging task in the Human-Computer Interface domain. The medical, automotive, or computer games domains are only several fields where this task can find practical applicabilities. Currently, the head movement recognition is performed using complex systems based on video information or using an IMU sensor with nine freedom degrees. In this paper, we describe a new approach for recognizing head movements using a new type of IMU sensor with six freedom degrees placed on top of a headphone pair. The system aims to provide an easy control method for people suffering from tetraplegia for a specific set of activities. The system collects data from the inertial sensor placed on top of the headphone to analyze and then extract the features for head movement recognition. We did construct and evaluated eight predictive models of classifying head movements activity to determine which one is the best fit for the proposed head movement recognition sy...

Biometrics is based on the analysis of biometric features and is usually used for person recognition. In this article we analyze biometric features related to touch screen gestures. The proposed approach to person identification has a great application potential due to the popularity of mobile devices, so research in this area will be an important direction of technological development. The aims of this research are to design and build a data acquisition system to collect surveys resulting from the execution of single finger gestures on a mobile device screen and to propose data preprocessing procedure and to indicate the best classification method for person recognition based on these surveys. The experiment with fifty persons, nine gestures and ten repetitions was performed, so 4 500 surveys acquired and analyzed within the developed system. Depending on the classification method the accuracy of the task of gesture-based person recognition was in the range from 96.60% to 99.29%. All the patterns, which achieved the highest performance in terms of the typical performance measures consist of a single connected movement of a finger over the touch screen. The patterns which include multiple disconnected segments and/or taps are less successful in the task of recognizing people.

In this research, there are three models of Artificial Neural Network (ANN), multi-layer feed-forward (MLFF), generalized regression neural network (GRNN), and radial basis (RB), are used for the prediction of the octane number of bio-gasoline. The models provide a solution for any ...

The paper deals with the analysis of synchronous generator protection for loss of field condition. The effect of loss of field on the generator and on the system is investigated. The proposed scheme provides the loss of field protection of generator based on simple measurements at the generator terminals instead of direct detection of loss of field. To this end generator terminal quantities are used. For feature extraction discrete wavelet transform (DWT) is applied to the obtained signals. Wavelet energy and wavelet entropy of generator terminal voltage, current, active power output and reactive power output are used as features for loss of field detection. Thereafter, support vector machine (SVM) is used for the classification of loss of field. The investigations suggests that SVM based technique is robust and capable of maintaining its performance under different loading conditions. For obtaining test signal for loss of field a power system model consisting of generator is simula...

The present work focuses on the development of a fast and cost effective method based on Laser Induced Breakdown Spectroscopy (LIBS) to the quality control, traceability and detection of adulteration in milk. Two adulteration cases have been studied; a qualitative analysis for the discrimination between different milk blends and quantification of melamine in adulterated toddler milk powder. Principal Component Analysis (PCA) and Neural Networks (NN) have been used to analyze LIBS spectra obtaining a correct classification rate of 98% with a 100% of robustness. For the quantification of melamine, two methodologies have been developed; univariate analysis using CN emission band and multivariate calibration NN model obtaining correlation coefficient (R 2) values of 0.982 and 0.999 respectively. The results of the use of LIBS technique coupled with chemometric analysis are discussed in terms of its potential use in the food industry to perform the quality control of this dairy product.

Modern computer systems practically cannot function without a computer network. New concepts of data transmission are emerging, e.g., programmable networks. However, the development of computer networks entails the need for development in one more aspect, i.e., the quality of the data transmission through the network. The data transmission quality can be described using parameters, i.e., delay, bandwidth, packet loss ratio and jitter. On the basis of the obtained values, specialists are able to state how measured parameters impact on the overall quality of the provided service. Unfortunately, for a non-expert user, understanding of these parameters can be too complex. Hence, the problem of translation of the parameters describing the transmission quality appears understandable to the user. This article presents the concept of using Machine Learning (ML) to solve the above-mentioned problem, i.e., a dynamic classification of the measured parameters describing the transmission quality...

The monitoring of hydraulic system condition change during its exploitation ran its complex problem. The main task is to identify early phase damage of hydraulic system elements (pumps, valves, ect.) in order to take decision which can avoid hydraulic system break down. This paper presents the possibility of phase trajectories use in detecting output pressure change of hydraulic system caused by positive-displacement pump wear. In hydraulic drive systems the positive-displacement pumps are one of the important elements. Proper work of these elements causes proper work of whole hydraulic system. The wear of pump elements causes pump's pressure lost and increase of volumetric losses, which lead to decreasing delivery of the pump and increasing in vibration and noise of its work. The run of vibroacustic diagnosis of pump is mainly guide to search: damage symptoms in vibration signals. In the case of high noise level and high level of mechanical complexity, the estimations of damage...

The ability to rapidly conduct in-situ chemical analysis of multiple samples of soil and other geological materials in the field offers many advantages over a traditional approach that involves collecting samples for subsequent examination in the laboratory. This study explores the application of complementary spectroscopic analyzers and a data fusion methodology for the classification/discrimination of >100 soil samples from sites across the United States. Commercially available, handheld analyzers for X-ray fluorescence spectroscopy (XRFS), Raman spectroscopy (RS), and laser-induced breakdown spectroscopy (LIBS) were used to collect data both in the laboratory and in the field. Following a common data pre-processing protocol, principal component analysis (PCA) and partial least squares discriminant analysis (PLSDA) were used to build classification models. The features generated by PLSDA were then used in a hierarchical classification approach to assess the relative advantage o...

Laser-induced breakdown spectroscopy (LIBS) is an important analysis technique with applications in many industrial branches and fields of scientific research. Nowadays, the advantages of LIBS are impaired by the main drawback in the interpretation of obtained spectra and identification of observed spectral lines. This procedure is highly time-consuming since it is essentially based on the comparison of lines present in the spectrum with the literature database. This paper proposes the use of various computational intelligence methods to develop a reliable and fast classification of quasi-destructively acquired LIBS spectra into a set of predefined classes. We focus on a specific problem of classification of paper-ink samples into 30 separate, predefined classes. For each of 30 classes (10 pens of each of 5 ink types combined with 10 sheets of 5 paper types plus empty pages), 100 LIBS spectra are collected. Four variants of preprocessing, seven classifiers (decision trees, random forest, k-nearest neighbor, support vector machine, probabilistic neural network, multi-layer perceptron, and generalized regression neural network), 5-fold stratified cross-validation, and a test on an independent set (for methods evaluation) scenarios are employed. Our developed system yielded an accuracy of 99.08%, obtained using the random forest classifier. Our results clearly demonstrates that machine learning methods can be used to identify the paper-ink samples based on LIBS reliably at a faster rate.

Turkey, ertanatarAgmail.com, ozyi maz(yiIdiz.edu.tr 2School of Electrical and Computer Engineering, Purdue University, Electrical Engineering Building, 465 Northwestern Ave., West Lafayette, IN 47907, USA, edu tissues for maintaining the viability of cells and to transport Abstract-It is necessary to know both the primary and carbon dioxide back to the lungs through the veins [3]. secondary structure of proteins in order to predict their Thehemoglobin molecule is made up of four polypeptide biological functions. Neural Networks are effective for subunits that are known as globins and four heme groups for secondary structure prediction of proteins. In this study, each globin chain [4]. The four globins of normal adult the self-organizing map (SOM) algorithm, and the self-hemoglobin consist of two identical alpha chains, each with organizing global ranking (SOGR) algorithm were 141 amino acids and two beta chains each with 146 amino investigated with different window sizes of amino acid acids. Defects can occur in the genetic code for hemoglobin sequences to predict the protein secondary structure and cause abnormalities such as sickle cell disease and from the protein primary structure. In this study, all of thalassemia [4]. the data were obtained from PDB (Protein Data Bank). ANN's are a useful approach for prediction of secondary Then, the letter data were converted to numerical data structures of proteins [5]. They have the capability of and processed with ANNs. 17 different types of data with mapping the relationships between the protein sequences. a number of sliding window lengths were used. In Studies on the protein secondary structure prediction by general, results were very satisfactory, and the SOGR computational and statistical methods started with Krigbaum had the highest testing accuracies and faster speed of and Kuntton as pointed out by Cai et al. [6]. They used learning. multiple linear regression algorithms to predict the amino acid composition of a protein.

Laser-induced breakdown spectroscopy (LIBS) is an important analysis technique with applications in many industrial branches and fields of scientific research. Nowadays, the advantages of LIBS are impaired by the main drawback in the interpretation of obtained spectra and identification of observed spectral lines. This procedure is highly time-consuming since it is essentially based on the comparison of lines present in the spectrum with the literature database. This paper proposes the use of various computational intelligence methods to develop a reliable and fast classification of quasi-destructively acquired LIBS spectra into a set of predefined classes. We focus on a specific problem of classification of paper-ink samples into 30 separate, predefined classes. For each of 30 classes (10 pens of each of 5 ink types combined with 10 sheets of 5 paper types plus empty pages), 100 LIBS spectra are collected. Four variants of preprocessing, seven classifiers (decision trees, random forest, k-nearest neighbor, support vector machine, probabilistic neural network, multi-layer perceptron, and generalized regression neural network), 5-fold stratified cross-validation, and a test on an independent set (for methods evaluation) scenarios are employed. Our developed system yielded an accuracy of 99.08%, obtained using the random forest classifier. Our results clearly demonstrates that machine learning methods can be used to identify the paper-ink samples based on LIBS reliably at a faster rate.

This paper describes the algorithms used to estimate the state of consumption of a pump, based on dynamic pressure or vibrations. To create algorithms, the author used computational intelligence methods, in the form of neural networks. In order to perform the analysis, data analysis systems were designed, based on three neural networks: multilayer perceptron neural network (MLP), generalized regression neural network (GRNN) and probabilistic neural network (PNN). Processing of the input signal in the final result of the analysis consisted of several steps. First, the measurement data were preprocessed (delete constant component, normalization, standardization, reduction, fast Fourier transform (FFT), etc.), and training and test sets were prepared using the matrices with the expected system answers. The last step was the analysis, consisting of design data analysis systems based on artificial neural networks and their learning and testing. On the basis of the obtained results were evaluated the effectiveness of neural networks and the methods of pre-processing of signals applied to approximate the state of consumption of the displacement pump. Design systems were evaluated based on accuracy (generated error) and complexity (number of parameters and training time) criteria. The main contribution of the paper is to design and compare methods for pre-processing the signal, and to design and compare the effectiveness of three neural networks in the diagnosis consumption of a positive displacement pump.

In this paper we introduce a novel technique for modeling and recognizing gesture signals in 2D space. This technique is based on measuring the direction of the gradient of the movement trajectory as features of the gesture signal. Each gesture signal is represented as a time series of gradient angle values. These features are classified by applying a given classification method. In this article we compared the accuracy of a feed forward Artificial Neural Network with a Support Vector Machine using a radial kernel. The comparison was based on the recorded data of 13 gesture signals as training and testing data. The average accuracy of the ANN and SVM were 98.27% and 96.34% respectively. The false detection ratio was 3.83% for ANN and 8.45% for SVM, which suggests the ANN is more suitable for gesture signal recognition.

This work presents analysis of chromatographic signal used to identify substances in samples. First part consists of chromatography overview and description of three classification methods (neural network with backpropagation, probabilistic neural network with Parzen window and support vector machines). Designed algorithm consists of several stages: signal filtering, peak detection and its approximation with sum of two Gaussian functions. The parameters of that two curves are the features vectors describing the peak of the substance. The last step is classification, for which two types of supervised machine learning were compared, based on the whole signal and on features vectors. Both types were tested for different classificators and their parameters.

As massively multi-player gaming environments become more detailed, developing agents to populate these virtual worlds as capable non-player characters poses an increasingly complex problem. Human players in many games must achieve their objectives through financial skills such as trading and supply chain management as well as through combat and diplomacy. In this paper, we examine the problem of creating intelligent trading agents for virtual markets. Using historical data from EVE Online, a science-fiction based MMORPG, we evaluate several strategies for buying, selling, and supply chain management. We demonstrate that using reinforcement learning to determine policies based on the market microstructure gives trading agents a competitive advantage in amassing wealth. Imbuing agents with the ability to adapt their trading policies can make them more resistant to exploitation by other traders and capable of participating in virtual economies on an equal footing with humans.

The Penn-Lehman automated trading project is a broad investigation of algorithms and strategies for automated trading in financial markets. The PLAT project's centerpiece is the Penn exchange simulator (PXS), a software simulator for automated stock trading that merges automated client orders for shares with real-world, real-time order data. PXS automatically computes client profits and losses, volumes traded, simulator and external prices, and other quantities of interest. To test the effectiveness of PXS and of various trading strategies, we've held three formal competitions between automated clients.

The Penn-Lehman automated trading project is a broad investigation of algorithms and strategies for automated trading in financial markets. The PLAT project's centerpiece is the Penn exchange simulator (PXS), a software simulator for automated stock trading that merges automated client orders for shares with real-world, real-time order data. PXS automatically computes client profits and losses, volumes traded, simulator and external prices, and other quantities of interest. To test the effectiveness of PXS and of various trading strategies, we've held three formal competitions between automated clients.

We describe the Penn-Lehman Automated Trading project, a broad investigation of algorithms and strategies for automated trading in financial markets that centers around the Penn Exchange Simulator (PXS). PXS performs a stock market simulation that integrates virtual client orders with limit order data from real-world electronic exchanges. We describe our motivations and interests, the design and architecture of PXS, and the results of several competitions we have designed and held among a diverse set of 14 automated trading clients submitted by over 30 participants.

Alteration and addition to valuable data on paper documents are among the fastest growing crimes around the globe. The loss due to these crimes is huge and is increasing with an alarming rate. The techniques, which are used by forensic document examiners, to examine such cases are still limited to manual examination of physical, chemical and microscopic characteristics. Moreover, it is very difficult to detect an alteration when the ink of similar color is involved. We could not find much in the literature to deal with this problem in an automated pattern recognition framework. In this paper, we restrict ourselves to alterations made with ball-point pen strokes and propose a scheme for detection of such alterations using pattern recognition tools. For this, a large set of color and texture based features is extracted. To choose an adequate set of useful features from the extracted ones, a multilayer perceptron (MLP) based feature analysis technique is used. For detection of the alteration, three different classifiers, namely, K-nearest neighbor (KNN), multilayer perceptron (MLP) and support vector machines (SVMs) are used. The results are quite promising.

This paper proposes a novel automated agent strategy for stock market trading, developed in the context of the Penn-Lehman Automated Trading (PLAT) simulation platform [1]. We provide a comprehensive experimental validation of our strategy using historic order book data from the NASDAQ market.

This paper documents the development of three autonomous stocktrading agents within the framework of the Penn Exchange Simulator (PXS), a novel stock-trading simulator that takes advantage of electronic crossing networks to realistically mix agent bids with bids from the real stock market [1]. The three approaches presented take inspiration from reinforcement learning, myopic trading using regression-based price prediction, and market making. These approaches are fully implemented and tested with results reported here, including individual evaluations using a fixed opponent strategy and a comparative analysis of the strategies in a joint simulation. The market-making strategy described in this paper was the winner in the fall 2003 PLAT live competition and the runnerup in the spring 2004 live competition, exhibiting consistent profitability. The strategy's performance in the live competitions is presented and analyzed.