Showing papers in "International Journal of Intelligent Systems and Applications in 2019"

Air Quality Prediction in Visakhapatnam with LSTM based Recurrent Neural Networks

Abstract: The research activity considered in this paper concerns about efficient approach for modeling and prediction of air quality. Poor air quality is an environmental hazard that has become a great challenge across the globe. Therefore, ambient air quality assessment and prediction has become a significant area of study. In general, air quality refers to quantification of pollution free air in a particular location. It is determined by measuring different types of pollution indicators in the atmosphere. Traditional approaches depend on numerical methods to estimate the air pollutant concentration and require lots of computing power. Moreover, these methods cannot draw insights from the abundant data available. To address this issue, the proposed study puts forward a deep learning approach for quantification and prediction of ambient air quality. Recurrent neural networks (RNN) based framework with special structured memory cells known as Long Short Term Memory (LSTM) is proposed to capture the dependencies in various pollutants and to perform air quality prediction. Real time dataset of the city Visakhapatnam having a record of 12 pollutants was considered for the study. Modeling of temporal sequence data of each pollutant was performed for forecasting hourly based concentrations. Experimental results show that proposed RNN-LSTM frame work attained higher accuracy in estimating hourly based air ambience. Further, this model may be enhanced by adopting bidirectional mechanism in recurrent layer.

Comparison of Predicting Student‘s Performance using Machine Learning Algorithms

Abstract: Predicting the student performance is playing vital role in educational sector so that the analysis of student‘s status helps to improve for better performance. Applying data mining concepts and algorithms in the field of education is Educational Data Mining. In recent days, Machine learning algorithms are very much useful in almost all the fields. Many researchers used machine learning algorithms only. In this paper we proposed the student performance prediction system using Deep Neural Network. We trained the model and tested with Kaggle dataset using different algorithms such as Decision Tree (C5.0), Naïve Bayes, Random Forest, Support Vector Machine, K-Nearest Neighbor and Deep neural network in R Programming and compared the accuracy of all other algorithms. Among six algorithms Deep Neural Network outperformed with 84% as accuracy.

A Trend Analysis of Machine Learning Research with Topic Models and Mann-Kendall Test

Abstract: This paper aims to systematically examine the literature of machine learning for the period of 1968~2017 to identify and analyze the research trends. A list of journals from well-established publishers ScienceDirect, Springer, JMLR, IEEE (approximately 23,365 journal articles) related to machine learning is used to prepare a content collection. To the best of our information, it is the first effort to comprehend the trend analysis in machine learning research with topic models: Latent Semantic Analysis (LSA), Latent Dirichlet Allocation (LDA), and LDA with Coherent Model (LDA_CM). The LDA_CM topic model gives the highest topic coherence amongst all topic models under consideration. This study provides a scientific ground that helps to overcome the subjectivity of collective opinion. The Mann-Kendall test is used to understand the trend of the topics. Our findings provide indicative of paradigmatic shifts in research methodology of significant patterns of topical prominence and the evolving research areas. It is used to highlight the evolution regarding the previous and recent trends in research topics in the area of machine learning. Understanding such an intellectual structure and future trends will assist the researchers to adopt the divergent developments of this research in one place. This paper analyzes the overall trends of the machine learning research since 1968, based on the latent topics identified in the period of 2007~2017 that may be helpful to the researchers exploring the recommended areas and publish their research articles.

Weight Matrix-Based Representation of Sub-Optimum Disturbance Cancellation Filters

Abstract: The disturbance cancellation techniques are investigated in this paper for Passive Bistatic Radars. The conventional procedure is to compute a clean signal by iteratively constructing an error vector from the residual of the surveillance samples after subtraction of a linear combination of clutters samples. A weight vector is eventually extracted in pure block algorithms, while a weight matrix is computed in iterative schemes. It is illustrated in this paper that the computed weight matrix in the latter case contains valuable information describing the clutters properties. The weight matrix-based disturbance attenuation technique is then innovated and its effectiveness is compared to the conventional errorbased procedure in the test bed of several available iterative algorithms. Moreover, a revision of the FBLMS algorithm is presented to cover the case of complex input signals.

Comparing the Performance of Naive Bayes And Decision Tree Classification Using R

Abstract: The use of technology is at its peak. Many companies try to reduce the work and get an efficient result in a specific amount of time. But a large amount of data is being processed each day that is being stored and turned into large datasets. To get useful information, the dataset needs to be analyzed so that one can extract knowledge by training the machine. Thus, it is important to analyze and extract knowledge from a large dataset. In this paper, we have used two popular classification techniquesDecision tree and Naive Bayes to compare the performance of the classification of our data set. We have taken student performance dataset that has 480 observations. We have classified these students into different groups and then calculated the accuracy of our classification by using the R language. Decision tree uses a divide and conquer method including some rules that makes it easy for humans to understand. The Naive Bayes theorem includes an assumption that the pair of features being classified are independent. It is based on the Bayes theorem.

An Enhanced Differential Evolution Algorithm with Multi-mutation Strategies and Self-adapting Control Parameters

Abstract: Differential evolution (DE) is a stochastic population-based optimization algorithm first introduced in 1995. It is an efficient search method that is widely used for solving global optimization problems. It has three control parameters: the scaling factor (F), the crossover rate (CR), and the population size (NP). As any evolutionary algorithm (EA), the performance of DE depends on its exploration and exploitation abilities for the search space. Tuning the control parameters and choosing a suitable mutation strategy play an important role in balancing the rate of exploration and exploitation. Many variants of the DE algorithm have been introduced to enhance its exploration and exploitation abilities. All of these DE variants try to achieve a good balance between exploration and exploitation rates. In this paper, an enhanced DE algorithm with multi-mutation strategies and self-adapting control parameters is proposed. We use three forms of mutation strategies with their associated self-adapting control parameters. Only one mutation strategy is selected to generate the trial vector. Switching between these mutation forms during the evolution process provides dynamic rates of exploration and exploitation. Having different rates of exploration and exploitation through the optimization process enhances the performance of DE in terms of accuracy and convergence rate. The proposed algorithm is evaluated over 38 benchmark functions: 13 traditional functions, 10 special functions chosen from CEC2005, and 15 special functions chosen from CEC2013. Comparison is made in terms of the mean and standard deviation of the error with the standard \"DE/rand/1/bin\" and five state-of-theart DE algorithms. Furthermore, two nonparametric statistical tests are applied in the comparison: Wilcoxon signed-rank and Friedman tests. The results show that the performance of the proposed algorithm is better than other DE algorithms for the majority of the tested functions.

Ensemble Feature Selection Algorithm

Abstract: In this paper, we propose a new feature selection algorithm based on ensemble selection. In order to generate the library of models, each model is trained using just one feature. This means each model in the library represents a feature. Ensemble construction returns a well performing subset of features associated to the well performing subset of models. Our proposed approaches are evaluated using eight benchmark datasets. The results show the effectiveness of our ensemble selection approaches.

Sugarcane Crop Yield Forecasting Model Using Supervised Machine Learning

Abstract: Agriculture is the most important sector in the Indian economy and contributes 18% of Gross Domestic Product (GDP). India is the second largest producer of sugarcane crop and produces about 20% of the world's sugarcane. In this paper, a novel approach to sugarcane yield forecasting in Karnataka(India) region using LongTerm-Time-Series (LTTS), Weather-and-soil attributes, Normalized Vegetation Index(NDVI) and Supervised machine learning(SML) algorithms have been proposed. Sugarcane Cultivation Life Cycle (SCLC) in Karnataka(India) region is about 12 months, with plantation beginning at three different seasons. Our approach divides yield forecasting into three stages, i)soil-and-weather attributes are predicted for the duration of SCLC, ii)NDVI is predicted using Support Vector Machine Regression (SVR) algorithm by considering soil-and-weather attributes as input, iii)sugarcane crop is predicted using SVR by considering NDVI as input. Our approach has been verified using historical dataset and results have shown that our approach has successfully modeled soil and weather attributes prediction as 24 steps LTTS with accuracy of 85.24% for Soil Temperature given by Lasso algorithm, 85.372% accuracy for Temperature given by Naive-Bayes algorithm, accuracy for Soil Moisture is 77.46% given by Naive-Bayes, NDVI prediction with accuracy of 89.97% given by SVR-RBF, crop prediction with accuracy of 83.49% given by SVR-RBF.

Football Match Prediction with Tree Based Model Classification

Abstract: This paper presents the football match prediction using a tree-based model algorithm (C5.0, Random Forest, and Extreme Gradient Boosting). Backward wrapper model was applied as a feature selection methodology to help select the best feature that will improve the accuracy of the model. This study used 10 seasons of football data match history (2007/2008 – 2016/2017) in the English Premier League with 15 initial features to predict the match results. With the tuning process, each model showed improvement in accuracy. Random Forest algorithm generated the best accuracy with 68,55% while the C5.0 algorithm had the lowest accuracy at 64,87% and Extreme Gradient Boosting algorithm produced accuracy of 67,89%. With the output produced in this study, the Decision Tree based algorithm is concluded as not good enough in predicting a football match history.

Development a New Crossover Scheme for Traveling Salesman Problem by aid of Genetic Algorithm

Abstract: ─This research work provides a detailed working principle and analysis technique of multioffspring crossover operator. The proposed approach is an extension of the basic partiallymapped crossover (PMX) based upon survival of the fittest theory. It improves the performance of the genetic algorithm (GA) for solving the well-known combinatorial optimization problem, the traveling salesman problem (TSP). This study is based on numerical experiments of the proposed with other traditional crossover operators for eighteen benchmarks TSPLIB instances. The simulation results show a considerable improvement because the proposed operator enhances the opportunity of having better offspring. Moreover, the t-test also establishes the improved significance of the proposed operator. Its preferable results not only confirm the advantages over others, but also show the long run survival of a generation having a number of offspring more than the number of parents with the help of mathematical ecology theory. Index Terms─NP-hard, Traveling salesman problems, Genetic algorithms, Multi-offspring, Crossover operators.

Balinese Historian Chatbot using Full-Text Search and Artificial Intelligence Markup Language Method

Abstract: In the era of technology, various information could be obtained quickly and easily. The history of Bali is one of the information that could be obtained. Balinese have known their history through Babad and stories which are told through generations. Babad is traditionalhistorical writing which tells important event that has happened. As technology evolves, Balinese‘s interest in studying their own history has been decreased. It is caused by people interest in studying history books and chronicles tend to decrease over time. Therefore, an innovation of technology, which able to convert historical data from printed media to digital media, is needed. The technology that could be used is Chatbot technology; a computer program that could carry out conversations. Chatbot technology is used to make people learning history easily by using Instant Messenger LINE as a platform to communicate. This Chatbot uses two methods, namely the Artificial Intelligence Markup Language method and the Full-Text Search method. The Artificial Intelligence Markup Language method is used as the process of making characteristic of questions and answers. The Full-Text Search method is the process of matching answers based on user input. This chatbot only uses Indonesian to communicate. The results of this study are a Chatbot that could be accessed by using Instant Messenger LINE and could communicate like historian expert.

A Hybrid Approach for Requirements Prioritization Using LFPP and ANN

Abstract: Requirements prioritization is a most important activity to rank the requirements as per their priority of order .It is a crucial phase of requirement engineering in software development process. In this research introduced a MCDM model for requirements prioritization. To select a best supplier firm of washing machine three important criteria are used. In this proposed model investigation for requirements prioritization, a case study adopted from Ozcan et al using LOG FAHP (Logarithmic fuzzy analytic hierarchy process) and ANN (Artificial Neural Network) based model to choose the best supplier firm granting the highest client satisfaction among all technical aspects. The test was conducted on MATLAB software and result evaluated on fuzzy comparison matrix with three supplier selection criteria based on FAHP and LOGANFIS that shows the decision making outcome for requirements prioritization is better than existing approaches with higher priority.

Application of Bird Swarm Algorithm for Allocation of Distributed Generation in an Indian Practical Distribution Network

Abstract: This article addresses an optimal allocation of multi Distributed Generation (DG) units in an Indian practical radial distribution network (RDN) for minimization of network loss and voltage deviation. For this work, combined sensitivity index (CSI) is utilized to identify the appropriate positions/locations of DG units. However, the appropriate size of DG is determined through a nature-inspired; population-based Bird Swarm Algorithm (BSA). Secondly, the influence of DG penetration level on network loss and voltage profile is investigated and presented. In this regard, two types of DG technologies (solar and biomass) are considered for loss reduction and voltage deviation reduction. The performance of CSI and BSA methodology is successfully evaluated on an Indian practical 52-bus RDN.

A Cyclic Attribution Technique Feature Selection Method for Human Activity Recognition

Abstract: Feature selection is a technique of selecting the most important features for predictive model construction. It is a key component in machine learning for many pattern recognition applications. The primary objective of this paper is to create a more precise system for Human Activity Recognition (HAR) by identifying the most appropriate features. We propose a Cyclic Attribution Technique (CAT) feature selection technique for recognition of human activity based on group theory and the fundamental properties of the cyclic group. We tested our model on UCI-HAR dataset focusing on six activities. With the proposed method, 561 features could be reduced to 63. Using an Artificial Neural Network (ANN), we compared performances of our new dataset with selected features and the original dataset classifier. Results showed that the model could provide an excellent overall accuracy of 96.7%. The proposed CAT technique can specify high-quality features to the success of HAR with ANN classifier. Two benefits support this technique by reducing classification overfitting and training time.

Optimal Machine learning Model for Software Defect Prediction

Abstract: Machine Learning is a division of Artificial Intelligence which builds a system that learns from the data. Machine learning has the capability of taking the raw data from the repository which can do the computation and can predict the software bug. It is always desirable to detect the software bug at the earliest so that time and cost can be reduced. Feature selection technique wrapper and filter method is used to find the most optimal software metrics. The main aim of the paper is to find the best model for the software bug prediction. In this paper machine learning techniques linear Regression, Random Forest, Neural Network, Support Vector Machine, Decision Tree, Decision Stump are used and comparative analysis has been done using performance parameters such as correlation, R-squared, mean square error, accuracy for software modules named as ant, ivy, tomcat, berek, camel, lucene, poi, synapse and velocity. Support vector machine outperform as compare to other machine learning model.

Fuzzy Clustering of Sequential Data

Abstract: With the increase in popularity of the Internet and the advancement of technology in the fields like bioinformatics and other scientific communities the amount of sequential data is on the increase at a tremendous rate. With this increase, it has become inevitable to mine useful information from this vast amount of data. The mined information can be used in various spheres; from day to day web activities like the prediction of next web pages, serving better advertisements, to biological areas like genomic data analysis etc. A rough set based clustering of sequential data was proposed by Kumar et al recently. They defined and used a measure, called Sequence and Set Similarity Measure to determine similarity in data. However, we have observed that this measure does not reflect some important characteristics of sequential data. As a result, in this paper, we used the fuzzy set technique to introduce a similarity measure, which we termed as Kernel and Set Similarity Measure to find the similarity of sequential data and generate overlapping clusters. For this purpose, we used exponential string kernels and Jaccard's similarity index. The new similarity measure takes an account of the order of items in the sequence as well as the content of the sequential pattern. In order to compare our algorithm with that of Kumar et al, we used the MSNBC data set from the UCI repository, which was also used in their paper. As far as our knowledge goes, this is the first fuzzy clustering algorithm for sequential data.

Segmentation of Soft Tissues and Tumors from Biomedical Images using Optimized K-Means Clustering via Level Set formulation

Abstract: Biomedical Image-segmentation is one of the ways towards removing an area of attentiveness by making various segments of an image. The segmentation of biomedical images is considered as one of the challenging tasks in many clinical applications due to poor illuminations, intensity inhomogeneity and noise. In this paper, we propose a new segmentation method which is called Optimized K-Means Clustering via Level Set Formulation. The proposed method diversified into two stages for efficient segmentation of soft tissues and tumor‘s from MRI brain Scans Images, which is called pre-processing and post-processing. In the first stage, a hybrid approach is considered as pre-processing is called Optimized K-Means Clustering which is the combined approach of Particle Swarm Optimization (PSO) as well as K-Means Clustering for improve the clustering efficiency. We choose the ̳optimal‘ cluster centers by Particle Swarm Optimization (PSO) algorithm for improving the clustering efficiency. During the process of pre-processing, these segmentation results suffer from few drawbacks such as outliers, edge and boundary leakage problems. In this regard, post-processing is necessary to minimize the obstacles, so we are implementing pre-processing results by using level-set method for smoothed and accurate segmentation of regions from biomedical images such as MRI brain images over existing level set methods.

Method for Unit Self-Diagnosis at System Level

Abstract: This paper suggests unconventional approach to system level self-diagnosis. Traditionally, system level self-diagnosis focuses on determining the state of the units which are tested by other system units. In contrast, the suggested approach utilizes the results of tests performed by a system unit to determine its own state. Such diagnosis is in many respects close to self-testing, since a unit evaluates its own state, which is inherent in selftesting. However, as distinct from self-testing, in the suggested approach a unit evaluates it on the basis of tests that it does not performs on itself, but on other system units. The paper considers different diagnosis models with various testing assignments and different faulty assumptions including permanent and intermittent faults, and hybridfault situations. The diagnosis algorithm for identifying the unit’s state has been developed, and correctness of the algorithm has been verified by computer simulation experiments.

Performance Assessment of Bacterial Foraging based Power System Stabilizer in Multi-Machine Power System

Abstract: This paper describes the process of power system stabilizer (PSS) optimization by using bacterial foraging (BG) to improve the power system stability and damping out the oscillation during large and small disturbances in a multi-machine power system. The proposed PSS type is P. Kundur (Lead-Lag) with speed deviation as the input signal. BG used to optimize the PSS gains. The proposed BG based delta w lead-lag PSS (P. Kundur structure) (BG-PSS) evaluated in the wellknown benchmark simulation problem P. Kundur 4machines 11-buses 2-areas power system. The BG-PSS compared with MB-PSS with simplified settings: IEEE® type PSS4B according to IEEE Std. 421.5, Conventional Delta w PSS (as the proposed PSS without optimization) from P. Kundur, and Conventional Acceleration Power (Delta Pa) PSS to demonstrate its robustness and superiority versus the three PSSs types to damp out the inter-area oscillations in a multi-machine power system. The damping ratio and the real part of the eigenvalues used as the fitness function in the optimization process. The nonlinear simulation results obtained in the MATLAB / SIMULINK environment prove that the proposed PSS is highly effective, robust, & superior to the other used controllers in restrictive the inter-area oscillation in a large power system & to maintain the wide-area stability of the system. Also, the performance indices eigenvalue analysis, peak overshoot, settling time, and steady-state error used to validate the superior oscillation damping and fast recovered transient dynamic behavior over the three considered controllers.

Analysis of Cyberbullying Incidence among Filipina Victims: A Pattern Recognition using Association Rule Extraction

Abstract: Cyberbullying is an intentional action of harassment along the complex domain of social media utilizing information technology online. This research experimented unsupervised associative approach on text mining technique to automatically find cyberbullying words, patterns and extract association rules from a collection of tweets based on the domain / frequent words. Furthermore, this research identifies the relationship between cyberbullying keywords with other cyberbullying words, thus generating knowledge discovery of different cyberbullying word patterns from unstructured tweets. The study revealed that the type of dominant frequent cyberbullying words are intelligence, personality, and insulting words that describe the behavior, appearance of the female victims and sex related words that humiliate female victims. The results of the study suggest that we can utilize unsupervised associative approached in text mining to extract important information from unstructured text. Further, applying association rules can be helpful in recognizing the relationship and meaning between keywords with other words, therefore generating knowledge discovery of different datasets from unstructured text.

Application of Particle Swarm based Neural Network to Predict Scour Depth around the Bridge Pier

Abstract: Scour around the bridge pier is one of the major factors which affect the safety and stability of the bridge structure. Due to the presence of complexity in the scour mechanism, there is no common and simple method to estimate the scour depth. The present paper gives an idea of hybridizing two techniques such as an artificial neural network with swarm intelligence technique particle swarm optimization to estimate the scour depth around the bridge pier and abbreviated as PSO-ANN. The present discussion covers the estimation of scour depth for clear water and live bed scour condition around circular and rectangular pier shapes. The independent variables, Sediment size (d50), sediment quantity (Sq), velocity (u) and time (t) are used as input to develop the models to estimate or quantify a dependent variable scour depth (ds). The efficiency and accuracy of the model are measured using model performances indicators such as Correlation Coefficient (CC), Normalized Root Mean Square Error (NRMSE), Nash Sutcliffe Error (NSE), and Normalized Mean Bias (NMB). The predicted results of both the models are compared with each other and also compared with measured scour depth. The study concludes that the proposed PSO-ANN model is suitable to estimate the scour depth in both the cases for circular and rectangular pier shapes.

Smart Warehouse Management using Hybrid Architecture of Neural Network with Barcode Reader 1D / 2D Vision Technology

Abstract: Manually, to manage stocks amounts spending the every day in the rays to count for each product the number which it remains in stores, or to record by a scanner head barcode information dependent of each product. However, the mission become increasingly difficult if several warehouses are found, that involves much time to pass from a product to another, moreover that requires agents to carry out these spots. In this article we use a network architecture neuron combined with the readers bar code of technology vision, this method allows to know in real time information concerning each product in stock. It will allow besides introducing the concept of real stocks rather than physical. However The basic classical use of data and to feed it will be completely changed by the spheres of knowledge which generates the NN (Neural Network) to store information on the quantity at a given time (Dynamic inventory), the entries(delivery of suppliers ) and the outputs ( delivery or sale with the customers and use of manufacturing pieces or repair ).

Enhanced Deep Feed Forward Neural Network Model for the Customer Attrition Analysis in Banking Sector

Abstract: In the present era with the development of the innovation and the globalization, attrition of customer is considered as the vital metric which decides the incomes and gainfulness of the association. It is relevant for all the business spaces regardless of the measure of the business notwithstanding including the new companies. As per the business organization, about 65% of income comes from the customer's client. The objective of the customer attrition analysis is to anticipate the client who is probably going to exit from the present business association. The attrition analysis also termed as churn analysis. The point of this paper is to assemble a precise prescient model using the Enhanced Deep Feed Forward Neural Network Model to predict the customer whittling down in the Banking Domain. The result obtained through the proposed model is compared with various classes of machine learning algorithms Logistic regression, Decision tree, Gaussian Naïve Bayes Algorithm, and Artificial Neural Network. The outcome demonstrates that the proposed Enhanced Deep Feed Forward Neural Network Model performs best in accuracy compared with the existing machine learning model in predicting the customer attrition rate with the Banking Sector.

Structural Transformations of Incoming Signal by a Single Nonlinear Oscillatory Neuron or by an Artificial Nonlinear Neural Network

Abstract: Structural transformations of incoming informational signal by a single nonlinear oscillatory neuron or an artificial nonlinear neural network are investigated. The neurons are modeled as threshold devices so that the artificial nonlinear neural network under consideration are systems of nonlinear van der Pol type oscillatory neurons. The neurons are coupled by synaptic weight coefficients to endow the systems with the configuration topology of a chain or a ring. It is shown that the morphology of the outgoing signal – with respect to the shape, amplitude and time dependence of the instantaneous frequency of the signal – at the output of such a neural network has a higher degree of stochasticity than the morphology of the signal at the output of a single neuron. We conclude that the process of coding by a single neuron or an entire chain-like or circular neural network may be considered in terms of frequency modulations, which are known in Physics as a way to transmit information. We conjecture that frequency modulations constitute one of the ways of coding of information by the neurons in these types of neural networks.

A New Hybrid Genetic and Information Gain Algorithm for Imputing Missing Values in Cancer Genes Datasets

Abstract: A DNA microarray can represent thousands of genes for studying tumor and genetic diseases in humans. Datasets of DNA microarray normally have missing values, which requires an undeniably crucial process for handling missing values. This paper presents a new algorithm, named EMII, for imputing missing values in medical datasets. EMII algorithm evolutionarily combines Information Gain (IG) and Genetic Algorithm (GA) to mutually generate imputable values. EMII algorithm is column-oriented not instance oriented than other implementation of GA which increases column correlation to the class in the same dataset. EMII algorithm is evaluated for imputing the generated missing values in four cancer gene expression standard medical datasets (Colon, Leukemia, Lung cancer-Michigan, and Prostate) via comparing the truth original complete datasets against the imputed datasets. The analysis of the experimental results reveals that the imputed values generated by EMII were almost the same as the original values besides having the same impact on the applied classifiers due to accuracy as similar as the original complete datasets. EMII has a running time of θ(n2), where n is the total number of columns.

Efficient Classification using Average Weighted Pattern Score with Attribute Rank based Feature Selection

Abstract: Classification is found to be an important field of research for many applications such as medical diagnosis, credit risk and fraud analysis, customer segregation, and business modeling. The main intention of classification is to predict the class labels for the unlabeled test samples using a labelled training set accurately. Several classification algorithms exist to classify the test samples based on the trained samples. However, they are not suitable for many real world applications since even a small performance degradation of classification algorithms may lead to substantial loss and crucial implications. In this paper, a simple classification method using the average weighted pattern score with attribute rank based feature selection has been proposed. Feature selection is carried out by computing the attribute score based ranking and the classification is performed using average weighted pattern computation. Experiments have been performed with 40 standard datasets and the results are compared with other classifiers. The outcome of the analysis shows the good performance of the proposed method with higher classification accuracy.

Ant Colony System Algorithm with Dynamic Pheromone Updating for 0/1 Knapsack Problem

Abstract: The 0/1 Knapsack (KP) is a combinatorial optimization problem that can be solved using various optimization algorithms. Ant Colony System (ACS) is one of these algorithms that is operated iteratively and converged emphatically to a matured solution. The convergence of the ACS depends mainly on the heuristic patterns that are used to update the pheromone trails throughout the optimization cycles. Although, ACS has significant advantages, it suffers from a slow convergence, as the pheromones, which are used to initiate the searching process are initialized randomly at the beginning. In this paper, a new heuristic pattern is proposed to speed up the convergence of ACS with 0/1 KP. The proposed heuristic enforces an order-critical item selection. As such, the proposed heuristic depends on considering the profit added by each item, as similar to the existing heuristics, besides the order of item selection. Accordingly, the proposed heuristic allows the items that are added at the end to get more value in order to be considered in the beginning of the next round. As such, with each cycle, the selected items are varied substantially and the pheromones are vastly updated in order to avoid long trapping with the initial values that are initialized randomly. The experiments showed that the proposed heuristic is converged more rapidly compared to the existing heuristics by reducing up to 30% of the cycles required to reach the optimal solution using difficult 0/1 KP datasets. Accordingly, the times required for convergence have been reduced significantly in the proposed work compared to the time required by the existing algorithms.

Prediction of Water Demand Using Artificial Neural Networks Models and Statistical Model

Abstract: The prediction of future water demand will help water distribution companies and government to plan the distribution process of water, which impacts on sustainable development planning. In this paper, we use a linear and nonlinear models to predict water demand, for this purpose, we will use different types of Artificial Neural Networks (ANNs) with different learning approaches to predict the water demand, compared with a known type of statistical methods. The dataset depends on sets of collected data (extracted from municipalities databases) during a specific period of time and hence we proposing a nonlinear model for predicting the monthly water demand and finally provide the more accurate prediction model compared with other linear and nonlinear methods. The applied models capable of making an accurate prediction for water demand in the future for the Jenin city at the north of Palestine. This prediction is made with a time horizon month, depending on the extracted data, this data will be used to feed the neural network model to implement mechanisms and system that can be employed to predicts a short-term for water demands. Two applied models of artificial neural networks are used; Multilayer Perceptron NNs (MLPNNs) and Radial Basis Function NNs (RBFNNs) with different learning and optimization algorithms Levenberg Marquardt (LM) and Genetic Algorithms (GAs), and one type of linear statistical method called Autoregressive integrated moving average ARIMA are applied to the water demand data collected from Jenin city to predict the water demand in the future. The execution results appear that the MLPNNs-LM type is outperformed the RBFNN-GAs and ARIMA models in the prediction the water demand values.

Fuzzy Sliding Mode Control Scheme for Vehicle Active Suspension System Optimized by ABC Algorithm

Abstract: This paper suggests a proposed control scheme of fuzzy sliding mode and PID controller tuned with Artificial bee colony (ABC) algorithm to control vehicle suspension system. Suspension systems are utilized to provide vehicles safety and improve comfortable driving. The effects of the road roughness transmitted by the tires to the vehicle body can be reduced by using suspension systems. Fuzzy system is used for estimating the unknown parameters and uncertainty in the suspension system components (spring, damper and actuator). This study combines sliding mode with fuzzy strategy to design a robust control method. The ABC technique is used to optimize the controller parameters. The suggested control scheme endeavors to limit the vibration of the vehicle body by creating a suitable force for the suspension systems when passing on disturbance. Passive and active suspension systems are compared to test efficiency and ability of the proposed control scheme to enhance the safety and comfortable driving for different road profiles.