Razana Alwee

Hybrid Support Vector Regression and Autoregressive Integrated Moving Average Models Improved by Particle Swarm Optimization for Property Crime Rates Forecasting with Economic Indicators

Crimes forecasting is an important area in the field of criminology. Linear models, such as regression and econometric models, are commonly applied in crime forecasting. However, in real crimes data, it is common that the data consists of both linear and nonlinear components. A single model may not be sufficient to identify all the characteristics of the data. The purpose of this study is to introduce a hybrid model that combines support vector regression (SVR) and autoregressive integrated moving average (ARIMA) to be applied in crime rates forecasting. SVR is very robust with small training data and high-dimensional problem. Meanwhile, ARIMA has the ability to model several types of time series. However, the accuracy of the SVR model depends on values of its parameters, while ARIMA is not robust to be applied to small data sets. Therefore, to overcome this problem, particle swarm optimization is used to estimate the parameters of the SVR and ARIMA models. The proposed hybrid model is used to forecast the property crime rates of the United State based on economic indicators. The experimental results show that the proposed hybrid model is able to produce more accurate forecasting results as compared to the individual models.

Study of cost functions in Three Term Backpropagation for classification problems

Three Term Backpropagation(BP) Network as proposed by Zweiri in 2003 has outperformed standard Two Term Backpropagation. However, further studies on Three Term Backpropagation in 2007 indicated that this network only surpassed standard BP for small scale datasets but not for medium and large scale datasets. It has also been observed that by using Mean Square Error (MSE) as a cost function in Three Term BP has some drawbacks, and these include incorrect saturation and tend to trap in local minima, resulting in slow convergence and poor performance. In this study, thorough experiments on implementing various cost functions are executed to probe the effectiveness of Three Term BP network. The cost functions under investigations include Mean Square Error (MSE), Bernoulli function, Modified cost function and Improved cost function. The results reveal that MSE is not an ideal cost function to be used for Three Term BP. Hence, alternative cost functions need to be considered when using BP network for classification problems.

Economic indicators selection for crime rates forecasting using cooperative feature selection

Features selection in multivariate forecasting model is very important to ensure that the model is accurate. The purpose of this study is to apply the Cooperative Feature Selection method for features selection. The features are economic indicators that will be used in crime rate forecasting model. The Cooperative Feature Selection combines grey relational analysis and artificial neural network to establish a cooperative model that can rank and select the significant economic indicators. Grey relational analysis is used to select the best data series to represent each economic indicator and is also used to rank the economic indicators according to its importance to the crime rate. After that, the artificial neural network is used to select the significant economic indicators for forecasting the crime rates. In this study, we used economic indicators of unemployment rate, consumer price index, gross domestic product and consumer sentiment index, as well as data rates of property crime and violent crime for the United States. Levenberg-Marquardt neural network is used in this study. From our experiments, we found that consumer price index is an important economic indicator that has a significant influence on the violent crime rate. While for property crime rate, the gross domestic product, unemployment rate and consumer price index are the influential economic indicators. The Cooperative Feature Selection is also found to produce smaller errors as compared to Multiple Linear Regression in forecasting property and violent crime rates.

Enhancement of Particle Swarm Optimization in Elman Recurrent Network with Bounded Vmax Function

As the widespread modus operandi in real applications, Backpropagation(BP) in Recurrent Neural Networks (RNN) is computationally more powerful than standard feedforward neural networks. In principle, RNN can implement almost any arbitrary sequential behavior. However, there are many drawbacks in BP network, for instance, confinement in finding local minimum and may get stuck at regions of a search space or trap in local minima. To solve these problems, various optimization techniques such as Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) have been executed to improve ANN performance. In this study, we exploit errors optimization of Elman Recurrent Network with Particle Swarm Optimization (ERNPSO) to probe the performance of both networks with bounded Vmax function. Main characteristics of Vmax function are to control the global exploration of particles in PSO. The results show that ERNPSO with bounded Vmax of hyperbolic tangent furnishes promising outcomes in terms of classification accuracy and convergence rate compared to bounded Vmax sigmoid function and standard Vmax function.

Kullback Leibler divergence for image quantitative evaluation

Medical imaging has been expanding ever since to give diagnostic information through different types of modalities. Currently, there are many types of modalities such as Computed Tomography (CT) scans, Magnetic Resonance Imaging (MRI), X-rays (plain radiography), Positron Emission Tomography (PET) scan and Ultrasonographic diagnostics (USG), available in the field of medical and surgical. These modalities are widely used in clinical diagnosis and development of research in education. In terms of image quality, the qualitative analysis was always used to evaluate the quality of output image from classification results. By qualitative analysis, the researchers were able to judge the precision of detected lesion and hence calculated the accuracy of detection through the testing cases. However, the qualitative analysis was sometimes subjective and the verification from more than one radiologist was needed to confirm the results of classification. Therefore, the quantitative analysis was also needed to ensure the results from the classification algorithm can be assessed objectively. In this study, we propose pixel-based approach of Kullback Leibler (KL) divergence in assessing the medical images. Unlike the standard statistical analysis, the evaluation using KL divergence does not require testing of hypothesis or confidence interval construction based on the mean and standard deviation. The proposed framework of KL is useful to provide a descriptive measure for the purpose of summarizing data. Firstly, both of the original and computed images are normalized where the sum of all intensities is equal to one. Then, the probability distribution is calculated by column using function of hist (HO) and hist (HA) and each of the column are expressed as data vector h0i = {h01, h02, h03, h0i} and hAi = {hA1, hA2, hA3, hAi} respectively. In the computation of probability distribution, the function of hist bins the elements in each data vector of and into 10 equally spaced containers and return the amount of elements in each container as row vector. The results have shown that the proposed framework of Kullback Leibler divergence is promising in presenting better final images quantitatively.

Eye Redness Image Processing Techniques

The use of photographs for the assessment of ocular conditions has been suggested to further standardize clinical procedures. The selection of the photographs to be used as scale reference images was subjective. Numerous methods have been proposed to assign eye redness scores by computational methods. Image analysis techniques have been investigated over the last 20 years in an attempt to forgo subjective grading scales. Image segmentation is one of the most important and challenging problems in image processing. This paper briefly outlines the comprehensive of image processing and the implementation of image segmentation in eye redness.

Swarm Optimized Grey SVR and ARIMA for Modeling of Larceny-Theft Rate with Economic Indicators

As real world data, larceny-theft rates are most likely to have both linear and nonlinear components. A single model such as the linear or nonlinear model may not be sufficient to model the larceny-theft rate. Thus, a hybridization of the linear and nonlinear models is proposed for modeling the larceny-theft rate. The proposed model combines Support Vector Regression (SVR) and Autoregressive Integrated Moving Average (ARIMA) models. Particle swarm optimization is used to optimize the parameters of SVR and ARIMA models. The proposed model is equipped with features selection that combines grey relational analysis and SVR to choose the significant economic indicators for the larceny-theft rate. The experimental results show that the proposed model has better accuracy than the linear, nonlinear, and existing hybrid models in modeling the larceny-theft rate of United States.

An overview on crime prediction methods

In the recent past, crime analyses are required to reveal the complexities in the crime dataset. This process will help the parties that involve in law enforcement in arresting offenders and directing the crime prevention strategies. The ability to predict the future crimes based on the location, pattern and time can serve as a valuable source of knowledge for them either from strategic or tactical perspectives. Nevertheless, to predict future crime accurately with a better performance, it is a challenging task because of the increasing numbers of crime in present days. Therefore, crime prediction method is important to identify the future crime and reduces the numbers of crime. Currently, some researchers have been conducted a study to predict crime based on particular inputs. The performance of prediction models can be evaluated using a variety of different prediction methods such as support vector machine, multivariate time series and artificial neural network. However, there are still some limitations on their findings to provide an accurate prediction for the location of crimes. A large number of research papers on this topic have already been published previously. Thus, in this paper, we thoroughly review each of them and summarized the outcomes. Our objective is to identify current implementations of crime prediction method and the possibility to enhance it for future needs.