Search results for: likelihood

Authors: Olesya Bolkhovskaya, Alexander Maltsev

Abstract:

In this work the characteristics of spatial signal detec¬tion from an antenna array in various sample cases are investigated. Cases for a various number of available prior information about the received signal and the background noise are considered. The spatial difference between a signal and noise is only used. The performance characteristics and detecting curves are presented. All test-statistics are obtained on the basis of the generalized likelihood ratio (GLR). The received results are correct for a short and long sample.

Keywords: GLR test-statistic, detection task, generalized likelihood ratio, antenna array, detection curves, performance characteristics.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1478

Authors: Y. N. Phang, E. F. Loh

Abstract:

The zero truncated model is usually used in modeling count data without zero. It is the opposite of zero inflated model. Zero truncated Poisson and zero truncated negative binomial models are discussed and used by some researchers in analyzing the abundance of rare species and hospital stay. Zero truncated models are used as the base in developing hurdle models. In this study, we developed a new model, the zero truncated strict arcsine model, which can be used as an alternative model in modeling count data without zero and with extra variation. Two simulated and one real life data sets are used and fitted into this developed model. The results show that the model provides a good fit to the data. Maximum likelihood estimation method is used in estimating the parameters.

Keywords: Hurdle models, maximum likelihood estimation method, positive count data.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1814

Authors: Maria C. Mariani, Md Al Masum Bhuiyan, Osei K. Tweneboah, Hector G. Huizar

Abstract:

This work is devoted to the study of modeling geophysical time series. A stochastic technique with time-varying parameters is used to forecast the volatility of data arising in geophysics. In this study, the volatility is defined as a logarithmic first-order autoregressive process. We observe that the inclusion of log-volatility into the time-varying parameter estimation significantly improves forecasting which is facilitated via maximum likelihood estimation. This allows us to conclude that the estimation algorithm for the corresponding one-step-ahead suggested volatility (with ±2 standard prediction errors) is very feasible since it possesses good convergence properties.

Keywords: Augmented Dickey Fuller Test, geophysical time series, maximum likelihood estimation, stochastic volatility model.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 821

Authors: Keunhong Chae, Seokho Yoon

Abstract:

This paper proposes frequency offset (FO) estimation schemes robust to the non-Gaussian noise for orthogonal frequency division multiplexing (OFDM) systems. A maximum-likelihood (ML) scheme and a low-complexity estimation scheme are proposed by applying the probability density function of the cyclic prefix of OFDM symbols to the ML criterion. From simulation results, it is confirmed that the proposed schemes offer a significant FO estimation performance improvement over the conventional estimation scheme in non-Gaussian noise environments.

Keywords: Frequency offset, cyclic prefix, maximum-likelihood, non-Gaussian noise, OFDM.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1977

Authors: Norwati Mustapha, Manijeh Jalali, Abolghasem Bozorgniya, Mehrdad Jalali

Abstract:

Web usage mining algorithms have been widely utilized for modeling user web navigation behavior. In this study we advance a model for mining of user-s navigation pattern. The model makes user model based on expectation-maximization (EM) algorithm.An EM algorithm is used in statistics for finding maximum likelihood estimates of parameters in probabilistic models, where the model depends on unobserved latent variables. The experimental results represent that by decreasing the number of clusters, the log likelihood converges toward lower values and probability of the largest cluster will be decreased while the number of the clusters increases in each treatment.

Keywords: Web Usage Mining, Expectation maximization, navigation pattern mining.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1532

Authors: J. Arneric, A. Rozga

Abstract:

In this paper usefulness of quasi-Newton iteration procedure in parameters estimation of the conditional variance equation within BHHH algorithm is presented. Analytical solution of maximization of the likelihood function using first and second derivatives is too complex when the variance is time-varying. The advantage of BHHH algorithm in comparison to the other optimization algorithms is that requires no third derivatives with assured convergence. To simplify optimization procedure BHHH algorithm uses the approximation of the matrix of second derivatives according to information identity. However, parameters estimation in a/symmetric GARCH(1,1) model assuming normal distribution of returns is not that simple, i.e. it is difficult to solve it analytically. Maximum of the likelihood function can be founded by iteration procedure until no further increase can be found. Because the solutions of the numerical optimization are very sensitive to the initial values, GARCH(1,1) model starting parameters are defined. The number of iterations can be reduced using starting values close to the global maximum. Optimization procedure will be illustrated in framework of modeling volatility on daily basis of the most liquid stocks on Croatian capital market: Podravka stocks (food industry), Petrokemija stocks (fertilizer industry) and Ericsson Nikola Tesla stocks (information-s-communications industry).

Keywords: Heteroscedasticity, Log-likelihood Maximization, Quasi-Newton iteration procedure, Volatility.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2600

Authors: Ehsan Bashiri

Abstract:

A 7-step method (with 25 sub-steps) to assess risk of air pollutants is introduced. These steps are: pre-considerations, sampling, statistical analysis, exposure matrix and likelihood, doseresponse matrix and likelihood, total risk evaluation, and discussion of findings. All mentioned words and expressions are wellunderstood; however, almost all steps have been modified, improved, and coupled in such a way that a comprehensive method has been prepared. Accordingly, the SADRA (Statistical Analysis-Driven Risk Assessment) emphasizes extensive and ongoing application of analytical statistics in traditional risk assessment models. A Sulfur Dioxide case study validates the claim and provides a good illustration for this method.

Keywords: Criteria air pollutants, Matrix of risk, Riskassessment, Statistical analysis.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1651

Authors: F. Z. Doğru, O. Arslan

Abstract:

In general, classical methods such as maximum likelihood (ML) and least squares (LS) estimation methods are used to estimate the shape parameters of the Burr XII distribution. However, these estimators are very sensitive to the outliers. To overcome this problem we propose alternative robust estimators based on the M-estimation method for the shape parameters of the Burr XII distribution. We provide a small simulation study and a real data example to illustrate the performance of the proposed estimators over the ML and the LS estimators. The simulation results show that the proposed robust estimators generally outperform the classical estimators in terms of bias and root mean square errors when there are outliers in data.

Keywords: Burr XII distribution, robust estimator, M-estimator, maximum likelihood, least squares.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2614

Authors: D. I. De Souza, D. R. Fonseca, R. Rocha

Abstract:

In this paper we will develop further the sequential life test approach presented in a previous article by [1] using an underlying two parameter Weibull sampling distribution. The minimum life will be considered equal to zero. We will again provide rules for making one of the three possible decisions as each observation becomes available; that is: accept the null hypothesis H0; reject the null hypothesis H0; or obtain additional information by making another observation. The product being analyzed is a new type of a low alloy-high strength steel product. To estimate the shape and the scale parameters of the underlying Weibull model we will use a maximum likelihood approach for censored failure data. A new example will further develop the proposed sequential life testing approach.

Keywords: Sequential Life Testing, Underlying Weibull Model, Maximum Likelihood Approach, Hypothesis Testing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1345

Authors: Keunhong Chae, Seokho Yoon

Abstract:

In this paper, frequency offset (FO) estimation schemes robust to the non-Gaussian noise environments are proposed for orthogonal frequency division multiplexing (OFDM) systems. First, a maximum-likelihood (ML) estimation scheme in non-Gaussian noise environments is proposed, and then, the complexity of the ML estimation scheme is reduced by employing a reduced set of candidate values. In numerical results, it is demonstrated that the proposed schemes provide a significant performance improvement over the conventional estimation scheme in non-Gaussian noise environments while maintaining the performance similar to the estimation performance in Gaussian noise environments.

Keywords: Frequency offset estimation, maximum-likelihood, non-Gaussian noise environment, OFDM, training symbol.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1902

Authors: Ali Gholami, M.Esfadiari, M.H.Masihabadi

Abstract:

In this research three methods of Maximum Likelihood, Mahalanobis Distance and Minimum Distance were analyzed in the Western part of Isfahan province in the Iran country. For this purpose, the IRS satellite images and various land preparation uses in region including rangelands, irrigation farming, dry farming, gardens and urban areas were separated and identified. In these methods, matrix error and Kappa index were calculated and accuracy of each method, based on percentages: 53.13, 56.64 and 48.44, were obtained respectively. Considering the low accuracy of these methods to separate land uses due to spread of the land uses, it-s suggested the visual interpretation of the map, to preparing the land use map in this region. The map prepared by visual interpretation is in high accuracy if it will be accompany with the visit of the region.

Keywords: Aghche Region, land use map, MaximumLikelihood, Mahalanobis Distance and Minimum Distance.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1772

Authors: N. Mamode Khan, V. Jowaheer

Abstract:

Number of breakdowns experienced by a machinery is a highly under-dispersed count random variable and its value can be attributed to the factors related to the mechanical input and output of that machinery. Analyzing such under-dispersed count observations as a function of the explanatory factors has been a challenging problem. In this paper, we aim at estimating the effects of various factors on the number of breakdowns experienced by a passenger car based on a study performed in Mauritius over a year. We remark that the number of passenger car breakdowns is highly under-dispersed. These data are therefore modelled and analyzed using Com-Poisson regression model. We use quasi-likelihood estimation approach to estimate the parameters of the model. Under-dispersion parameter is estimated to be 2.14 justifying the appropriateness of Com-Poisson distribution in modelling under-dispersed count responses recorded in this study.

Keywords: Breakdowns, under-dispersion, com-poisson, generalized linear model, quasi-likelihood estimation

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1501

Authors: Sagaren Pillay

Abstract:

This study empirically examines the long run equilibrium relationship between South Africa’s exports and imports using quarterly data from 1985 to 2012. The theoretical framework used for the study is based on Johansen’s Maximum Likelihood cointegration technique which tests for both the existence and number of cointegration vectors that exists. The study finds that both the series are integrated of order one and are cointegrated. A statistically significant cointegrating relationship is found to exist between exports and imports. The study models this unique linear and lagged relationship using a Vector Error Correction Model (VECM). The findings of the study confirm the existence of a long run equilibrium relationship between exports and imports.

Keywords: Cointegration lagged, linear, maximum likelihood, vector error correction model.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2727

Authors: Somkiat Lerkvaranyu, Yoshikazu Miyanaga

Abstract:

In this paper, a new technique of signal detection has been proposed for detecting the orthogonal frequency-division multiplexing (OFDM) signal in the presence of nonlinear distortion.There are several advantages of OFDM communications system.However, one of the existing problems is remain considered as the nonlinear distortion generated by high-power-amplifier at the transmitter end due to the large dynamic range of an OFDM signal. The proposed method is the maximum likelihood detection with the symbol estimation. When the training data are available, the neural network has been used to learn the characteristic of received signal and to estimate the new positions of the transmitted symbol which are provided to the maximum likelihood detector. Resulting in the system performance, the nonlinear distortions of a traveling wave tube amplifier with OFDM signal are considered in this paper.Simulation results of the bit-error-rate performance are obtained with 16-QAM OFDM systems.

Keywords: OFDM, TWTA, nonlinear distortion, detection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1628

Authors: Darren Thompson, Philip Morrow, Bryan Scotney, John Winder

Abstract:

Patients with diabetes are susceptible to chronic foot wounds which may be difficult to manage and slow to heal. Diagnosis and treatment currently rely on the subjective judgement of experienced professionals. An objective method of tissue assessment is required. In this paper, a data fusion approach was taken to wound tissue classification. The supervised Maximum Likelihood and unsupervised Multi-Modal Expectation Maximisation algorithms were used to classify tissues within simulated wound models by weighting the contributions of both colour and 3D depth information. It was found that, at low weightings, depth information could show significant improvements in classification accuracy when compared to classification by colour alone, particularly when using the maximum likelihood method. However, larger weightings were found to have an entirely negative effect on accuracy.

Keywords: Classification, data fusion, diabetic foot, stereophotogrammetry, tissue colour.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1664

Authors: X. Mamaqi, J.Miguel, P. Olave

Abstract:

The aim of this paper is to provide an empirical evidence about the effects that the management of continuous training have on employability (or employment stability) in the Spanish labour market. With this purpose a binary logit model with interaction effect is been used. The dependent variable includes two situations of the active workers: continuous and discontinuous employability. To distinguish between them an Employability Index Stability (ESI) was calculated taking into account two factors: time worked and job security. Various aspects of the continuous training and personal workers data are used as independent variables. The data obtained from a survey of a sample of 918 employed have revealed a relationship between the likelihood of continuous employability and continuous training received. The empirical results support the positive and significant relationship between various aspects of the training provided by firms and employability likelihood of the workers, postulate alike from a theoretical point of view.

Keywords: training management, employability/employmentstability, binary logit model, interaction effect, Spanish marketlabour.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1549

Authors: E. B. Jo, J. H. Lee, J. Y. Park, S. M. Kim

Abstract:

In this study, we present an advanced detection technique for mass type breast cancer based on texture information of organs. The proposed method detects the cancer areas in three stages. In the first stage, the midpoints of mass area are determined based on AHE (Adaptive Histogram Equalization). In the second stage, we set the threshold coefficient of homogeneity by using MLE (Maximum Likelihood Estimation) to compute the uniformity of texture. Finally, mass type cancer tissues are extracted from the original image. As a result, it was observed that the proposed method shows an improved detection performance on dense breast tissues of Korean women compared with the existing methods. It is expected that the proposed method may provide additional diagnostic information for detection of mass-type breast cancer.

Keywords: Mass Type Breast Cancer, Mammography, Maximum Likelihood Estimation (MLE), Ranklets, SVM

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1939

Authors: Tanapong Khomyat, Peerapong Uthansakul, Monthippa Uthansakul

Abstract:

In this paper, we evaluate the performance of the Hybrid-MIMO Receiver Scheme (HMRS) in Cognitive Radio network (CR-network). We investigate the efficiency of the proposed scheme which the energy level and user number of primary user are varied according to the characteristic of CR-network. HMRS can allow users to transmit either Space-Time Block Code (STBC) or Spatial-Multiplexing (SM) streams simultaneously by using Successive Interference Cancellation (SIC) and Maximum Likelihood Detection (MLD). From simulation, the results indicate that the interference level effects to the performance of HMRS. Moreover, the exact closed-form capacity of the proposed scheme is derived and compared with STBC scheme.

Keywords: Hybrid-MIMO, Cognitive radio network (CRnetwork), Symbol Error Rate (SER), Successive interference cancellation (SIC), Maximum likelihood detection (MLD).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1591

Authors: D. I. De Souza, G. P. Azevedo, D. R. Fonseca

Abstract:

In this paper we will develop further the sequential life test approach presented in a previous article by [1] using an underlying two parameter Inverse Weibull sampling distribution. The location parameter or minimum life will be considered equal to zero. Once again we will provide rules for making one of the three possible decisions as each observation becomes available; that is: accept the null hypothesis H0; reject the null hypothesis H0; or obtain additional information by making another observation. The product being analyzed is a new electronic component. There is little information available about the possible values the parameters of the corresponding Inverse Weibull underlying sampling distribution could have.To estimate the shape and the scale parameters of the underlying Inverse Weibull model we will use a maximum likelihood approach for censored failure data. A new example will further develop the proposed sequential life testing approach.

Keywords: Sequential Life Testing, Inverse Weibull Model, Maximum Likelihood Approach, Hypothesis Testing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1377

Authors: Tom Snir, Eitan Rubin

Abstract:

Mendelian Disease Genes represent a collection of single points of failure for the various systems they constitute. Such genes have been shown, on average, to encode longer proteins than 'non-disease' proteins. Existing models suggest that this results from the increased likeli-hood of longer genes undergoing mutations. Here, we show that in saturated mutagenesis experiments performed on model organisms, where the likelihood of each gene mutating is one, a similar relationship between length and the probability of a gene being lethal was observed. We thus suggest an extended model demonstrating that the likelihood of a mutated gene to produce a severe phenotype is length-dependent. Using the occurrence of conserved domains, we bring evidence that this dependency results from a correlation between protein length and the number of functions it performs. We propose that protein length thus serves as a proxy for protein cardinality in different networks required for the organism's survival and well-being. We use this example to argue that the collection of Mendelian Disease Genes can, and should, be used to study the rules governing systems vulnerability in living organisms.

Keywords: Systems Biology, Protein Length

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1741

Authors: Yeon-Jea Cho, Sang-Uk Park, Won-Chul Choi, Dong-Jo Park

Abstract:

This paper proposes a smart design strategy for a sequential detector to reliably detect the primary user-s signal, especially in fast fading environments. We study the computation of the log-likelihood ratio for coping with a fast changing received signal and noise sample variances, which are considered random variables. First, we analyze the detectability of the conventional generalized log-likelihood ratio (GLLR) scheme when considering fast changing statistics of unknown parameters caused by fast fading effects. Secondly, we propose an efficient sensing algorithm for performing the sequential probability ratio test in a robust and efficient manner when the channel statistics are unknown. Finally, the proposed scheme is compared to the conventional method with simulation results with respect to the average number of samples required to reach a detection decision.

Keywords: Cognitive radio, fast fading, sequential detection, spectrum sensing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1703

Authors: Chathurika D. Abeyrathne, Malka N. Halgamuge, Peter M. Farrell

Abstract:

There is an ongoing controversy in the literature related to the biological effects of weak, low frequency electromagnetic fields. The physical arguments and interpretation of the experimental evidence are inconsistent, where some physical arguments and experimental demonstrations tend to reject the likelihood of any effect of the fields at extremely low level. The problem arises of explaining, how the low-energy influences of weak magnetic fields can compete with the thermal and electrical noise of cells at normal temperature using the theoretical studies. The magnetoreception in animals involve radical pair mechanism. The same mechanism has been shown to be involved in the circadian rhythm synchronization in mammals. These reactions can be influenced by the weak magnetic fields. Hence, it is postulated the biological clock can be affected by weak magnetic fields and these disruptions to the rhythm can cause adverse biological effects. In this paper, likelihood of altering the biological clock via the radical pair mechanism is analyzed to simplify these studies of controversy.

Keywords: Bio-effect, biological clock, magnetoreception, radical pair mechanism, weak magnetic field.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2278

Authors: Siti Aisyah Zakaria, Nor Azrita Mohd Amin, Noor Fadhilah Ahmad Radi, Nasrul Hamidin

Abstract:

Higher ground-level ozone (GLO) concentration adversely affects human health, vegetations as well as activities in the ecosystem. In Malaysia, most of the analysis on GLO concentration are carried out using the average value of GLO concentration, which refers to the centre of distribution to make a prediction or estimation. However, analysis which focuses on the higher value or extreme value in GLO concentration is rarely explored. Hence, the objective of this study is to classify the tail behaviour of GLO using generalized extreme value (GEV) distribution estimation the return level using the corresponding modelling (Gumbel, Weibull, and Frechet) of GEV distribution. The results show that Weibull distribution which is also known as short tail distribution and considered as having less extreme behaviour is the best-fitted distribution for four selected air monitoring stations in Peninsular Malaysia, namely Larkin, Pelabuhan Kelang, Shah Alam, and Tanjung Malim; while Gumbel distribution which is considered as a medium tail distribution is the best-fitted distribution for Nilai station. The return level of GLO concentration in Shah Alam station is comparatively higher than other stations. Overall, return levels increase with increasing return periods but the increment depends on the type of the tail of GEV distribution’s tail. We conduct this study by using maximum likelihood estimation (MLE) method to estimate the parameters at four selected stations in Peninsular Malaysia. Next, the validation for the fitted block maxima series to GEV distribution is performed using probability plot, quantile plot and likelihood ratio test. Profile likelihood confidence interval is tested to verify the type of GEV distribution. These results are important as a guide for early notification on future extreme ozone events.

Keywords: Extreme value theory, generalized extreme value distribution, ground-level ozone, return level.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 450

Authors: Ehab Mourtaga, Ahmad Sharieh, Mousa Abdallah

Abstract:

An automatic speech recognition system for the formal Arabic language is needed. The Quran is the most formal spoken book in Arabic, it is spoken all over the world. In this research, an automatic speech recognizer for Quranic based speakerindependent was developed and tested. The system was developed based on the tri-phone Hidden Markov Model and Maximum Likelihood Linear Regression (MLLR). The MLLR computes a set of transformations which reduces the mismatch between an initial model set and the adaptation data. It uses the regression class tree, as well as, estimates a set of linear transformations for the mean and variance parameters of a Gaussian mixture HMM system. The 30th Chapter of the Quran, with five of the most famous readers of the Quran, was used for the training and testing of the data. The chapter includes about 2000 distinct words. The advantages of using the Quranic verses as the database in this developed recognizer are the uniqueness of the words and the high level of orderliness between verses. The level of accuracy from the tested data ranged 68 to 85%.

Keywords: Hidden Markov Model (HMM), MaximumLikelihood Linear Regression (MLLR), Quran, Regression ClassTree, Speech Recognition, Speaker-independent.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1870

Authors: D. I. De Souza, G. P. Azevedo, R. Rocha

Abstract:

In this paper we will develop a sequential life test approach applied to a modified low alloy-high strength steel part used in highway overpasses in Brazil.We will consider two possible underlying sampling distributions: the Normal and theInverse Weibull models. The minimum life will be considered equal to zero. We will use the two underlying models to analyze a fatigue life test situation, comparing the results obtained from both.Since a major chemical component of this low alloy-high strength steel part has been changed, there is little information available about the possible values that the parameters of the corresponding Normal and Inverse Weibull underlying sampling distributions could have. To estimate the shape and the scale parameters of these two sampling models we will use a maximum likelihood approach for censored failure data. We will also develop a truncation mechanism for the Inverse Weibull and Normal models. We will provide rules to truncate a sequential life testing situation making one of the two possible decisions at the moment of truncation; that is, accept or reject the null hypothesis H0. An example will develop the proposed truncated sequential life testing approach for the Inverse Weibull and Normal models.

Keywords: Sequential life testing, normal and inverse Weibull models, maximum likelihood approach, truncation mechanism.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1388

Authors: O. Abiodun Adeyinka, B. Adeyemo Adesesan

Abstract:

The forensic use of handwriting depends on the analysis, comparison, and evaluation decisions made by forensic document examiners. When using biometric technology in forensic applications, it is necessary to compute Likelihood Ratio (LR) for quantifying strength of evidence under two competing hypotheses, namely the prosecution and the defense hypotheses wherein a set of assumptions and methods for a given data set will be made. It is therefore important to know how repeatable and reproducible our estimated LR is. This paper evaluated the accuracy and reproducibility of examiners' decisions. Confidence interval for the estimated LR were presented so as not get an incorrect estimate that will be used to deliver wrong judgment in the court of Law. The estimate of LR is fundamentally a Bayesian concept and we used two LR estimators, namely Logistic Regression (LoR) and Kernel Density Estimator (KDE) for this paper. The repeatability evaluation was carried out by retesting the initial experiment after an interval of six months to observe whether examiners would repeat their decisions for the estimated LR. The experimental results, which are based on handwriting dataset, show that LR has different confidence intervals which therefore implies that LR cannot be estimated with the same certainty everywhere. Though the LoR performed better than the KDE when tested using the same dataset, the two LR estimators investigated showed a consistent region in which LR value can be estimated confidently. These two findings advance our understanding of LR when used in computing the strength of evidence in handwriting using forensics.

Keywords: Logistic Regression LoR, Kernel Density Estimator KDE, Handwriting, Confidence Interval, Repeatability, Reproducibility.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 404

Authors: Yeonjeong Lee, Kyoung-jae Kim, Youngtae Kim

Abstract:

This study proposes a novel recommender system that uses data mining and multi-model ensemble techniques to enhance the recommendation performance through reflecting the precise user’s preference. The proposed model consists of two steps. In the first step, this study uses logistic regression, decision trees, and artificial neural networks to predict customers who have high likelihood to purchase products in each product group. Then, this study combines the results of each predictor using the multi-model ensemble techniques such as bagging and bumping. In the second step, this study uses the market basket analysis to extract association rules for co-purchased products. Finally, the system selects customers who have high likelihood to purchase products in each product group and recommends proper products from same or different product groups to them through above two steps. We test the usability of the proposed system by using prototype and real-world transaction and profile data. In addition, we survey about user satisfaction for the recommended product list from the proposed system and the randomly selected product lists. The results also show that the proposed system may be useful in real-world online shopping store.

Keywords: Product recommender system, Ensemble technique, Association rules, Decision tree, Artificial neural networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4175

Authors: J. P. Dubois, O. M. Abdul-Latif

Abstract:

Support Vector Machine (SVM) is a recent class of statistical classification and regression techniques playing an increasing role in applications to detection problems in various engineering problems, notably in statistical signal processing, pattern recognition, image analysis, and communication systems. In this paper, SVM is applied to an infrared (IR) binary communication system with different types of channel models including Ricean multipath fading and partially developed scattering channel with additive white Gaussian noise (AWGN) at the receiver. The structure and performance of SVM in terms of the bit error rate (BER) metric is derived and simulated for these channel stochastic models and the computational complexity of the implementation, in terms of average computational time per bit, is also presented. The performance of SVM is then compared to classical binary signal maximum likelihood detection using a matched filter driven by On-Off keying (OOK) modulation. We found that the performance of SVM is superior to that of the traditional optimal detection schemes used in statistical communication, especially for very low signal-to-noise ratio (SNR) ranges. For large SNR, the performance of the SVM is similar to that of the classical detectors. The implication of these results is that SVM can prove very beneficial to IR communication systems that notoriously suffer from low SNR at the cost of increased computational complexity.

Keywords: Least square-support vector machine, on-off keying, matched filter, maximum likelihood detector, wireless infrared communication.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1911

Authors: D. I. De Souza, D. R. Fonseca, G. P. Azevedo

Abstract:

Sometimes the amount of time available for testing could be considerably less than the expected lifetime of the component. To overcome such a problem, there is the accelerated life-testing alternative aimed at forcing components to fail by testing them at much higher-than-intended application conditions. These models are known as acceleration models. One possible way to translate test results obtained under accelerated conditions to normal using conditions could be through the application of the “Maxwell Distribution Law.” In this paper we will apply a combined approach of a sequential life testing and an accelerated life testing to a low alloy high-strength steel component used in the construction of overpasses in Brazil. The underlying sampling distribution will be three-parameter Inverse Weibull model. To estimate the three parameters of the Inverse Weibull model we will use a maximum likelihood approach for censored failure data. We will be assuming a linear acceleration condition. To evaluate the accuracy (significance) of the parameter values obtained under normal conditions for the underlying Inverse Weibull model we will apply to the expected normal failure times a sequential life testing using a truncation mechanism. An example will illustrate the application of this procedure.

Keywords: Sequential Life Testing, Accelerated Life Testing, Underlying Three-Parameter Weibull Model, Maximum Likelihood Approach, Hypothesis Testing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1605

Authors: Rupesh K. Gopal, Saroj K. Meher

Abstract:

In this report we present a rule-based approach to detect anomalous telephone calls. The method described here uses subscriber usage CDR (call detail record) data sampled over two observation periods: study period and test period. The study period contains call records of customers- non-anomalous behaviour. Customers are first grouped according to their similar usage behaviour (like, average number of local calls per week, etc). For customers in each group, we develop a probabilistic model to describe their usage. Next, we use maximum likelihood estimation (MLE) to estimate the parameters of the calling behaviour. Then we determine thresholds by calculating acceptable change within a group. MLE is used on the data in the test period to estimate the parameters of the calling behaviour. These parameters are compared against thresholds. Any deviation beyond the threshold is used to raise an alarm. This method has the advantage of identifying local anomalies as compared to techniques which identify global anomalies. The method is tested for 90 days of study data and 10 days of test data of telecom customers. For medium to large deviations in the data in test window, the method is able to identify 90% of anomalous usage with less than 1% false alarm rate.

Keywords: Subscription fraud, fraud detection, anomalydetection, maximum likelihood estimation, rule based systems.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2768