Open Access. Powered by Scholars. Published by Universities.®
- Discipline
- Keyword
-
- Stability (21)
- Convex functions (19)
- Genetic algorithm (18)
- Fixed point (16)
- Classification (13)
-
- Optimization (13)
- Clustering (11)
- Convergence (11)
- Data mining (11)
- Simulation (11)
- Authentication (10)
- Cloud computing (10)
- Magnetic field (10)
- Maximum likelihood estimation (10)
- Optimal control (10)
- Cloud Computing (9)
- Prediction (9)
- Soft set (9)
- Chaos (8)
- Fractional calculus (8)
- Genetic Algorithm (8)
- Initial value problems (8)
- Moments (8)
- Multistep (8)
- Particle swarm optimization (8)
- Security (8)
- Symmetric (8)
- Time delay (8)
- Variational inequalities (8)
- Wireless sensor networks (8)
Articles 631 - 660 of 2320
Full-Text Articles in Entire DC Network
Improved Certificateless Signcryption For Iot Smart Devices, Balasubramanian V., Mala T.
Improved Certificateless Signcryption For Iot Smart Devices, Balasubramanian V., Mala T.
Applied Mathematics & Information Sciences
Signcryption achieves both encryption and digital signature at the same time. The cost of the computations involved and the overheads involved in the communication is smaller than the conventional sign-then-encrypt method. Certificateless cryptosystem intends to solve the disadvantage of the identity-based cryptosystem: key escrow problem. It also simplifies the public key management. In this paper, an enhanced new certificateless signcryption technique is proposed by employing bilinear pairings. The scheme performance is also demonstrated to be more effective and efficient for smart devices. The proposed scheme fulfills public ciphertext verifiability and satisfies indistinguishability against adaptively-chosen ciphertext attacks. It is also immune …
Finite Element Analysis Of 125 Mm Tank Gun Smooth Bore Barrel Behaviour During Sand Ingress, J. Balasubramanian, P. Vasundhra, S. Ramesh, V. Sivaramakrishnan
Finite Element Analysis Of 125 Mm Tank Gun Smooth Bore Barrel Behaviour During Sand Ingress, J. Balasubramanian, P. Vasundhra, S. Ramesh, V. Sivaramakrishnan
Applied Mathematics & Information Sciences
This work presents a detailed study of 125 mm tank gun smooth bore barrel behavior during sand ingress while firing with High Explosive Fragmentation (HEF) projectile. The field and metallurgical investigation of failed muzzle end barrels confirms that the presence of sintered sand deposit at the defective location. Hence, finite element analysis using ANSYS–Autodyn software is done to understand the barrel failure during sand ingress. In this analysis, the barrel and projectile are modeled using 3D brick elements and the sand particles are modeled using SPH approach. The present work involves a series of simulation for studying the effect of …
Bayesian Model Averaging For Benchmark Dose Analysis In Developmental Toxicology, Eman Khorsheed, Mehdi Razzaghi
Bayesian Model Averaging For Benchmark Dose Analysis In Developmental Toxicology, Eman Khorsheed, Mehdi Razzaghi
Applied Mathematics & Information Sciences
To reduce uncertainty due to model selection when a large number of potential candidate models is available, the use of Bayesian Model Averaging (BMA) has emerged as an important tool. As known, the BMA methodology is a coherent approach since we can express the desired quantities as a weighted average of model specific quantities with the weights determined based on how much the data supports each model. In toxicological studies, a wide range of statistical models have been utilized for dose-response modeling and risk assessment with no particular model receiving a universal acceptance. Here, we consider the application of BMA …
On Distributions Of Order Statistics For Nonidentically Distributed Variables, Fahrettin Ozbey, Mehmet Gungor, Yunus Bulut
On Distributions Of Order Statistics For Nonidentically Distributed Variables, Fahrettin Ozbey, Mehmet Gungor, Yunus Bulut
Applied Mathematics & Information Sciences
In this study, distribution and probability density functions of any d order statistics of innid continuous random variables are expressed. Then, some results connecting distributions of order statistics of innid random variables to that of order statistics of iid random variables are given.
Common Fixed-Point Theorems Of Caristi-Type Mappings By Using Its Absolute Derivative, Mohamad Muslikh, Adem Kılıcman, Siti Hasana Bt Sapar, Norfifah Bt Bachok
Common Fixed-Point Theorems Of Caristi-Type Mappings By Using Its Absolute Derivative, Mohamad Muslikh, Adem Kılıcman, Siti Hasana Bt Sapar, Norfifah Bt Bachok
Applied Mathematics & Information Sciences
In this article, we introduce common fixed-point theorems of Caristi-type mappings by using the absolute derivative of the mapping as a generator of its Caristi-type maps. The common fixed-point theorem that we obtain covers the single-valued and set-valued mappings. Some of the examples are given to support usability of our result.
Exponential Stability For A Transmission Problem With Locally Indirect Stabilization, Carlos A. Raposo, Manoel J. Santos
Exponential Stability For A Transmission Problem With Locally Indirect Stabilization, Carlos A. Raposo, Manoel J. Santos
Applied Mathematics & Information Sciences
In this manuscript we consider the transmission problem, in one space dimension, for linear dissipative waves with locally indirect stabilization. We study the wave propagation in a medium with a component with attrition and another being simply elastic. We show that for this type of material, the dissipation produced by the frictional part is strong enough to produce exponential decay of the solution.
Heterogeneous Image Transcoding Using Interconversion Matrices, V. Mohan, P. Shanmugapriya, Y. Venkataramani
Heterogeneous Image Transcoding Using Interconversion Matrices, V. Mohan, P. Shanmugapriya, Y. Venkataramani
Applied Mathematics & Information Sciences
We propose a new scheme for heterogeneous transcoding of image files from JPEG to JPEG 2000 format with directional filter analysis and modified SPIHT coding scheme. The optimal parameters QF out(I,D) and Z(I,D) that produce high perceived quality in terms of SSIM are selected using a prediction algorithm. Furthermore, the heterogeneous image transcoding are performed using inter-conversion matrices which reduce the computational overheads. The resultant coefficients, using BDCT to DWT, are analyzed by Directional Filter banks (DFBs). The DWT-DFB coefficients are encoded via progressive SPIHT algorithm to meet the devices constraints. In the experimental set up, the quality of reconstructed …
An Analysis On Web-Service-Generated Data To Facilitate Service Retrieval, I.R. Praveen Joe, P. Varalakshmi
An Analysis On Web-Service-Generated Data To Facilitate Service Retrieval, I.R. Praveen Joe, P. Varalakshmi
Applied Mathematics & Information Sciences
In this paper, a novel layered clustering approach is proposed to cluster web services in order to facilitate web service selection process. The process of web service selection from a rapidly growing number of functionally similar services in the internet, results in an increase of service discovery cost, transforming data time between services and service searching time. Though suitable technologies for web services clustering are being developed, blending neural networks and swarm-based algorithms is not prevailing. A novel two-phase clustering approach involving ART (Adaptive Resonance Theory) network for primary clustering with functional data and swarm algorithms (BOIDs, ABC and PSO) …
Modelling Of Hchs System For Optimal E-O-L Combination Section And Disassembly In Reverse Logistics, T. Sathish, P. Periyasamy
Modelling Of Hchs System For Optimal E-O-L Combination Section And Disassembly In Reverse Logistics, T. Sathish, P. Periyasamy
Applied Mathematics & Information Sciences
We propose a novel strategy for the multi period disassembly in reverse logistic. The proposed system combines the naturalinspired cuckoo search with the artificial inspired harmony search algorithms for disassembly of end of life products. The new strategy is tested by considering a test case and its performance is compared with the conventional optimization techniques. The obtained results are validated based on the cost, disassembly time, and execution time. It is shown that the performance of proposed technique outperforms the conventional technique. Hence the proposed hybrid metaheuristics approach is suitable for the real time reverse logistics management.
Multiscale Wavelet Transformation And Amplitude Zone Time Epoch Coding For Ecg Data Extraction And Compression, S. Velmurugan, A. Mahabub Basha
Multiscale Wavelet Transformation And Amplitude Zone Time Epoch Coding For Ecg Data Extraction And Compression, S. Velmurugan, A. Mahabub Basha
Applied Mathematics & Information Sciences
Tele-health monitoring plays a vital role in healthcare applications. Electrocardiogram (ECG) signals are an essential part in diagnosis and analysis of cardiac diseases. It consumes large amount of time and space during transmission and data storage. To reduce the time and space complexity, an Integrated Multiscale Gabor Wavelet Transformation (IMAGWT) and Amplitude Zone Time Epoch Coding (IMGWT–AZTEC) mechanism is introduced. Multiscale Approximated Gabor Wavelet Transformation (MAGWT) is used in IMAGWT–AZTEC mechanism to decompose the ECG signals into many sub-bands and to extract the P, T waves and QRS complex. After extraction of signals, AZTEC and decoding techniques are introduced for …
An Enhanced Proactive Transmission Protocol For Optical Burst Switching Networks, S. Manisekar, J. Arputha Vijaya Selvi
An Enhanced Proactive Transmission Protocol For Optical Burst Switching Networks, S. Manisekar, J. Arputha Vijaya Selvi
Applied Mathematics & Information Sciences
Due to provision of connectionless and tremendous transmission service, Optical Burst Switching Network (OBSN) becomes the most challenging task for enabling a reliable as well as congestion free communication. So, different communication protocols are developed in the traditional works, but it has the following drawbacks: inefficient, reduced throughput, increased delay, and bandwidth consumption. To solve these problems, this new protocol, namely, Multiple Path Transmission Protocol (MPTP) is developed in this work. It integrates the functionalities of multiple-path routing and Transmission Control Protocol (TCP) Vegas. Here, the wavelength reservation is mainly performed for channel reservation during communication. Moreover, the proposed MPTP …
Harmonic Analysis Of Photovoltaic-Fed Symmetric Multilevel Inverter Using Modified Artificial Neural Network, N. I. Maheshwari, M. Chandrasekaran
Harmonic Analysis Of Photovoltaic-Fed Symmetric Multilevel Inverter Using Modified Artificial Neural Network, N. I. Maheshwari, M. Chandrasekaran
Applied Mathematics & Information Sciences
This paper presents a control scheme to eliminate the harmonics in a symmetric cascaded multilevel inverter- connected grid system. It is controlled by modified artificial neural networks and genetic algorithm. The solar photovoltaic acts as input source and the output power is synchronized with grid. The switching angles are acquired such that the fundamental component of output voltage is kept steady and lower- order harmonics are limited or removed. The equipped system is coordinated with sun-oriented photovoltaic framework to decrease the harmonics. Also, this work’s target is to achieve the proposed idea using genetic algorithm strategies to get the ideal …
A New Digital Signature Scheme Based On Chaotic Maps And Quadratic Residue Problems, Nedal Tahat, Mohammad S. Hijazi
A New Digital Signature Scheme Based On Chaotic Maps And Quadratic Residue Problems, Nedal Tahat, Mohammad S. Hijazi
Applied Mathematics & Information Sciences
Securing electronic signature gives the contracting parties, especially the consumer, safety and security, which positively reflects on trade exchange. Digital-signature algorithms can be categorized based on the type of security suppositions, for example discrete logarithm, factorization of hard-problems, and elliptic-curve cryptography, which are all currently believed to be unsolvable in a reasonable time period. In recent years, a variety of chaotic cryptographic schemes have been proposed. The idea of chaotic systems with applications to cryptography has received a great deal of attention from researchers from a variety of disciplines. Therefore, in this paper, we propose a new signature scheme based …
Ostrowski-Type Inequalities For Functions Whose Derivative Modulus Is Relatively Convex., Miguel Vivas Cortez, Carlos Garcıa, Jorge E. Hernandez H.
Ostrowski-Type Inequalities For Functions Whose Derivative Modulus Is Relatively Convex., Miguel Vivas Cortez, Carlos Garcıa, Jorge E. Hernandez H.
Applied Mathematics & Information Sciences
We found some new Ostrowski-type inequalities for functions whose derivative module is relatively convex, also some others of the same type making use of relatively s−convex functions in the second sense. With these results we obtain generalizations of results found by M. Alomari et. al. using convex and s−convex in the second sense ([1,2]).
Alpha Power Transformed Frechet Distribution, Suleman Nasiru, Peter N. Mwita, Oscar Ngesa
Alpha Power Transformed Frechet Distribution, Suleman Nasiru, Peter N. Mwita, Oscar Ngesa
Applied Mathematics & Information Sciences
The Fr´echet distribution has several applications in different fields of study and is most commonly used for modeling extreme events. In recent time, modifications of the Fr´echet distribution have been proposed to improve its fit when used for modeling lifetime data. In this paper, a new modification called the alpha power transformed Fr´echet distribution is proposed and studied. The parameters of the model are estimated using maximum-likelihood estimation and simulation studies are performed to investigate the properties of the estimators for the parameters. Applications of the model are demonstrated using two-real data sets. Finally, bivariate and multivariate extensions of the …
A Cpw-Fed Hexagonal Antenna With Fractal Elements For Uwb Applications, V. Dinesh, G. Murugesan
A Cpw-Fed Hexagonal Antenna With Fractal Elements For Uwb Applications, V. Dinesh, G. Murugesan
Applied Mathematics & Information Sciences
We propose a coplanar waveguide-fed hexagonal antenna design for ultra wideband applications. The size of the antenna is miniaturized to 25×25×1.588 mm3 and the impedance bandwidth is improved by adding fractal elements at the edges. Parametric analysis of the fractal elements and the impedance behavior is performed to realize the radiation characteristics and bandwidth coverage as required for microwave imaging. The addition of fractal elements introduces multi-resonance at different frequency and covers a large bandwidth of 1.7 GHz–11 GHz respectively. The effect of surface currents in the ground plane reduces the antennas operating bandwidth which is reduced by introducing defective …
Two Dimensional Harmonic Quasi Convex Functions And Integral Inequalities, Muhammad Uzair Awan, Muhammad Aslam Noor, Khalida Inayat Noor
Two Dimensional Harmonic Quasi Convex Functions And Integral Inequalities, Muhammad Uzair Awan, Muhammad Aslam Noor, Khalida Inayat Noor
Applied Mathematics & Information Sciences
The objective of this paper is to introduce the notion of two-dimensional harmonic quasi-convex functions. We obtain some new Hermite-Hadamard type integral inequalities for two-dimensional harmonic quasi convex functions. Some special cases are also discussed.
Performance Analysis Of Surface Roughness Modeling Using Soft Computing Approaches, B. Radha Krishnan, V. Vijayan, G. Senthilkumar
Performance Analysis Of Surface Roughness Modeling Using Soft Computing Approaches, B. Radha Krishnan, V. Vijayan, G. Senthilkumar
Applied Mathematics & Information Sciences
In this paper, classification algorithms are used to classify the test data samples for determining the error rate by comparing its classification response with actual response. In this paper, Random Forest (RF) and Adaptive Neuro Fuzzy Inference System (ANFIS) classification algorithms are used as soft computing techniques to determine the error rate for the prediction of surface roughness of the materials. The parameters feed, depth of cut, speed and mean are extracted from the test sample materials and they are given to classification mode of the ANFIS classifier which produces vision measurement value. The error rate is determined by subtracting …
Study On A New (3+1)-Dimensional Extensions Of The Konopelchenko-Dubrovsky Equation, Abdul-Majid Wazwaz
Study On A New (3+1)-Dimensional Extensions Of The Konopelchenko-Dubrovsky Equation, Abdul-Majid Wazwaz
Applied Mathematics & Information Sciences
In this work we introduce new (3+1)-dimensional extensions of the Konopelchenko-Dubrovsky (KD) equation. We use the simplified Hirota’s method to study these new extensions. We derive the dispersion relation and the multiple soliton solutions for each developed model.
The Optimal Retailer’S Economic Production Quantity (Epq) Policies With Two-Level Trade Credit Under Alternate Due Date Of Payment And Limited Storage Capacity, H. M. Srivastava, Ghi-Feng Yen, An-Kuo Lee, Yi-Xiu Wu, Shy-Der Lin
The Optimal Retailer’S Economic Production Quantity (Epq) Policies With Two-Level Trade Credit Under Alternate Due Date Of Payment And Limited Storage Capacity, H. M. Srivastava, Ghi-Feng Yen, An-Kuo Lee, Yi-Xiu Wu, Shy-Der Lin
Applied Mathematics & Information Sciences
In recent years, many researchers investigated and developed the Economic Production Quantity (EPQ) model under permissible delay in payment. There are two payment terms which are usually being used. If a customer buys one item from a retailer at time t belonging to the time-interval [0,N], then the customer will have a trade-credit period N −t and will make the payment at time N. The other payment term (alternate due date) is when a customer buys one item from a retailer at time t belonging to [0,T], the customer will have a trade-credit period N and will make the payment …
Unitary Aberrations On Pixellated Screens, Kenan Uriostegui, Kurt Bernardo Wolf
Unitary Aberrations On Pixellated Screens, Kenan Uriostegui, Kurt Bernardo Wolf
Applied Mathematics & Information Sciences
The finite oscillator based on the Lie group of spin U(2) provides a model for finite one-dimensional (1D) arrays of N = 2 j +1 pixels, for j integer or half-integer which, as j → ¥, deforms to the continuous 1D model of geometric optics. Translations, linear transformations and aberrations in the latter are canonical and have their N×N unitary counterparts in the former. Since in U(N) there are only N2 independent transformations, we identify the finite counterparts of translations, linear transformations and aberrations within the finite model, applicable to the correction of aberrated images or signals on N-pixel linear …
Temperature Forecasting: A Comparison Between Parametric And Non-Parametric Models, Alaa Sheta, Ajay Katangur, Abdelkarim Baareh
Temperature Forecasting: A Comparison Between Parametric And Non-Parametric Models, Alaa Sheta, Ajay Katangur, Abdelkarim Baareh
Applied Mathematics & Information Sciences
The development of accurate temperature prediction models is essential for not only human life but also for agricultural, animal life, tourism, and many others. Power consumption and achieving energy efficiency in buildings also depends on temperature. Although modeling-based regression is one of the most popular approaches, it still suffers from many difficulties related to the number of available measurements, the order of the model and the non-linearity of the data. In this paper, we provide a comparison between parametric and non-parametric models for temperature forecasting. We propose three-model structures to estimate the temperature in Mumbai, the business capital of India. …
Exponentially-Modified Logistic Distribution With Application To Mining And Nutrition Data, Jimmy Reyes, Osvaldo Venegas, Hector W. Gomez
Exponentially-Modified Logistic Distribution With Application To Mining And Nutrition Data, Jimmy Reyes, Osvaldo Venegas, Hector W. Gomez
Applied Mathematics & Information Sciences
In this work we introduce a modification of the exponentially-modified Gaussian distribution. This new distribution is obtained by combining a logistic distribution with an exponential distribution, and is more flexible than other similar distributions. We provide a closed expression for the density function and obtain some important properties useful for making inferences, such as moment estimators and maximum likelihood estimators. By way of illustration, and using real data to show the effectiveness of the new model, we compare it with known related models, showing that the new model achieves a better fit.
Ostrowski Type Inequalities For Functions Whose Second Derivatives Are Convex Generalized., Yenny Rangel-Oliveros, Miguel Vivas-Cortez
Ostrowski Type Inequalities For Functions Whose Second Derivatives Are Convex Generalized., Yenny Rangel-Oliveros, Miguel Vivas-Cortez
Applied Mathematics & Information Sciences
In this paper we introduce the notion of s−j−convex functions as generalization of convex functions. Some basic results under various conditions for the function j are investigated. Moreover, we establish Ostrowski type inequalities for twice differentiable mappings which are s−j−convex.
The Effect Of Immigration On The Persistence And Ergodicity Of A Stochastic Sis Model For Transmission Of Disease, Youness El Ansari, Ali El Myr, Lahcen Omari, Aadil Lahrouz
The Effect Of Immigration On The Persistence And Ergodicity Of A Stochastic Sis Model For Transmission Of Disease, Youness El Ansari, Ali El Myr, Lahcen Omari, Aadil Lahrouz
Applied Mathematics & Information Sciences
This paper aims to study the dynamic behavior of a stochastic SIS (Susceptible-Infected-Susceptible) epidemic model with varying population size and constant flow of new members of whom a specified fraction is infective. First, we show that the solution is stochastically ultimately bounded and permanent. Then, we investigate the persistence in the mean of the variable I(t) (the number of infected members). Next, we use the Markov semigroups theory to investigate the ergodicity of the solution. Mainly, we show that a stationary distribution for the solution always exists for all values of the parameters in this model. Finally, Numerical simulations are …
Non-Linear Machine Learning Techniques For Multi-Label Image Data Classification, D. Senthilkumar, A. K. Reshmy, M. G. Kavitha
Non-Linear Machine Learning Techniques For Multi-Label Image Data Classification, D. Senthilkumar, A. K. Reshmy, M. G. Kavitha
Applied Mathematics & Information Sciences
In this paper, we propose non-linear Machine Learning Techniques (MLT) for Multi-label Image Classification (MLIC) problems. Multi-label Learning requires MLT to identify the complex non-linear relationship between the features and class labels. Also, Multi-label data degrades the performance of the classifiers and processing of this data with a large number of features is too complex while using traditional methods. Therefore, we propose two approaches namely ensemble Deep Learning Network (DLN) and Multivariate Adaptive Regression Splines (MARS) for MLIC. The experimental results show that the proposed (DLN and MARS) algorithms achieves a superior predictive performance rate of 94.77% and 81.68% respectively, …
Deep Neural Network Based Speaker Verification System Using Features From Glottal Activity Regions, P. Shanmugapriya, V. Mohan, T. Jayasankar, Y. Venkataramani
Deep Neural Network Based Speaker Verification System Using Features From Glottal Activity Regions, P. Shanmugapriya, V. Mohan, T. Jayasankar, Y. Venkataramani
Applied Mathematics & Information Sciences
We propose a Deep Neural Network (DNN)-based Speaker Verification (SV) system using features derived from Glottal Activity (GA) regions. Glottal activity regions are detected through Glottal Closure Instant (GCI), Normalized Autocorrelation Peak Strength (NAPS) and Higher Order Statistics (HOS) from speech signal. For the detection of GA regions, the speech signal is represented in terms of Zero Frequency Filtered Signal (ZFFS) and Integrated Linear Predicted Residual (ILPR). Mel Frequency Cepstral Coefficient (MFCC) and Wavelet Transformed Residual Coefficients extracted from the detected GA regions are used for analysing the performance of speaker verification system based on DNN and i-vector DNN. The …
Spatial Vector Autoregressive Model With Calendar Variation For East Java Inflation And Money Supply, Eni Sumarminingsih, Setiawan Matoha, Agus Suharsono, Budi Nurani Ruchjana
Spatial Vector Autoregressive Model With Calendar Variation For East Java Inflation And Money Supply, Eni Sumarminingsih, Setiawan Matoha, Agus Suharsono, Budi Nurani Ruchjana
Applied Mathematics & Information Sciences
Inflation is a macroeconomic variable that can indicate the economy of a country, so it needs to be managed to a stability point. One way to control inflation is to regulate the money supply that was made by the Central Bank. This shows the relationship between inflation and money supply. In addition, inflation and money supply have space-time relationship. One character of inflation and money supply in Indonesia is the effect of Eid al Fitr on these two variables. The purpose of this study is to model the inflation and the money supply simultaneously with the effect of calendar variation …
Parallel Concatenation Of Ldpc Codes With Lte Modulation Schemes, Mohanad Alfiras, Wael A. H. Hadi, Amjad Ali Jassim
Parallel Concatenation Of Ldpc Codes With Lte Modulation Schemes, Mohanad Alfiras, Wael A. H. Hadi, Amjad Ali Jassim
Applied Mathematics & Information Sciences
low-density parity checks (LDPC) codes have received significant interest for communication system applications for their performance as error correction codes. LDPC codes outperform parallel turbo codes, which are based on a convolutional encoder. This paper proposes parallel concatenation of LDPC codes of small length. The proposed method employs a summation of the estimation output technique from the LDPC decoder. The study performed simulations to evaluate the parallel concatenation of two and three irregular LDPC codes at code rates of 1/2 and 1/3. The simulations evaluated a low number of iterations about ten only for each decoder. We further compare the …
Simplified Equation For Received Power Calculation Of The Lte Signal In 2.13 Ghz., Edgar Alan Degollado-Rea, Sergio Vidal-Beltran, Jose Luis Lopez-Bonilla
Simplified Equation For Received Power Calculation Of The Lte Signal In 2.13 Ghz., Edgar Alan Degollado-Rea, Sergio Vidal-Beltran, Jose Luis Lopez-Bonilla
Applied Mathematics & Information Sciences
In this paper, an analysis of path loss is made using an empirical method that allows obtaining a simplified equation for the received power calculation, which depends on fewer variables than the propagation models traditionally-used in 2.13 GHz. The final part is about the simplified equations obtained for two areas of M´exico city, which can be used to generalize the estimation of losses to other areas with similar characteristics.