Search

Distributed linguistic representations are powerful tools for modelling the uncertainty and complexity of preference information in linguistic decision making. To provide a comprehensive perspective on the development of distributed linguistic representations in decision making, we present the taxonomy of existing distributed linguistic representations. Then, we review the key elements and applications of distributed linguistic information processing in decision making, including the distance measurement, aggregation methods, distributed linguistic preference relations, and distributed linguistic multiple attribute decision making models. Next, we provide a discussion on ongoing challenges and future research directions from the perspective of data science and explainable artificial intelligence.

The “surprisingly popular” method (SP) of aggregating individual judgments has shown promise in overcoming a weakness of other crowdsourcing methods—situations in which the majority is incorrect. This method relies on participants’ estimates of other participants’ judgments; when an option is chosen more often than the average metacognitive judgments of that option, it is “surprisingly popular” and is selected by the method. Although SP has been shown to improve group decision making about factual propositions (e.g., state capitals), its application to future outcomes has been limited. In three preregistered studies, we compared SP to other methods of aggregating individual predictions about future events. Study 1 examined predictions of football games, Study 2 examined predictions of the 2018 US midterm elections, and Study 3 examined predictions of basketball games. When applied to judgments made by objectively assessed experts, SP performed slightly better than other aggregation methods. Although there is still more to learn about the conditions under which SP is effective, it shows promise as a means of crowdsourcing predictions of future outcomes.

espanolEn el siglo xix, Nueva Zelanda y Uruguay fueron sociedades de nuevo asentamiento europeo, ricas en recursos naturales, que se insertaron en la economia mundial como productoras y exportadoras de bienes ganaderos, pero que mostraron desempenos divergentes a largo plazo. El objetivo del articulo es identificar las principales tendencias historicas del desempeno relativo de la productividad fisica de la ganaderia de ambos paises y comprender los procesos de convergencia-divergencia, a la luz del concepto de trayectoria tecnologica. Para ello, se realiza una revision de los indicadores sinteticos unidad ganadera y carne equivalente, se proponen correcciones a los metodos de agregacion usuales, teniendo en cuenta las especificidades de cada sistema ganadero, y se comparan las trayectorias tecnologicas involucradas en el crecimiento de la productividad de la ganaderia. EnglishIn the nineteenth century, New Zealand and Uruguay were new European settler societies and rich in natural resources. Both countries engaged with the world economy as producers and exporters of livestock products. However, these countries exhibited divergent long-term development. The aim of this article is to identify the main historical tendencies associated with the relative physical productivity of livestock in both countries and to understand convergent-divergent processes in light of their technological trajectories. The main synthetic indicators used in agrarian sciences, such as Livestock Units and Meat Equivalents, are reviewed and a series of corrections that account for the specific features of each agrarian system is proposed for the usual aggregation methods. The technological trajectories that affected livestock productivity growth are also compared. The main results indicate that Uruguayan livestock production fell behind that of New Zealand due to the technological trajectory followed by each country. In the nineteenth century, Uruguay had very favourable natural conditions for livestock production. It did not develop technologies to improve soil and land productivity until the second half of the twentieth century. In contrast, New Zealand began to improve grassland productivity during the nineteenth century in order to increase and diversify livestock production.

Lot layering may help hedge funds improve the alignment between tax outcomes and the economic experience of their investors. Although lot layering is considered by most tax experts to be the most precise method of partnership allocations, this commonly understood precision is reduced upon redemptions due to the cumbersome and uneconomic basis adjustment method stipulated by Treasury regulations. We propose that changes be made to the current regulations that could remedy this problem. Despite its unavoidable deficiency caused by the basis adjustment requirements under the current regulations, we believe that lot layering aligns tax and economics more closely than do any of the aggregation methods presently used by most hedge funds. TOPICS:Wealth management, legal/regulatory/public policy Key Findings • Lot layering is considered by most tax experts to be the most precise method of partnership allocations, where the precision relates to the alignment between tax and economic outcomes of partners in a partnership. • We explain how this commonly understood precision is reduced upon redemptions due to the cumbersome and uneconomic basis adjustment method stipulated by Treasury regulations and propose changes to the current regulations that could remedy this problem. • Despite its unavoidable deficiency caused by the basis adjustment requirements under the current regulations, lot layering both eliminates the need for stuffing allocations and aligns tax and economics more closely than do any of the aggregation methods presently used by most hedge funds.

Recommender Systems are tools designed to help users find relevant information from the myriad of content available online. They work by actively suggesting items that are relevant to users according to their historical preferences or observed actions. Among recommender systems, top- N recommenders work by suggesting a ranking of N items that can be of interest to a user. Although a significant number of top- N recommenders have been proposed in the literature, they often disagree in their returned rankings, offering an opportunity for improving the final recommendation ranking by aggregating the outputs of different algorithms. Rank aggregation was successfully used in a significant number of areas, but only a few rank aggregation methods have been proposed in the recommender systems literature. Furthermore, there is a lack of studies regarding rankings’ characteristics and their possible impacts on the improvements achieved through rank aggregation. This work presents an extensive two-phase experimental analysis of rank aggregation in recommender systems. In the first phase, we investigate the characteristics of rankings recommended by 15 different top- N recommender algorithms regarding agreement and diversity. In the second phase, we look at the results of 19 rank aggregation methods and identify different scenarios where they perform best or worst according to the input rankings’ characteristics. Our results show that supervised rank aggregation methods provide improvements in the results of the recommended rankings in six out of seven datasets. These methods provide robustness even in the presence of a big set of weak recommendation rankings. However, in cases where there was a set of non-diverse high-quality input rankings, supervised and unsupervised algorithms produced similar results. In these cases, we can avoid the cost of the former in favor of the latter.

Crowdsourcing is the process of obtaining input or enlisting the services of a crowd of people with the aim of finding the best solution to a proposed problem. The method with which the collected input is aggregated affects the outcome of the crowdsourcing process. In this paper, we introduce a modelling framework through which we compare several aggregation methods for crowdsourcing systems. The work reveals a surprising result where confidence-related approaches lack in performance when compared to other approaches such as simple plurality voting or approaches which consider respondent competence. This inadequacy stems from a psychological phenomenon brought forth by David Dunning and Justin Kruger related to people’s bias in assessing their own cognitive abilities.

Constructs that reflect differences in variability are of interest to many researchers studying workplace phenomena. The aggregation methods typically used to investigate “variability-based” constructs suffer from several limitations, including the inability to include Level 1 predictors and a failure to account for uncertainty in the variability estimates. We demonstrate how mixed-effects location-scale (MELS) and heterogeneous variance models, which are direct extensions of traditional mixed-effects (or multilevel) models, can be used to test mean (location)- and variability (scale)-related hypotheses simultaneously. The aims of this article are to demonstrate (a) how the MELS and heterogeneous variance models can be estimated with both nested cross-sectional and longitudinal data to answer novel research questions about constructs of interest to organizational researchers, (b) how a Bayesian approach allows for the inclusion of random intercepts and slopes when predicting both variability and mean levels, and finally (c) how researchers can use a multilevel approach to predict between-group heterogeneous variances. In doing so, this article highlights the added value of viewing variability as more than a statistical nuisance in organizational research.

One of the most essential parts of every sentiment analysis application is the aggregation mechanism used to combine results obtained from a lower granularity level into an overall result. In this ...

The problem of quality estimation of crowdsourced work is of great importance. Although a variety of aggregation methods have been proposed to find high-quality structured claims in multiple-choice crowdsourcing tasks such as item labeling, they do not apply to more general tasks, such as article writing and brand design with unstructured submissions. One possibility to tackle this problem is to ask another set of crowd workers to review and grade each submission, essentially transforming unstructured submissions into structured ratings. Nevertheless, such an approach incurs unnecessary monetary cost and delay. In this paper, we address this problem by exploiting task requesters’ historical feedback and directly modeling the submission quality. We propose two embedding-based methods where the first one learns worker embedding and the second one learns both worker embedding and meta information embedding, with additional consideration of neighborhood similarity. Experimental results on three large-scale crowdsourcing data sets demonstrate that our embedding-based feature-learning methods perform much better than feature-engineering methods that use popular learning-to-rank algorithms. At the same time, our methods do not require additional crowdsourced grading.

This paper examines the common practice of aggregating choice alternatives within discrete choice models. We carry out a Monte Carlo study based on realistic vehicle choice data for sample sizes ranging from 500–10,000 individuals. We consider methods for aggregation proposed by McFadden (1978) and Brownstone and Li (2017) as well as the more commonly used methods of choosing a representative disaggregate alternative or averaging the attributes across disaggregate alternatives. The results show that only the “broad choice” aggregation method proposed by Brownstone and Li provides unbiased parameter estimates and confidence bands. Finally, we apply these aggregation methods to study households’ choices of new 2008 model vehicles from the National Household Travel Survey (NHTS) where 1120 unique vehicles are aggregated into 235 make/model classes. Consistent with our Monte Carlo results we find large differences between the resulting estimates across different aggregation methods.

In this work we introduce attention as a state of the art mechanism for classification of radio galaxies using convolutional neural networks. We present an attention-based model that performs on par with previous classifiers while using more than 50% fewer parameters than the next smallest classic CNN application in this field. We demonstrate quantitatively how the selection of normalisation and aggregation methods used in attention-gating can affect the output of individual models, and show that the resulting attention maps can be used to interpret the classification choices made by the model. We observe that the salient regions identified by the our model align well with the regions an expert human classifier would attend to make equivalent classifications. We show that while the selection of normalisation and aggregation may only minimally affect the performance of individual models, it can significantly affect the interpretability of the respective attention maps and by selecting a model which aligns well with how astronomers classify radio sources by eye, a user can employ the model in a more effective manner., 18 pages, 16 figures, Published in MNRAS

Forecasts support decision making in a variety of applications. Statistical models can produce accurate forecasts given abundant training data, but when data is sparse or rapidly changing, statistical models may not be able to make accurate predictions. Expert judgmental forecasts—models that combine expert-generated predictions into a single forecast—can make predictions when training data is limited by relying on human intuition. Researchers have proposed a wide array of algorithms to combine expert predictions into a single forecast, but there is no consensus on an optimal aggregation model. This review surveyed recent literature on aggregating expert-elicited predictions. We gathered common terminology, aggregation methods, and forecasting performance metrics, and offer guidance to strengthen future work that is growing at an accelerated pace.

Countries that supply the majority of their energy needs from fossil fuels are turning to new, renewable, and eco-friendly energy sources. Wind energy technology has been the most promising alternative to traditional energy systems. The aim of this case study was the selection of the best wind turbine for one of the leading companies in Turkey. First, the studies on wind turbines in the literature were reviewed. The criteria used in these studies were listed. Interviews were conducted with the experts from the company and 21 criteria and the turbine alternatives were determined. Because there are multiple criteria, and some of them conflict with each other, it has been preferable to use multi-criteria decision making (MCDM) methods. The criteria were weighted by the SWARA method, which was not used in the previous turbine studies. The criterion weights of alternatives were given by experts using linguistic variable-defined neutrosophic numbers for the first time to select a wind turbine. TOPSIS and EDAS methods were used as the solution methods integrated with single valued neutrosophic numbers. The best wind turbine was selected with the assistance of the aggregation method of Borda. © 2020 Elsevier Ltd

Subjective Bayesian prior distributions elicited from experts can be aggregated together to form group priors. This paper compares aggregated priors formed by equal weight aggregation, the classical method and the Sheffield elicitation framework to each other and individual expert priors, using an expert elicitation carried out for a clinical trial. Aggregation methods and individual expert prior distributions are compared using proper scoring rules to compare the informativeness and calibration of the distributions. The three aggregation methods outperform the individual experts, and the Sheffield elicitation framework performs best among them.

Recently, interest in sentiment analysis has grown exponentially. Many studies have developed a wide variety of algorithms capable of classifying texts according to the sentiment conveyed in them. Such sentiment is usually expressed as positive, neutral or negative. However, neutral reviews are often ignored in many sentiment analysis problems because of their ambiguity and lack of information. In this paper, we propose to empower neutrality by characterizing the boundary between positive and negative reviews, with the goal of improving the model’s performance. We apply different sentiment analysis methods to different corpora extracting their sentiment and, hence, detecting neutral reviews by consensus to filter them, i.e., taking into account different models based on weighted aggregation. We finally compare classification performance on single and aggregated models. The results clearly show that aggregation methods outperform single models in most cases, which led us to conclude that neutrality is key for distinguishing between positive and negative and, then, for improving sentiment classification.

The growth of forests and the feedbacks between forests and environmental changes are central issues in the planetary carbon cycle, global climate change, and basic plant ecology. A challenge to understanding both growth and feedbacks from local to global scales is that many critical metabolic processes vary among species. An innovation in solving this challenge is the recognition that species can be lumped into “functional groups” based on metabolic similarity, and these functional groups can then be studied in computational models that simulate ecosystem function. Despite the vast resources devoted to functional group studies and the progress made by them, an important logical and biological question has not been formally addressed, “How do the groupings alter the results of modeling studies?” To what extent do modeling results depend on the choices made in aggregating taxa into functional groups. Here, we consider the effects of using different aggregation strategies in simulating the carbon dynamics of a deciduous forest. Understanding the impacts that aggregation strategy has on efforts to simulate regional-to-global-scale forest dynamics offers insights into both ecosystem regulation and model function and addresses this central problem in the study of carbon dynamics.

Ranking is a common task for selecting and evaluating alternatives. In the past few decades, combining rankings results from various sources into a consensus ranking has become an increasingly active research topic. In this study, we focus on the evaluation of rank aggregation methods. We first develop an experimental data generation method, which can provide ground truth ranking for alternatives based on their “inherent ability.” This experimental data generation method can generate the required individual synthetic rankings with adjustable accuracy and length. We propose characterizing the effectiveness of rank aggregation methods by calculating the Kendall tau distance between the aggregated ranking and the ground truth ranking. We then compare four classical rank aggregation methods and present some useful findings on the relative performances of the four methods. The results reveal that both the accuracy and length of individual rankings have a remarkable effect on the comparison results between rank aggregation methods. Our methods and results may be helpful to both researchers and decision-makers.

In this paper, we present logic-based aggregation models used for ranking student applicants and we compare them with a number of existing aggregation methods, each more complex than the previous one. The proposed models aim to include depen- dencies in the data using Logical aggregation (LA). LA is a aggregation method based on interpolative Boolean algebra (IBA), a consistent multi-valued realization of Boolean algebra. This technique is used for a Boolean consistent aggregation of attributes that are logically dependent. The comparison is performed in the case of student applicants for master programs at the University of Belgrade. We have shown that LA has some advantages over other presented aggregation methods. The software realization of all applied aggregation methods is also provided. This paper may be of interest not only for student ranking, but also for similar problems of ranking people e.g. employees, team members, etc.

HighlightsWe proposed a multi source product reputation model which offers several benefits over single source based approach.A method which uses four factors to rank the reputability of a source is introduced.Robust and strategy proof aggregation methods are introduced which provide a good approximation of the true reputation value even if some ratings are false or biased.Decay principle method is proposed in order to reflect the recent opinions about the product, which are more suitable in product reputation context.A balance between sensitivity and robustness is maintained, which makes the model resilient to malicious users but still considers the recent opinions. Product reputation model is very important for customers and manufacturers in order to make decisions. Several product reputation models are proposed in literature which use customer reviews in order to compute reputation values. However, the aggregation methods used are not able to estimate a good reputation value when some ratings are false. Some of these aggregation methods are not robust to false and biased ratings because a single false rating is enough to change the result. Others are robust to false ratings but not able to reflect the recent opinions about product quickly. In addition, most of the product reputation models are based on single source, therefore suffer from availability and vulnerability issues. In this paper, we propose a multi-source product reputation model where robust and strategy proof aggregation methods are used. A source credibility measure method is proposed, which uses four factors to determine malicious sources. Furthermore, a suitable decay principle for product reputation is also introduced in order to reflect the newest opinions quickly. The results show that proposed model is robust, strategy proof and able to provide a good estimation even if some ratings are false.

Planning models are widely used to provide insight in the transition of energy systems. With the increasing levels of intermittent renewables, the temporal representation in such models requires careful attention. However, representing all 8760 h in each modeled year while at the same time integrating detailed technical constraints of power plants might lead to an overall computationally intractable problem. In this context, several time aggregation methods to select a number of typical periods that together form an adequate representation of the full year have recently been developed.

Substantial progress has been made in assessing marine health in an integrative way. However, managers are still reluctant in undertaking such assessments, because: (i) lack of indicators; (ii) absence of targets; (iii) difficulty of aggregating indicators from different ecosystem components, habitats, and areas; (iv) absence of criteria on the number of indicators to be used; (v) discussion on the use of “one-out, all-out” (OOAO) principle in aggregating; and (vi) lack of traceability when integrating data. Our objective was, using open access databases with indicators across all the European seas with agreed targets, to demonstrate if the Nested Environmental status Assessment Tool (NEAT), can be used at the European scale, serving to managers and policy-makers as a tool to assess the environmental status under the Marine Strategy Framework Directive (MSFD). We have used MSFD Descriptor D3 (commercial fish) from 341 stocks, 119 species and two indicators from each of them (years 2013–2015); D5 (eutrophication) with 90th percentile of Chlorophyll-a (years 2009–2014); and D8 (contaminants), with Anthracene, Fluoranthene, Naphthalene, Cadmium, Nickel, and Lead as indicators (years 2009–2013). We have calculated the environmental status for each European subdivision, subregion, and regional sea, nested at different levels. The analyses include weighting and no-weighting by each assessment area; for ecosystem component (water column, phytoplankton, fish, crustaceans, and molluscs); descriptor (three), and habitat (pelagic, demersal/benthic), with the confidence value of the status. A sensitivity analysis was undertaken to determine a minimum number of indicators to include for a robust assessment. We demonstrated that using NEAT in assessing the status of large marine areas, by aggregating indicators, ecosystem components and descriptors, at different spatial scales, can remove at least four out of the six barriers that managers and policy-makers confront when undertaking such assessments. This can be done by using open-access databases and already established targets. Aggregating indicators of different origin is possible. Around 40 indicators seem to be enough to obtain robust assessments. It is better to integrate the assessment items using an ecosystem-based approach, rather than using the OOAO principle. Using NEAT, this approach supports identifying the problematic environmental issues needing management attention and measures.

Archaeozoologists commonly use Number of Identified SPecimens (NISP) and Minimum Number of Elements (MNE) as measures of anatomical abundances. According to a blind test examining the reproducibility and accuracy of identifications of ungulate remains (Morin et al., Part I, Journal of Archaeological Method and Theory, doi: 10.1007/s10816-016-9300-4), NISP provides estimates of skeletal abundances that are less robust than those based on MNE. However, although results were improved with the latter method, MNE is not free of problems. Here, we show through an analysis of paired NISP-MNE data for 24 classes of elements that MNE is prone to inflate the representation of rare parts (as measured by NISP), a phenomenon more strongly expressed in certain elements than in others. Moreover, some elements show a wide scatter of points, which raises issues of data reproducibility. MNE is also known for being seriously affected by aggregation methods. These fundamental problems severely undermine the value of MNE as a measure of abundance. This article introduces an alternative counting method that avoids many of the weaknesses of MNE. This counting method, called the Number of Distinct Elements (NDE), focuses on the occurrence of pre-determined, invariant landmarks counted on mutually exclusive specimens. Preliminary experimental results suggest that NDE counts are robust predictors of skeletal, and perhaps taxonomic, abundances. Moreover, the NDE approach eliminates the complex and time-consuming task of spreading or drawing specimens to identify fragment overlap. Furthermore, NDE values are additive and easy to calculate. Given these features, the NDE approach represents a compelling alternative to MNE in archaeozoological analysis.

In this paper we review literature and suggest a framework to examine and measure urban water security as it interacts with urbanization and urban-regional systems. We develop a comprehensive framework to start bridging that gap. In this framework, urban water security is shaped by five interacting social and environmental domains. These are Socio-demographic, Economic, Technological, Ecological, and Governance (SETEG). We suggest a few indicators and aggregation methods that can shed light on the multidimensional and interconnected nature of urban water security, and illuminate different levels of influence among the five SETEG domains. By improving the selection of indicators for the multiple SETEG domains and interactions creating urban water security (or insecurity), combined approaches such as the one outlined above might help move a scattered array of water security goals towards the creation of informed, cohesive and relevant policy interventions.

One of the crucial problems of the classifier ensemble is the so-called combination rule which is responsible for establishing a single decision from the pool of predictors. The final decision is made on the basis of the outputs of individual classifiers. At the same time, some of the individuals do not contribute much to the collective decision and may be discarded. This paper discusses how to design an effective combination rule, based on support functions returned by individual classifiers. We express our interest in aggregation methods which do not require training, because in many real-life problems we do not have an abundance of training objects or we are working under time constraints. Additionally, we show how to use proposed operators for simultaneous classifier combination and ensemble pruning. Our proposed schemes have embedded classifier selection step, which is based on weight thresholding. The experimental analysis carried out on the set of benchmark datasets and backed up with a statistical analysis, proved the usefulness of the proposed method, especially when the number of class labels is high.

This paper presents an accurate real-time stereo matching method, which is based on the extended binary weighted aggregation. The accuracy of the proposed stereo matching method was significantly enhanced by extending its binary weighted aggregation so that remote connections of support regions can be allowed for aggregation. The extended binary weighted aggregation is based on the following two new ideas. First, the extended binary weighted aggregation connects distant regions over color boundaries, making them one large support region for a given pixel. This approach induces more aggregation targets, and, thus, makes the aggregation step more robust. Second, it excludes cost outliers in the support region to prevent them from being propagated during the aggregation step, making a quality support region. With the extended binary weighted aggregation, the proposed stereo matching method obtains more accurate disparity maps than existing stereo matching methods using binary weighted aggregation methods, while maintaining the speed advantage of binary weighted aggregation. Experimental results illustrated that the proposed stereo matching method outperformed all existing real-time stereo matching methods in terms of accuracy, providing the average bad pixel rate of 5.12%, for the Middlebury stereo test images. The proposed stereo matching method was implemented on a CUDA platform with a high-end GPU. The implemented system operated at up to 300 fps for the stereo images with 320 × 240 pixel resolution and a disparity range of 32 pixels.

This paper extends the previous virtual synchronous generator (VSG) control for voltage-source converters to modular multilevel converters (MMCs). Compared with the ac synchronous generator (SG), the VSG-based MMC is embedded with a power-angle characteristic. Starting from this feature and with reference to the existing ac system coherency-based equivalence methods, this paper proposes a novel coherency equivalence method for MMC used in large-scale renewable energy systems connected to the ac grid. This novel method mainly focuses on the coherency criterion and parameter aggregation methods which have significant differences compared with the ac system methods. The proposed equivalence method is validated on PSCAD/EMTDC and the study results indicate that: 1) the VSG-based MMC model does show the power-angle characteristic similar to the SG model; 2) the proposed coherency criterion can accurately and efficiently classify the paralleled MMC into coherent groups; and 3) the dynamic behaviors of the external systems are preserved after the aggregations of all previous coherency MMC groups.

Owing to the irreparable damage caused by conventional agriculture and its industry, a shift to sustainable agriculture is a more strategic and essential need now than ever before. To boost the transition, this study presents a new and comprehensive tool for assessing agricultural sustainability of paddy fields in Guilan Province, Iran. Based on relevant literature review, an investigation was made into the sustainability indicators which had theoretically been adopted by researchers in recent years. Then a set of indicators was introduced for examination and prioritization. The proposed structural model included three factors (social, economic, and environmental) and 62 indicators. All of the 62 indicators were developed into a valid and reliable questionnaire, and agricultural experts and researchers familiar with agricultural sustainability concepts and the situation of Guilan’s paddy fields were surveyed about the feasibility of these factors as agricultural sustainability indicators. After th...

Machine learning has been a thriving topic in recent years, with many practical applications and active research aspects. In machine learning, model aggregation is an important area. The idea of model aggregation is to aggregate a global model from trained local models. However, traditional aggregation methods based on parameter averaging can not aggregate models which have different types and structures. Because parameter averaging will fail to average different types of values (parameters). To address this problem, we propose a new aggregation method which will suit for different types of local models. To achieve our goal, we transfer knowledge from local models to the global model. To do so, firstly, we propose differentially private GANs, let local parties generate synthetic data related to their training data. Secondly, we use the majority of prediction votes from local models to label those synthetic samples. Finally, use the labelled synthetic datato train the global model. By combining synthetic data and labels from local models, knowledge can be transferred from local models to the global model. We evaluate our scheme on Adult, MNIST and Fashion MNIST datasets under different settings, experimental results show that our scheme can achieve an accurate global model with low privacy loss. Besides, the easily implemented building blocks make our scheme efficient and practical for applications.

Fuzzy-based self-organizing aggregation method was suggested in the present study for swarm robots. In the suggested method, Swarm robots evaluate their limited sensor input via rules of fuzzy logic and display aggregation behavior with the suggested aggregation method. In the meantime, swarm robots also have the ability to escape the obstacles in a bounded arena with this method. Swarm robots perceive the neighboring robots with this method, make individual decisions and display aggregation behaviors. Different from the traditional self-organizing aggregation methods, the suggested approach utilizes fuzzy logic controllers to evaluate limited sensor data. Systematic experiments were applied on different number of swarm robots with different detection areas in arenas of different sizes. Moreover, noise was applied on the sensor inputs for examining the performance of the fuzzy logic based self-organizing aggregation method. The scalability and flexibility of the self-organizing aggregation behaviors of swarm robots were evaluated by way of systematic experiments. The swarm robots displayed aggregation behavior during the systematic experiments applied despite the changes in the number of robots and detection distances.

Buildings have both high as well as flexible energy demands and play an important role in the energy internet solution. The buildings' energy flexibility (BEF) is a widely recognized concept; however, how to unlock its potential is a relatively new research topic. In this paper, the authors provide an overview of the latest research related to BEF. An introduction to BEF is provided, methods developed for identifying and characterizing BEF are presented, and several key influencing factors are identified. The overview also covers various aggregation methods to scale up BEF impacts and service-oriented solutions for enabling BEF applications in different energy sectors. This work lays the groundwork for designing and developing seamless integration strategies for BEF use in both present and future energy systems.

The quality of crowdsourced work varies drastically from superior to inferior. As a consequence, the problem of automatically finding high-quality crowdsourced work is of great importance. A variety of aggregation methods have been proposed for multiple-choice tasks such as item labeling with structured claims. However, they do not apply to more general tasks, such as article writing and brand design, with unstructured submissions that cannot be aggregated. Recent work tackles this problem by asking another set of crowd workers to review and grade each submission, essentially transforming unstructured submissions into structured ratings that can be aggregated. Nevertheless, such an approach incurs unnecessary monetary cost and delay. In this paper, we address this problem by exploiting task requesters’ historical feedback and directly modeling the submission quality, without the need of additional crowdsourced ratings. We first propose three sets of features, which try to characterize the submission quality from various perspectives, including the submissions themselves, the workers who make the submissions, and the interactions between task requesters and workers. We then propose two quality models, where one judges the submission quality independently and the other judges comparatively. These models not only incorporate features, but also take worker-specific factors into consideration. Experimental results on three large-scale data sets demonstrate that our models outperform general-purpose learning-to-rank methods such as Logistic Regression, RankBoost, and ListNet for finding high-quality crowdsourced submissions.

Composite indicators (CIs) are a useful communication tool for conveying summary performance information in a relatively simple way. They are used widely in various sectors in public services. In Italy, they are also employed to allocate a quota of public funding to the universities. Despite increasing use, CIs value and benefit are still under debate and their suitability for the benchmark of universities is challenged by methodological issues and the multi-dimensional facet of higher education institutions. The issue addressed in the present work is whether the composite measure employed by the Italian Government is an accurate tool for evaluating, ranking and funding universities. For this purpose—after a revision of the indicators, weighting and aggregation methods put forward at the national level—CIs have been computed for the overall university performance, employing the indicators chosen at the national level and combining various weighting and normalisation schemes. The variation in the rankings assigned by the CIs to the given universities has been then explored to gauge its robustness. In the light of the results, some technical and analytical issues involved in the design of CIs and their policy implications are also discussed.

Search result diversification is one of the key techniques to cope with the ambiguous and underspecified information needs of web users. In the last few years, strategies that are based on the explicit knowledge of query aspects emerged as highly effective ways of diversifying search results. Our contributions in this article are two-fold. First, we extensively evaluate the performance of a state-of-the-art explicit diversification strategy and pin-point its potential weaknesses. We propose basic yet novel optimizations to remedy these weaknesses and boost the performance of this algorithm. As a second contribution, inspired by the success of the current diversification strategies that exploit the relevance of the candidate documents to individual query aspects, we cast the diversification problem into the problem of ranking aggregation. To this end, we propose to materialize the re-rankings of the candidate documents for each query aspect and then merge these rankings by adapting the score(-based) and rank(-based) aggregation methods. Our extensive experimental evaluations show that certain ranking aggregation methods are superior to existing explicit diversification strategies in terms of diversification effectiveness. Furthermore, these ranking aggregation methods have lower computational complexity than the state-of-the-art diversification strategies.

In this paper we aim to investigate different deep learning techniques for automatic extraction of valuable information from large sized satellite image data. We focus on the problem of semantic segmentation which attaches a class label to each pixel from the image. We investigate two semantic segmentation architectures based an convolutional neural networks: segnet and u-net. We analyse different tiling strides with reverse aggregation methods. We compare two classical methods (averaging and maximum) and propose a new method based on entropy. We test the models with distinct types of images, emphasizing the need to predict the results using information from all of them. We discuss various fusion strategies and introduce a fusion strategy based on the observations obtained from separately analysing the distinct image types.

In order to comply with climate goals in the future, all relevant sectors must implement extensive decarbonisation measures. For power system models, this implies that more complex interactions of bi-and multivalent technologies in the heat, industry and transport sector have to be included. Coupled with higher shares of renewable generation, these interactions lead to significant modelling challenges for large-scale power system models. Focusing on the heat sector, this study develops two aggregation methods for representing hybrid boilers, monovalent heat pumps and CHP systems in an adequate but computationally efficient way. The performance of the aggregation methods is evaluated and compared to the disaggregated reference case in terms of result accuracy and solution time for two load scenarios. It has been shown that the proposed aggregation method 2 which is based on the heating technology efficiencies yields a sound approximation of the disaggregated reference.

The last decade have witnessed a profusion of research work on the crowdsourcing topic. Human skills are essential in achieving high quality answers in crowdsourcing solving tasks. The current paper aims to introduce an innovative crowdsourcing-based solution for a scientific meta-journal. An overall architecture of the proposed system is introduced with a focus on the aggregation of the reviewers’ evaluations to produce a final decision. We introduce several aggregation methods adapted to the nature of data to fusion and discuss them. In addition, we discuss future challenges that cope with the proposed system.

International audience; This work falls within the context of predicting the value of a real function at some input locations given a limited number of observations of this function. The Kriging interpolation technique (or Gaussian process regression) is often considered to tackle such a problem, but the method suffers from its computational burden when the number of observation points is large. We introduce in this article nested Kriging predictors which are constructed by aggregating sub-models based on subsets of observation points. This approach is proven to have better theoretical properties than other aggregation methods that can be found in the literature. Contrarily to some other methods it can be shown that the proposed aggregation method is consistent. Finally, the practical interest of the proposed method is illustrated on simulated datasets and on an industrial test case with (Formula presented.) observations in a 6-dimensional space.

Spatially distributed regions may have different influences that affect the underlying physical processes and make it inappropriate to directly relocate learned models. We may also be aiming to detect rare events for which we have examples in some regions, but not others. Three novel voting methods are presented for combining classifiers trained on regions with available examples for predicting rare events in new regions; specifically the closure of shellfish farms. The ensemble methods introduced are consistently more accurate at predicting closures. Approximately 63% of locations were successfully learned with Class Balance aggregation compared with 37% for the Expert guidelines, and 0% for One Class Classification.

In this study, we develop a spatial decision support system (SDSS) for water management in a basin characterised by existing and projected dams to satisfy freshwater demand while preserving a particular ecosystem. The territorial and functional interests along with management scenarios were integrated into consistent stages of the SDSS. The developed prototype allowed comparisons of multiple water allocation options to competing users in the catchment by two aggregation methods. Through a simulation modelling exercise and stakeholder's involvement, the main outputs are the development of water management options and a set of criteria/subcriteria to evaluate these options related to socio-economic, water availability and ecological factors. Both aggregation methods reveal the positive effect of water transfer on overall evaluations. New dam construction would cause an increase in the overall evaluation from the SDSS by at best 34% when water availability criteria are favoured, while a decrease in overall evaluation by at worst 75% is indicated when ecological criteria are favoured.

The task of classifying videos of natural dynamic scenes into appropriate classes has gained a lot of attention in recent years. The problem especially becomes challenging when the camera used to capture the video is dynamic. In this paper, we analyse the performance of statistical aggregation (SA) techniques on various pre-trained convolutional neural network(CNN) models to address this problem. The proposed approach works by extracting CNN activation features for a number of frames in a video and then uses an aggregation scheme in order to obtain a robust feature descriptor for the video. We show through results that the proposed approach performs better than the-state-of-the art works for the Maryland and YUPenn dataset. The final descriptor obtained is powerful enough to distinguish among dynamic scenes and is even capable of addressing the scenario where the camera motion is dominant and the scene dynamics are complex. Further, this paper shows an extensive study on the performance of various aggregation methods and their combinations. We compare the proposed approach with other dynamic scene classification algorithms on two publicly available datasets — Maryland and YUPenn to demonstrate the superior performance of the proposed approach.

Different linguistic aggregation methods have been proposed and applied in the linguistic decision making problems. Generally, weights for experts or criteria are considered in linguistic aggregation processes. In this paper, we provide a method to discovery new forms to compute weights and new interpretations in the linguistic ordered weighted averaging operator. In linguistic decision analysis, it can be noticed that some of initial linguistic values used by experts have priority over others linguistic values in evaluation processes. We formalize the priority over initial linguistic values as weights for linguistic values, by considering weights for linguistic values as well as weights for experts, we provide an alternative method to discovery weights information of the linguistic ordered weighted averaging operator, its properties show that such linguistic aggregation operator is extensions of the 2-tuple arithmetic mean, the 2-tuple weighted aggregation operator and the 2-tuple ordered weighted averaging operator. By an illustrative example, we compare the linguistic aggregation operator with the 2-tuple weighted aggregation operator and the 2-tuple ordered weighted averaging operator in a decision making problem. From the practical point of view, we provide an optimization model to obtain such weights information in linguistic aggregation processes, examples show the linguistic aggregation operator as an alternative linguistic ordered weighted averaging operator in practice.

Every secure system has the possibility to fail. Therefore, extra effort should be taken to protect these systems. Intrusion Detection Systems (IDSs) had been proposed with the aim of providing extra protection to security systems. These systems trigger thousands of alerts per day, which prompt security analysts to verify each alert for relevance and severity based on an aggregation criterion. Several aggregation methods have been proposed to collect these alerts. This paper presents our threshold aggregation system (TAS). Results shows that TAS aggregates IDS alerts accurately based on user demands and threshold value.

Soft set theory was originally proposed by Molodtsov as a general mathematical tool for dealing with uncertainty in 1999. Recently, researches of decision making based on soft sets have got some progress, but few people consider multi-experts situation. As such, this paper discusses multi-experts group decision making problems. Firstly, we give a concept of intuitionistic fuzzy soft matrix (IFSM) and prove some relevant properties of IFSM. Then, an adjustable approach is presented by means of median level soft set and p-quantile level soft set for dealing with decision making problems based on IFSM. Thirdly, we study aggregation methods of IFSM, give two kinds of aggregation operators and methods that how to determine experts’ weights under different situation with programming models, four corresponding algorithms have been proposed, too. Finally, a practical example has been demonstrated the reasonability and efficiency of these new algorithms.

The growing interests in optimizing the generation, distribution, and delivery of electric power have motivated the implementation of several smart grid functions in many power systems around the globe. Among such functions are the peak-load management, demand response, direct load control, and integration of distributed power generation. Nowadays, smart grid functions are being implemented for industrial, residential, and/or commercial loads. One of the key requirements for implementing smart grid functions is the accurate and reliable load aggregation. The bottom-up, coordinated, and bus-split aggregation methods have been found applicable for different load types that are included in smart grid functions. Such a requirement has made aggregation methods main components in power system operation, as well as smart grid functions. This paper overviews the methods and approaches for performing the load aggregation, and provides a discussion for the critical role of load aggregation in power systems operating and smart grid functions. In addition, this paper discusses the applicability of the load aggregation methods in smart grid functions for residential loads.

Wireless sensor networks commonly implemented in such areas like volcano, fire observing, remote sensing and in army etc. Communication cost as well as energy utilization is minimized by implementing strategy called as data aggregation in wireless sensor networks. At the time of data routing, partial outcomes are consolidated by the data aggregation method from lower layer nodes. Related work illustrates number of researchers are proposed different aggregation methods such as synopsis diffusion, predicate count and sum technique. Issue is that these aggregation techniques don't have capability to recognize fake sub aggregate of compromise nodes. Problem is that this can causes huge errors at base station. Logical OR operation is performed by this system for aggregation of the digest at mediatory node. This system causes more risk at the time of information modification. To solve this problem, our system implements appending method to decrease the risk in data aggregation. In our proposed system, base station has capability to recognize the optional path through eliminating the attacker by using collected MAC. Experimental outcomes are demonstrates the proposed system execution is best as compare with the previous system.

Recommender systems are methods built to actively suggest personalized items to users based on their explicit declared preferences (ratings of movies in Netflix), or implicitly observed actions (purchase history). Although a great number of recommendation methods have been previously proposed in the literature, in many problems these methods present a high degree of disagreement in their recommendations. In this scenario, rank aggregation methods are an interesting solution. They can help finding a consensus on which items should be recommended to the user by taking into account the opinion of all available methods. In this direction, this paper proposes ERA (Evolutionary Rank Aggregation), a genetic programming method that outputs an aggregated ranking function built from information extracted from individual input rankings. ERA was tested in four large scale datasets, and obtained better results than other rank aggregation methods in three datasets, improving the results of mean average ranking precision in up to 9.5%.

Recently many works using deep convolutional neural networks for image classification and retrieval have suggested that activations of the network layers (CNN activations) are able to serve as good descriptors. In this paper, we investigate two popular activations: fully connected and convolutional activations, and propose a simple but effective method for aggregating those local activations to compact global descriptors (MPFC-SPC). Firstly, we use max pooling of fully connected activations (MPFC) and sum pooling of convolutional activations (SPC) as aggregation methods. Then we blend MPFC and SPC descriptors according to the complementary properties between these two activations. Our method achieves competitive results compared with other related state-of-the-art CNN descriptors and almost doesn't increase time overhead. In addition, we perform multi-scale vector of locally aggregated descriptors (VLAD) on MPFC and SPC. Further improved performance with ensemble of multiple levels' descriptors verifies the effectiveness of our method.

Discovering correspondences between schema elements is a crucial task for data integration. Most schema matching tools are semi-automatic, e.g., an expert must tune certain parameters thresholds, weights, etc.. They mainly use aggregation methods to combine similarity measures. The tuning of a matcher, especially for its aggregation function, has a strong impact on the matching quality of the resulting correspondences, and makes it difficult to integrate a new similarity measure or to match specific domain schemas. In this paper, we present YAM Yet Another Matcher, a matcher factory which enables the generation of a dedicated schema matcher for a given schema matching scenario. For this purpose we have formulated the schema matching task as a classification problem. Based on this machine learning framework, YAM automatically selects and tunes the best method to combine similarity measures e.g., a decision tree, an aggregation function. In addition, we describe how user inputs, such as a preference between recall or precision, can be closely integrated during the generation of the dedicated matcher. Many experiments run against matchers generated by YAM and traditional matching tools confirm the benefits of a matcher factory and the significant impact of user preferences.

Subjective judgment with human rating has been an important way of constructing ground truth for the evaluation in the research areas including information retrieval. Researchers aggregate the ratings of an instance into a single score by statistical measures or label aggregation methods to evaluate the proposed approaches and baselines. However, the rating distributions of instances are diverse even if the aggregated scores are same. We define a term of confusability which represents how confusable the reviewers are on the instances. We find that confusability has prominent influence on the evaluation results with a exploration study. We thus propose a novel evaluation solution with several effective confusability measures and confusability aware evaluation methods. They can be used as a supplementary to existing rating aggregation methods and evaluation methods.