Entropy-based Algorithm Research Articles

Research on entropy-based collaborative filtering algorithm and personalized recommendation in e-commerce

Based on the introduction to the user-based and item-based collaborative filtering algorithms, the problems related to the two algorithms are analyzed, and a new entropy-based recommendation algorithm is proposed. Aiming at the drawbacks of traditional similarity measurement methods, we put forward an improved similarity measurement method. The entropy-based collaborative filtering algorithm contributes to solving the cold-start problem and discovering users’ hidden interests. Using the data selected from Movielens and Book-Crossing datasets and MAE accuracy metric, three different collaborative filtering recommendation algorithms are compared through experiments. The experimental scheme and results are discussed in detail. The results show that the entropy-based algorithm provides better recommendation quality than user-based algorithm and achieves recommendation accuracy comparable to the item-based algorithm. At last, a solution to B2B e-commerce recommendation applications based on Web services technology is proposed, which adopts entropy-based collaborative filtering recommendation algorithm.

Entropy-based associative classification algorithm for mining manufacturing data

This paper presents a new associative classification algorithm for data mining. The algorithm uses elementary set concepts, information entropy and database manipulation techniques to develop useful relationships between input and output attributes of large databases. These relationships (knowledge) are represented using IF–THEN association rules, where the IF portion of the rule includes a set of input attributes features and THEN portion of the rule includes a set of output attributes that represent decision outcome. Application of the algorithm is presented with a thermal spray process control case study. Thermal spray is a process of forming a desired shape of material by spraying melted metal on a ceramic mould. The goal of the study is to identify spray process input parameters that can be used to effectively control the process with the purpose of obtaining better characteristics for the sprayed material. Detailed discussion on the source and characteristics of the data sets is also presented.

A novel conformation optimization model and algorithm for structure-based drug design

In this paper, we present a multi-scale optimization model and an entropy-based genetic algorithm for molecular docking. In this model, we introduce to the refined docking design a concept of residue groups based on induced-fit and adopt a combination of conformations in different scales. A new iteration scheme, in conjunction with multi-population evolution strategy, entropy-based searching technique with narrowing down space and the quasi-exact penalty function, is developed to address the optimization problem for molecular docking. A new docking program that accounts for protein flexibility has also been developed. The docking results indicate that the method can be efficiently employed in structure-based drug design.

X-ray absorption tomography of a high-pressure metal-halide lamp with a bent arc due to Lorentz-forces

Due to its practical importance x-ray absorption (XRA) tomography is becoming a widely applied technique for reconstruction of gas temperature distribution in high intensity discharge lamps. Most of the previous papers have dealt with the cylindrically symmetric plasmas that normally occur when lamps are burned vertically. In this work, a XRA tomographic diagnostic is applied for the first time to a high-pressure metal-halide lamp with a bent arc. X-ray lamp-off and lamp-on measurements have been made for two orthogonal directions. The maximum entropy-based algorithm was applied for the reconstruction of local absorption coefficients which are directly related to the local values of mercury density and plasma temperature. For correct reconstruction of the temperature profile at the edges a new numerical strategy based on the adaptive regularization parameter is developed. This strategy allows use of the measured value of the wall temperature as a reference point to obtain a quantitative temperature distribution.

A cross entropy based algorithm for reliability problems

The Cross Entropy method has recently been applied to combinatorial optimization problems with promising results. This paper proposes a Cross Entropy based algorithm for reliability optimization of complex systems, where one wants to maximize the reliability of a system through optimal allocation of redundant components while respecting a set of budget constraints. We illustrate the effectiveness of the proposed algorithm on two classes of problems, software system reliability optimization and complex network reliability optimization, by testing it on instances from the literature as well as on randomly generated large scale instances. Furthermore, we show how a Cross Entropy-based algorithm can be fine-tuned by using a training scheme based upon the Response Surface Methodology. Computational results show the effectiveness as well as the robustness of the algorithm on different classes of problems.

Development of entropy based algorithm for cardiac beat detection in 12-lead electrocardiogram

This paper presents an entropy-based algorithm for the detection of QRS complexes (cardiac beat) in the 12-lead Electrocardiogram (ECG) using support vector machine (SVM). Digital filtering techniques are used to remove power line interference and baseline wander in the ECG signal. Entropy criterion is used to enhance the QRS complexes. SVM is used as a classifier to delineate QRS and non-QRS regions. The performance of the proposed algorithm was tested using 12-lead real ECG recordings from the standard CSE ECG database. The numerical results indicated that the algorithm achieved 99.93% of detection rate. The percentage of false positive and false negative is 0.87% and 0.06%, respectively. The performance of the proposed algorithm is found to be better than published results of the other QRS detectors tested on the same database.

A simple entropy-based algorithm for planar point location

Given a planar polygonal subdivision S , point location involves preprocessing this subdivision into a data structure so that given any query point q , the cell of the subdivision containing q can be determined efficiently. Suppose that for each cell z in the subdivision, the probability p z that a query point lies within this cell is also given. The goal is to design the data structure to minimize the average search time. This problem has been considered before, but existing data structures are all quite complicated. It has long been known that the entropy H of the probability distribution is the dominant term in the lower bound on the average-case search time. In this article, we show that a very simple modification of a well-known randomized incremental algorithm can be applied to produce a data structure of expected linear size that can answer point-location queries in O ( H ) average time. We also present empirical evidence for the practical efficiency of this approach.

Entropy-Based Time Window for Artifact Removal in UWB Imaging of Breast Cancer Detection

In this letter, we propose an entropy-based time window function for artifact removal in ultra-wideband (UWB) imaging of breast cancer detection. The artifact is due to the incident and skin-breast backscattered signal in such an imaging system. Observing that artifacts at different antennas are very similar, we define an entropy function in the antenna domain to measure the similarity of different antenna signals and design a rectangular window function to eliminate the artifacts. This window function is applied to the computed finite-difference time-domain (FDTD) data of the breast model with a tumor embedded. Compared with the Wiener filtering algorithm, this entropy-based algorithm is computationally simple. In addition, it requires no prior knowledge about the breast and the tumor and brings no distortion to the tumor reflection

Tomographic diagnostics of high-frequency electrodeless lamps in argon–mercury mixtures

Tomographic reconstruction of spatial profiles of the mercury atom density in the excited state 7 3S1 in high-frequency electrodeless lamps (HFELs) has been performed. The measurements of the Hg 546.1 nm line emission intensity have been made for the HFELs in argon–mercury mixture depending on the operation regime with different cold spot temperatures in the range 31–98 °C. The maximum entropy-based algorithm was applied for the reconstruction of local emission coefficients from the integrated intensities. The emission coefficients are directly related to the local values of the mercury atom density in the excited state 7 3S1, the upper state of the 546.1 nm transition. Such an investigation has been performed first for the HFEL. We have found that the emitting mercury atoms in the state 7 3S1 are concentrated in a thin layer located close to the lamp wall. The radial profiles have demonstrated a strong depletion of the population density in the state 7 3S1 from the lamp centre at high generator currents and low mercury vapour density. The obtained results are analysed theoretically in the context of the radial cataphoresis phenomenon. We found a qualitative agreement between the reconstructed density profiles and theoretical model predictions.

Measurement of climate complexity using sample entropy

A climate system is a complex nonlinear system. Estimation of the complexity is of great interest in climatic forecast and prediction. In this paper, we propose the use of sample entropy (SampEn), an entropy-based algorithm, to measure the complexity of daily temperature series. Estimations of SampEn were calculated for 50 meteorological stations in the mountains of Southwest China, particularly in Yunnan Province. On the basis of these data, stations were grouped in climatically homogenous regions (climate provinces), and the spatial pattern of SampEn for each climate province was investigated. The SampEn value of spatial distribution of climate provinces reflects the varying degree of influence of the monsoonal air masses. High SampEn values occur in interactive regions of different air masses, owing to large regional differences in weather processes, while the southwest region is under the influence of the Southwest Monsoon leading to a homogenous climatic environment, low SampEn values and small spatial variations of SampEn. The results suggest that SampEn is an alternative nonlinear approach for analyzing and predicting complexity of climatic time series. Copyright (C) 2006 Royal Meteorological Society.

Assessment of the effect of land use patterns on hydrologic landscape functions: a comprehensive GIS‐based tool to minimize model uncertainty resulting from spatial aggregation

AbstractIt is a common technique to predict hydrologic effects of land use changes by simulation models. On the scale of watersheds, hydrologic models comprise different approaches for many hydrologic components. One important source of model uncertainty results from errors of measured input data and their spatial aggregation. Whereas simulations on the basis of measured land use can be validated by measuring resulting hydrologic components, model uncertainty on land use scenarios cannot be quantified because of their virtuality. An entropy‐based algorithm is proposed and validated, which should serve for the identification of an appropriate level of spatial aggregation. This technique is statistically based and applicable for common methods of subwatershed and hydrologic response unit (HRU) delineation. The main criterion of this algorithm is the heterogeneity of data distribution within the investigated watershed, which might vary for different land use scenarios.For spatial output analysis of a long‐term hydrologic model, a reassignment technique of statistically distributed output values to certain grids was developed. It combines efficient model calculations for aggregated spatial areas without losing spatial output heterogeneity. This procedure enables model validation on a grid‐based resolution and serves for the identification of spatial differences of uncertainty. Besides these new approaches for spatial analysis, different common tools for data pre‐ and post‐processing were integrated into the comprising software package IOSWAT, linked to a GIS. The results of the new approaches prove them to be valid for different watersheds and hydrologic conditions. The application of the comprehensive software package is presented in Fohrer et al. (this issue). Copyright © 2005 John Wiley & Sons, Ltd.

Event detection and period extraction using multi-scale symmetry and entropy

We present a system for detecting discrete periodic events over a short interval and in the presence of interference. In the first stage symmetries are identified using scale-space representation. This process detects signal events with a low signal-to-noise ratio but has the potential to introduce a number of false responses. This process is followed by an entropy-based algorithm that can robustly extract periodicities from a set of observed discrete events in the presence of a large number of false alarms. The event detection and period extraction processes have a low computational cost and can extract signal periodicity after a short observation time. This scheme was evaluated against four previously reported methods. Results demonstrate that the period extraction algorithm presented here is more reliable than three of the previously reported algorithms. The reliability of the algorithm presented here was similar to that of the fourth method but the computational cost was much less.

An in-silico method for prediction of polyadenylation signals in human sequences.

This paper presents a machine learning method to predict polyadenylation signals (PASes) in human DNA and mRNA sequences by analysing features around them. This method consists of three sequential steps of feature manipulation: generation, selection and integration of features. In the first step, new features are generated using k-gram nucleotide acid or amino acid patterns. In the second step, a number of important features are selected by an entropy-based algorithm. In the third step, support vector machines are employed to recognize true PASes from a large number of candidates. Our study shows that true PASes in DNA and mRNA sequences can be characterized by different features, and also shows that both upstream and downstream sequence elements are important for recognizing PASes from DNA sequences. We tested our method on several public data sets as well as our own extracted data sets. In most cases, we achieved better validation results than those reported previously on the same data sets. The important motifs observed are highly consistent with those reported in literature.

ENPCO: an entropy-based partition-codec algorithm to reduce power for bipartition-codec architecture in pipelined circuits

This work proposes a new algorithm to synthesize low power bipartition-codec architecture for pipelined circuits. The bipartition-codec architecture has been introduced as an effective power reduction technique for circuit design. The entropy-based partition-codec (ENPCO) algorithm extends this approach as it optimizes for both: power and area. It uses entropy as a criterion to balance between power and area. The ENPCO algorithm is composed of two phases: first, it clusters the output vectors with high occurrence into a group, moving all remaining output vectors into another group. The first group will be encoded in order to save power. Secondly, based on circuit entropy, output patterns are moved between both groups in order to balance power consumption and area overhead. A number of Microelectronic Center of North Carolina (MCNC) benchmarks were used to verify the effectiveness of our algorithm. Results demonstrate that ENPCO algorithm can achieve low power with less area overhead than the single-phase algorithm introduced previously by Shanq-Jang Ruan et al. (1999).

An entropy-based algorithm for detecting clusters of cases and controls and its comparison with a method using nearest neighbours

A new method for detecting disease clustering based on entropy is presented. For this method cases and controls are plotted on a map. The map is divided into regions. The entropy of the space is calculated as the log of the number of possible ways of placing the cases and controls in the various regions given the total number of cases and controls and the number of cases and controls in each region. The power of the entropy technique is tested against the power of the nearest neighbour technique (NNT). The entropy method is shown to be substantially more powerful than the NNT when there is more than one cluster in the space or when the clusters are near the boundary of the space.

Guest editorial

This paper presents a numerical method to compute the cable stays prestressing forces on concrete cable-stayed bridges to achieve the desired final geometry. The procedure includes a discrete direct sensitivity analysis module and an entropy-based algorithm for structural optimization. The structural analysis includes the most relevant effects namely the concrete time-dependent effects, the construction sequence and the geometrical nonlinearities. Numerical examples are presented and the influence of these effects in the cable forces is analyzed. The results show the relevance of considering these effects to find adequate values of the cable forces.

Electrospray ionization mass spectrometry: Deconvolution by an entropy-based algorithm

Summary The more demanding complication of the entropy processing of ES1 spectra is the presence of large, unresolved backgrounds on which many individual peaks are superimposed. In working with these spec- tra, we generally employ a combination of Fourier deconvolution and baseline subtraction. Such steps can be managed as a preftlter to the entropy deconvo- lution since the optimization (Fourier cutoff, exponen- tial factor) does not require knowledge of the entropy fit parameters. Figure 4 illustrates the application of these strategies to an ES1 mass spectrum of a gly- copeptide with two glycosolation sites. ties of the data. rates which are modeled as multiply charged Gauss- ian ion profiles. The other noise is shot noise corre- The computer spectra were generated by a numeri- cal simulation of a Poisson process in which a param- sponding to the finite ion counts. Of course, back- eter representing the expected number of counts for each mass value mi is defined. Independent scans are ground peaks or chemical “noise” are, from the obtained by sampling the Poisson distribution corre- sponding to this parameter [7]. In these simulations standpoint of the simulation, just additional real noise can arise in two ways. The ftrst is the relative fraction of counts that are real ions to counts that are peaks. representing background. From the standpoint of the simulation program they are distinguished only be- cause the background has a mass-independent arrival rate while the signal corresponds to mass-dependent This report has discussed a novel algorithm for ex- tracting parent masses from spectra containing multi- ply charged ions, a common feature of ESI. The algo- rithm works with raw data and does not require the

Entropy-Based Algorithm For Reducing Artifacts In Image Restoration

We introduce a simple algorithm to reduce artifacts that often appear in image restoration techniques such as Wiener filtering. The algorithm starts with the inverse filter solution and iteratively calculates the correction term. At each iteration we use an entropy gradient and an analytically calculated step size. The algorithm uses two Fourier transforms per iteration. We show both 1 -D and 2-D examples to illustrate the algorithm.