Large Number Of Attributes Research Articles

Derived fuzzy importance of attributes based on the weakest triangular norm-based fuzzy arithmetic and applications to the hotel services

The correlation between the performance of attributes and the overallsatisfaction such as they are perceived by the customers is often used tocalculate the importance of attributes in the crisp case. Recently, the methodwas extended, based on the standard Zadeh extension principle, to the fuzzycase, taking into account the specificity of the human thinking. Thedifficulties of calculation are important and only approximations of theanalytic results can be obtained. In the present paper we give a simplifiedand exact method to compute the derived importance of the attributes in thecase of input data given by triangular fuzzy numbers. The effectivecalculation is based on the $T_{W}$-extension principle and it uses reasonablecomputer resources even if a large number of attributes and customers isconsidered. The proposed derived method is later on compared with othermethods of calculation of the fuzzy importance of attributes. The results ofa survey with respect to the quality of hotel services in Oradea (Romania)are subject to the application of the proposed method.

An integrated algorithm for the optimal design of stated choice experiments with partial profiles

Stated choice experiments are conducted to identify the attributes that drive people’s preferences when choosing between competing options. They are widely used in transportation in order to support the decision making of companies and governmental authorities. A large number of attributes might increase the complexity of the choice task in a choice experiment, and have a detrimental effect on the quality of the results obtained. In order to reduce the cognitive effort required by the experiment, researchers may resort to experimental designs where the levels of some attributes are held constant within a choice situation. These designs are called partial profile designs. In this paper, we propose an integrated algorithm for the generation of D-optimal designs for stated choice experiments with partial profiles. This algorithm optimizes the set of constant attributes and the levels of the varying attributes simultaneously. An extensive computational experiment shows that the designs produced by the integrated algorithm outperform those obtained by existing algorithms, and match the optimal designs that have been analytically derived for a number of benchmark instances. Additionally, we evaluate the performance of the algorithm under varying experimental conditions and study the structure of the designs generated. We also revisit two stated choice experiments in transportation, and describe how the integrated algorithm could help to improve their designs.

Concurrent design of nanofluid for x-abilities using MADM approach

Purpose – Multiple attribute decision making (MADM) is a conceptual agenda used for evaluation and selection of optimal nanofluid to assure best performance of heat exchanger. Most of the studies focus on nanofluids focus on individual ability at one time. Relatively, not even a single study is available for selection of nanofluid for heat exchanger using concurrent design and MADM approach. The purpose of this paper is to propose a concurrent design methodology using MADM approach to assist improved design of heat exchanger concurrently for all the x-abilities in an integrated manner. Design/methodology/approach – A combined methodology of applying MADM approach using concurrent design for x-abilities is called CE-MADM approach. Implementation of nanofluid to improve thermal performance of heat exchanger entails thorough evaluation of nanofluids in various x-abilities (performance, maintenance, thermophysical properties and modelisation) to make exhaustive management decision. Sensitivity analysis is also proposed to study the behaviour of height of variation of density, heat capacity, thermal expansion and thermal conductivity with varying particle volume fraction and variation of relative closeness of available alternates from ideally best possible solution. Findings – MADM approach considering various x-abilities concurrently provide an approach for relative ranking of available nanofluids for optimum performance. Fishbone diagrams of all x-abilities are constructed to identify all the attributes and converge large number of attributes into single numerical index that are concurrently responsible for the cause thus saving time for easy evaluation, comparison and ranking by decision makers. Sensitivity analysis to demonstration height of variation of pertinent attributes with varying particle volume fraction. A MATLAB programming is established to execute calculations involved in the procedure. Originality/value – This paper comprises a predictable and effective mathematical approach to improve design of heat exchanger with nanofluid bearing in mind all the required x-abilities concurrently. This combined approach of CE-MADM is never applied before in the field of nanofluid to predict best possible results in feasible conditions considering all the x-abilities. Sensitivity analysis is also presented from the assumed mathematical equations of thermophysical properties.

NeuroPigPen: A Scalable Toolkit for Processing Electrophysiological Signal Data in Neuroscience Applications Using Apache Pig.

The recent advances in neurological imaging and sensing technologies have led to rapid increase in the volume, rate of data generation, and variety of neuroscience data. This “neuroscience Big data” represents a significant opportunity for the biomedical research community to design experiments using data with greater timescale, large number of attributes, and statistically significant data size. The results from these new data-driven research techniques can advance our understanding of complex neurological disorders, help model long-term effects of brain injuries, and provide new insights into dynamics of brain networks. However, many existing neuroinformatics data processing and analysis tools were not built to manage large volume of data, which makes it difficult for researchers to effectively leverage this available data to advance their research. We introduce a new toolkit called NeuroPigPen that was developed using Apache Hadoop and Pig data flow language to address the challenges posed by large-scale electrophysiological signal data. NeuroPigPen is a modular toolkit that can process large volumes of electrophysiological signal data, such as Electroencephalogram (EEG), Electrocardiogram (ECG), and blood oxygen levels (SpO2), using a new distributed storage model called Cloudwave Signal Format (CSF) that supports easy partitioning and storage of signal data on commodity hardware. NeuroPigPen was developed with three design principles: (a) Scalability—the ability to efficiently process increasing volumes of data; (b) Adaptability—the toolkit can be deployed across different computing configurations; and (c) Ease of programming—the toolkit can be easily used to compose multi-step data processing pipelines using high-level programming constructs. The NeuroPigPen toolkit was evaluated using 750 GB of electrophysiological signal data over a variety of Hadoop cluster configurations ranging from 3 to 30 Data nodes. The evaluation results demonstrate that the toolkit is highly scalable and adaptable, which makes it suitable for use in neuroscience applications as a scalable data processing toolkit. As part of the ongoing extension of NeuroPigPen, we are developing new modules to support statistical functions to analyze signal data for brain connectivity research. In addition, the toolkit is being extended to allow integration with scientific workflow systems. NeuroPigPen is released under BSD license at: https://sites.google.com/a/case.edu/neuropigpen/.

Hybrid local diffusion maps and improved cuckoo search algorithm for multiclass dataset analysis

Data clustering is a meaningful tool that can, help people classify mixed data automatically. With rapid technological development, data in modern applications become large scale and high dimensional. Some original clustering methods are not suitable for complicated datasets. To improve the performance of the popular kernel fuzzy C-means (KFCM), this study proposed a local density adaptive diffusion maps (LDM) technique to obtain a reliable similarity description and dimensionality reduction. To find the valid cluster centroids of the dataset, this study also proposed an improved cuckoo search (ICS) to optimize the unknown parameters of the KFCM model. The ICS algorithm utilized quaternions to represent individuals who will be optimized. Variable step length of Lévy flights and discovery probability were also proposed, which were adjusted by the evolutional ratio of the cuckoo search process. To verify the availability of the ICS, 5 benchmark functions were tested. Finally, the proposed hybrid ICS and LDM based on KFCM (ICS-LDM-KFCM) was used to identify 4 standard artificial and 6 real world datasets. Compared with other clustering methods, the proposed method obtained more accurate results. This method is verified to be more suitable for complicated datasets with large number of attributes and clusters.

The Incorporation of Passenger Connectivity and Intermodal Considerations in Intercity Transport Planning

Congested road and air networks, coupled with the idea that transport systems should be balanced, have spawned the field of intercity intermodal passenger transport which is based on travel across connected networks that is as seamless as possible. There is a very large number of attributes that affect the objective and subjective quality of travel. Intercity public transport planning therefore requires an excellent understanding of connections between different services and how the quality of these connections affects passengers, operators and the society at large. This article reviews the vast literature over the last 20 years from Europe, North America and Asia that is related to intercity travel, the connectivity of transport and cooperation between different modes of transport and presents a simple yet powerful way to perceive connected transport systems. The article then reviews planning practice at a variety of states worldwide, and suggests a set of focus areas of research or gaps that once filled, the authors expect, will allow further development of connected intercity passenger travel.

ISPM: Improved Snow Prediction Model to Nowcast Snow/No-Snow

Till date, many of the practioners, meteorologists, researchers, academicians, scientists across the globe proposed many methodologies and tools to nowcast snow/no-snow using satellite imagery, radar imagery, physical instruments, various algorithms, models and so on, adding to it some researchers estimated the amount of snow while some researchers detected the density of snow and few discriminated the differences between wet snow and dry snow. The main crux of the present research is to nowcast the presence of snow/no-snow more accurately by making use of historical weather datasets by adopting decision trees. In this paper, we are proposing a new algorithm Improved Snow Prediction Model (ISPM), an improvement to our earlier algorithms Snow Prediction Model (SPM), Improved Supervised Learning in Quest (ISLIQ), Supervised Learning using Gain Ratio as Attribute Selection Measure (SLGAS) and Supervised Learning using Entropy as Attribute Selection Measure (SLEAS). The ISPM algorithm out performs in terms of various performance measures like sensitivity, specificity, precision, dice, error rate and accuracy when compared with other decision tree models. The proposed method provides less computational complexity by evaluating the interval range, which significantly decreases the number of split points. Experimental results show that the ISPM algorithm scales up well to both large and small datasets with large number of attributes and class labels.

Comparison of intensity scales and CATA questions in new product development: Sensory characterisation and directions for product reformulation of milk desserts

Product development and product optimisation are important activities for all food companies. The methodologies applied for uncovering consumers’ perception during product development are diverse, but many of them share the common underlying assumption that an ideal product exists. One possible approach for product optimisation consists of asking consumers to describe the sensory characteristics of tasted and ideal products. In the Ideal Profile Method (IPM) consumers use intensity scales to describe products. However, this approach can be tedious for consumers when a large number of attributes are considered. Check-all-that-apply (CATA) questions have been reported to be a simple alternative to gather information about consumers’ perception of the sensory characteristics of products and have been previously used to identify ideal products. In this context, the aim of the present work was to compare intensity scales and CATA questions using a case study with vanilla milk desserts. Nine samples of vanilla milk desserts were formulated using a fractional factorial design. A trained assessors’ panel evaluated samples using descriptive analysis. A consumer study with 200 participants was carried out. Consumers were randomly divided into 2 groups: one group evaluated the samples and described their ideal using intensity scales and the second group used CATA questions. Data were analysed using multivariate techniques and partial least squares regression. Results showed that both CATA questions intensity scales provided similar information to descriptive analysis with trained assessors. Sample configurations, description of the ideal product, drivers of liking and directions for product reformulation were similar for both approaches.

Public Preferences for Prioritizing Preventive and Curative Health Care Interventions: A Discrete Choice Experiment

BackgroundSetting fair health care priorities counts among the most difficult ethical challenges our societies are facing. ObjectiveTo elicit through a discrete choice experiment the Belgian adult population’s (18–75 years; N = 750) preferences for prioritizing health care and investigate whether these preferences are different for prevention versus cure. MethodsWe used a Bayesian D-efficient design with partial profiles, which enables considering a large number of attributes and interaction effects. We included the following attributes: 1) type of intervention (cure vs. prevention), 2) effectiveness, 3) risk of adverse effects, 4) severity of illness, 5) link between the illness and patient’s health-related lifestyle, 6) time span between intervention and effect, and 7) patient’s age group. ResultsAll attributes were statistically significant contributors to the social value of a health care program, with patient’s lifestyle and age being the most influential ones. Interaction effects were found, showing that prevention was preferred to cure for disease in young adults, as well as for severe and lethal disease in people of any age. However, substantial differences were found in the preferences of respondents from different age groups, with different lifestyles and different health states. ConclusionsOur study suggests that according to the Belgian public, contextual factors of health gains such as patient’s age and health-related lifestyle should be considered in priority setting decisions. The studies, however, revealed substantial disagreement in opinion between different population subgroups.

A comprehensive evaluation of machine learning techniques for cancer class prediction based on microarray data.

Prostate cancer is among the most common cancer in males and its heterogeneity is well known. The genomic level changes can be detected in gene expression data and those changes may serve as standard model for any random cancer data for class prediction. Various techniques were implied on prostate cancer data set in order to accurately predict cancer class including machine learning techniques. Large number of attributes but few numbers of samples in microarray data leads to poor training; therefore, the most challenging part is attribute reduction or non-significant gene reduction. In this work, a combination of interquartile range and t-test is used for attribute reduction. Further, a comprehensive evaluation of ten state-of-the-art machine learning techniques for their accuracy in class prediction of prostate cancer is done. Out of these techniques, Bayes Network outperformed with an accuracy of 94.11% followed by Naïve Bayes with an accuracy of 91.17%.

Attribute Signatures: Dynamic Visual Summaries for Analyzing Multivariate Geographical Data.

The visual analysis of geographically referenced datasets with a large number of attributes is challenging due to the fact that the characteristics of the attributes are highly dependent upon the locations at which they are focussed, and the scale and time at which they are measured. Specialized interactive visual methods are required to help analysts in understanding the characteristics of the attributes when these multiple aspects are considered concurrently. Here, we develop attribute signatures-interactively crafted graphics that show the geographic variability of statistics of attributes through which the extent of dependency between the attributes and geography can be visually explored. We compute a number of statistical measures, which can also account for variations in time and scale, and use them as a basis for our visualizations. We then employ different graphical configurations to show and compare both continuous and discrete variation of location and scale. Our methods allow variation in multiple statistical summaries of multiple attributes to be considered concurrently and geographically, as evidenced by examples in which the census geography of London and the wider UK are explored.

Combining data mining algorithm and object-based image analysis for detailed urban mapping of hyperspectral images

Image classification of roofing types, road pavements, and natural features can assist land-cover maps in further examining the effects of such features on health, pollution, and the microclimate in urban settings. Airborne hyperspectral sensors with high spectral and spatial resolutions can be employed for detailed characterization of urban areas. This study aims to develop a procedure that is instrumental for automated knowledge discovery and mapping of urban surface materials from a large feature space of hyperspectral images. Two different images over Universiti Putra Malaysia (UPM) and Kuala Lumpur (KL), Malaysia, were captured by using hyperspectral sensors with 20 and 128 bands. The images were used to explore the combined performance of a data mining (DM) algorithm and object-based image analysis (OBIA). A large number of attributes were discovered with the C4.5 DM algorithm, which also generated the classification model as a decision tree. The UPM and KL classified images achieved 93.42 and 88.36% overall accuracy. The high accuracy of object-based classification can be linked to the knowledge discovery produced by the DM algorithm. This algorithm increased the productivity of OBIA, expedited the process of attribute selection, and resulted in an easy-to-use representation of a knowledge model from a decision tree structure.

A HYBRID ALGORITHM FOR DECISION TREE GENERATION

The paper discusses the experiments performed with Machine Learning algorithms (ID3, C4.5, Bagged-C4.5, Boosted-C4.5 and Naive Bayes) and an algorithm made on the basis of a combination of genetic algorithms (GA) and ID3. The latter algorithm is implemented as an extension of the MLC++ Library of Stanford University. The behaviour of the algorithm is tested using 24 databases including those with a large number of attributes. It is shown that owing to “hill-climbing” problem solving, the characteristics of the classifier made with the help of the new algorithm became significantly better. The behaviour of the algorithm is examined when constructing pruned classifiers. The ways to improve standard Machine Learning algorithms are suggested.

Resource Aware Monitoring in Distributed System using Tabu Search Algorithm

search algorithm like simulated annealing or evolutionary algorithm or genetic algorithm and guided local search algorithm is a effective solution of optimization problem. This is the most comprehensive combinatorial optimization technique available for treating difficult problems. It is a neighborhood based search method which is very useful in distributed system for monitoring application. Distributed operation of Applications involve: Multiple applications deployed over different sets of hosts e.g. Datacenters. Application State monitored the performance of both systems and applications running on large-scale distributed systems. It is constantly collecting detailed performance attribute values as a large number of nodes & a large number of attributes. Tricky task of Resource aware application state monitoring is the monitoring overlay construction. In this method first, it jointly considers inter-task cost sharing opportunity and node-level resource constraints. Further, it clearly models the per-message processing overhead which can be extensive but is often ignored by earlier works. Second, REMO produces a forest of optimized monitoring trees through iterations of two phases. One stage explores cost-sharing opportunities between tasks, and the other refines the tree with resource-sensitive construction schemes. REMO also included an adaptive algorithm that balances the profit and costs of cover adaptation. This is helpful for large systems with continuously changing monitoring tasks.

Acquisition of optimal attribute subset through genetic algorithm using GNP-based class association rule mining

Attribute selection is a technique to prune less relevant information and discover high-quality knowledge. It is especially useful for the classification of a large database, because the preprocessing of data increases the possibility that predictor attributes given to the mining algorithm become more relevant to the class attribute. In this paper, a method to acquire the optimal attribute subset for the genetic network programming (GNP) based class association rule mining has been proposed, and this attribute selection process using genetic algorithm (GA) leads to a higher accuracy for classification. Class association rule mining through GNP is conducted with a small subset of data rather than the original large number of attributes; thus simple but important rules are obtained for classification while the local optimal problem is avoided. Simulation results with educational data show that the classification accuracy is largely improved from 52.73 to 74.54%, when classification is made using the optimal attribute subset. © 2014 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Scalable CAIM discretization on multiple GPUs using concurrent kernels

Class-attribute interdependence maximization (CAIM) is one of the state-of-the-art algorithms for discretizing data for which classes are known. However, it may take a long time when run on high-dimensional large-scale data, with large number of attributes and/or instances. This paper presents a solution to this problem by introducing a graphic processing unit (GPU)-based implementation of the CAIM algorithm that significantly speeds up the discretization process on big complex data sets. The GPU-based implementation is scalable to multiple GPU devices and enables the use of concurrent kernels execution capabilities of modern GPUs. The CAIM GPU-based model is evaluated and compared with the original CAIM using single and multi-threaded parallel configurations on 40 data sets with different characteristics. The results show great speedup, up to 139 times faster using four GPUs, which makes discretization of big data efficient and manageable. For example, discretization time of one big data set is reduced from 2 h to $$<$$ < 2 min.

CPM Algorithm for Mining Association Rules from Databases of Engineering Design Instances

In this paper, we propose an algorithm for mining associating rules based on transaction combination, attribute combination, pattern comparison and comparative pattern mapping (CPM), aiming at the databases with a large number of attributes but a small number of transactions which are common in engineering design. There are four main steps in the CPM algorithm. First, it scans and expands the database and converts it into a Boolean matrix. Second, it compresses the Boolean matrix to construct a transaction combination matrix (TCM) and an attribute combination matrix (ACM) for further calculation. Third, it generates comparative patterns by comparing every transaction with other transactions in the ACM and stores the comparative patterns in a CP-tree. Finally, it obtains all frequent closed itemsets by picking up the frequent nodes of each branch in the last layer of the CP-tree and eliminating false frequent closed nodes, and all frequent itemsets are found by disassembling the frequent closed itemsets. By comparing CPM with Apriori, FP-Growth, nonordfp and FPgrowth, it is indicated that CPM has a satisfactory performance in mining associating rules from databases with multiple attributes, especially for associating rules with low minimum support degree.

Evaluating the Wald Test for Item‐Level Comparison of Saturated and Reduced Models in Cognitive Diagnosis

This article used the Wald test to evaluate the item‐level fit of a saturated cognitive diagnosis model (CDM) relative to the fits of the reduced models it subsumes. A simulation study was carried out to examine the Type I error and power of the Wald test in the context of the G‐DINA model. Results show that when the sample size is small and a larger number of attributes are required, the Type I error rate of the Wald test for the DINA and DINO models can be higher than the nominal significance levels, while the Type I error rate of the A‐CDM is closer to the nominal significance levels. However, with larger sample sizes, the Type I error rates for the three models are closer to the nominal significance levels. In addition, the Wald test has excellent statistical power to detect when the true underlying model is none of the reduced models examined even for relatively small sample sizes. The performance of the Wald test was also examined with real data. With an increasing number of CDMs from which to choose, this article provides an important contribution toward advancing the use of CDMs in practical educational settings.

Speeding up incremental wrapper feature subset selection with Naive Bayes classifier

This paper deals with the problem of wrapper feature subset selection (FSS) in classification-oriented datasets with a (very) large number of attributes. In high-dimensional datasets with thousands of variables, wrapper FSS becomes a laborious computational process because of the amount of CPU time it requires. In this paper we study how under certain circumstances the wrapper FSS process can be speeded up by embedding the classifier into the wrapper algorithm, instead of dealing with it as a black-box. Our proposal is based on the combination of the NB classifier (which is known to be largely beneficial for FSS) with incremental wrapper FSS algorithms. The merit of this approach is analyzed both theoretically and experimentally, and the results show an impressive speed-up for the embedded FSS process.

Security Solutions for Privacy Preserving Improved Data Mining

Approaches of data analysis in the context of Business Intelligence solutions are presented, when the data is scarce with respect to the needs of performing an analysis. Several scenarios are presented: usage of an initial dataset obtained from primary data as a reference for the quality of the results, enriching the dataset through decoration with derived attributes and enriching the dataset with external data. Each type of dataset decoration is used to improve the quality of the analysis ' results. After being subject to improvement using the presented methods, the improved dataset contains a large number of attributes regarding a subject. As some attributes refer to sensitive information or imply sensitive information about the subject, therefore dataset storage needs to prevent unwanted analysis that could reveal such information. A method for dataset partitioning is presented with respect to the predictive capacity of a set of attributes over a sensitive attribute. The proposed partitioning includes also means to hide the link between the real subject and stored data.Keywords: Business Intelligence, Data Mining, Security, Privacy, Dataset Partitioning, Secret Sharing1 IntroductionBusiness Intelligence (BI) helps the de- cision making process. It relies on various data to offer reports, estimations and support [1], At the base of BI, there are mechanisms for data processing. The results of data anal- ysis are highly dependent on the hypothesis made for the analysis, the quality of the data, the algorithms used for processing. There are areas where for a proper analysis, available data is not sufficient or there is room for im- provement. Such cases are discussed in the following. Presented methods improve the results of analysis. Some derived information is sensitive. We take into account that unau- thorized access to the dataset could trigger the case when analysis may disclose sensitive information about the subject represented by a certain instance in the dataset. Security must be assessed and a solution is needed to cover this risk.2 Problem Formulation of Security Issues Generated by Improved Prediction Using Enlarged DatasetsVarious situations arise when BI tries to solve difficult problems that heavily impact the organization. In telecom industry, efforts are made to predict and to prevent existing customers to migrate to another operator from the market. In the electronic commerce, efforts are made to predict what clients need and to send them incentives, offers and bo- nuses. In stock markets and currency mar- kets, predictions are made to estimate quotes and exchange rates.Business intelligence has its power based on several aspects:* the available data used to derive useful information for the business;* the instruments used to process the available data;* the instruments used to present the re- sults to the decision maker.The available data are mainly taken from the records made by the organization, from in- ternal documents such as contracts, orders, invoices etc., activity history. The available data depends on the degree of how much electronic support is used over manual opera- tions. Ideally, data is stored in files in data- bases and readily available when needed.The instruments used to process data are used to:* transform primary data according to var- ious needs;* apply algorithms in order to obtain re- sults.The power of such instruments is given by the flexibility in accessing heterogeneous da- ta and the quality and variety of algorithms used to obtain results.The instruments used to present the results to the decision maker have to hide the complex processing from the previous type of instru- ments and show useful information in a handful way. Reports have to be easy to un- derstand and conclusions must be highlighted in order to bring business value to the organ- ization.A common case is when a certain variable Y is studied. …