Manish Aggarwal

Bio: Manish Aggarwal is an academic researcher from Indian Institute of Technology, Jodhpur. The author has contributed to research in topics: Binary entropy function & Fuzzy logic. The author has an hindex of 2, co-authored 5 publications receiving 11 citations.

Redefining fuzzy entropy with a general framework

Abstract: A general entropy framework is proposed, which could lead to intuitive, interpretable and comparable entropy functions in both probabilistic or fuzzy domain, alike. Based on the proposed general entropy framework, the existing fuzzy entropy functions are redefined. The proposed entropy functions are also extended to the probabilistic-fuzzy domain. The usefulness of the work is shown in a real world case-study.

Attitude-based entropy function and applications in decision-making

Abstract: The popular entropy functions are rigorously analysed in the context of uncertainty in the real world decision making. Based on the findings, a new entropy function is introduced specifically for the human decision making. The proposed function considers an agent’s degree of sensitivity towards uncertainty, i.e., the tendency to exaggerate or downplay the inherent uncertainty. The proposed entropy function is equipped to deal with both the subjective and probabilistic uncertainties alike, which are often interlinked in a decision-making context. The properties of the proposed entropy function are rigorously studied. A real case-study in portfolio diversification highlights the usefulness of the entropy function. It was found that the attitude plays a profound role, when there are a large number of uncertain systems (portfolios) to compare and choose from, or when the portfolios are more diversified.

Representing uncertainty about fuzzy membership grade

Abstract: A novel uncertainty representation framework is introduced based on the inter-linkage between the inherent fuzziness and the agent’s confusion in its representation. The measure of fuzziness and this confusion is considered to be directly related to the lack of distinction between membership and non-membership grades. We term the proposed structure as confidence fuzzy set (CFS). It is further generalized as generalized CFS, quasi CFS and interval-valued CFS to take into consideration the DM’s individualistic bias in the representation of the underlying fuzziness. The operations on CFSs are investigated. The usefulness of CFS in multi-criteria decision making is discussed, and a real application in supplier selection is included.

An integrated and discriminative approach for group decision-making with probabilistic linguistic information

Abstract: Group decision-making (GDM) is a complex process. The diversity, discrimination, and inevitable uncertainty due to human intervention characterize such problems that add to this complexity. To circumvent this challenge, there is an urge for an appropriate knowledge representation and decision-making approaches. The present paper is concerned with a prescriptive approach to GDM that can aid a group of decision-makers (DMs) to arrive at a decision. To this end, the recent concept of probabilistic linguistic term set is utilized. The discrimination among the alternatives, as in the real world, are mimicked using an integrated framework that adopts CRITIC and variance methods for attribute weight calculation, Gini index for calculating the weights of DMs, Maclaurin symmetric mean for aggregating preferences, and weighted distance-based approximation for prioritization of alternatives. A real-world problem on electric bike selection illustrates the usefulness of the proposed work. Finally, comparative analysis with extant methods demonstrates the technical results, and it is inferred that the proposed work is (i) highly consistent (from Spearman correlation) and (ii) produces broad rank values (from standard deviation) that could be efficiently discriminated for rational decision-making and backup management during critical situations.

Human Decision Making through an Entropic Framework

Abstract: In this paper, new entropy functions are formulated based on an agent’s perceived uncertainty that inevitably affects the agent’s choice. The role of the decision-maker’s (DM’s) attitude is emphasized as one of the key determinants of such an entropy function. More specifically, new attitude-based variants of Shannon’s, Pal & Pal, and Aggarwal’s probabilistic entropies are introduced. The extant fuzzy entropies are also extended to consider the agent’s specific attitude. The proposed entropy functions provide a wide range of entropy values with the conventional entropy functions as their special cases. The special cases of the proposed entropies are examined. The wide applicability of the proposed entropy functions in multi criteria decision making is highlighted. A case-study is included to showcase the usefulness of the proposed entropy functions in the real world.

Short-term wind speed forecasting based on long short-term memory and improved BP neural network

Abstract: Accurate and reasonable wind speed prediction system has a significant impact on the utilization of wind energy. A novel combination forecasting model based on Long Short-Term Memory (LSTM) network and BP neural network is designed in this paper. This model combines the principle of deep learning algorithm and the improved BP neural network to deal with nonlinear wind speed prediction. Before the prediction, singular spectrum analysis (ssa) and complete ensemble empirical model decomposition adaptive noise (CEEMDAN) are selected as the data pretreatment part to de-noise the original wind speed data and decompose it into multiple components. This part is conducive to improving the signal-to-noise ratio (SNR) of wind speed data and simplifying the characteristics of wind speed data. Then, in order to reduce the error accumulation and computation redundancy, fuzzy entropy (FE) is used to calculate the time complexity of each component, according to the Spearman correlation, the inherent mode function (IMF) components are recombined to form a new subsequence. Experimental results show that the error accumulation can be reduced by 48.65% for dataset 1 and 29.53% for dataset 2, and the operation time can be reduced by about 50% for two datasets. To avoid the limitation of a single model, introducing the LSTM and improved BPNN which improved by sparrow search algorithm (SSA) two different prediction models are used to predict the sub sequences with high complexity and the low complexity subsequences, respectively. Finally, the predicted values of the models are superimposed to get the final values. In order to verify the validity of the proposed model, the final predictions, compared with six different prediction models, show that this model can achieve the best performance and obtain higher prediction accuracy. Such as the performance evaluation indexes (RMSE = 0.051, MAPE = 0.929%) are smallest obtained from dataset1 by one-step prediction, and (RMSE = 0.086, MAPE = 0.966%) are smallest obtained from dataset2 by one-step prediction. In addition, the Pearson correlation between the predicted value and the true wind speed value obtained by the prediction model applied to the two data sets is the highest 99.17% and 98.73%, respectively.

PT-TODIM Method for Probabilistic Linguistic MAGDM and Application to Industrial Control System Security Supplier Selection

Abstract: The close combination of internet technology and traditional industrial control system (ICS) is a double-edged sword, which not only improves the accuracy of the control system, but also brings great danger. According to the statistics of the authoritative industrial security incident information database, Repository of Industrial Security Incidents (RISI), as of October 2011, there have been more than 200 attacks on industrial control systems around the world. It is obviously a multi-attribute decision-making (MADM) problem to select the appropriate industrial control system security supplier (ICSSS) for ensuring the safety of ICS. In this paper, the traditional TODIM method is improved and reconstructed in the probabilistic linguistic environment by incorporating the prospect theory (PT) which has received close attention recently. In this new model, the entropy weight method is used to obtain the attribute weight under complete information. And based on the ideas of PT as well as traditional TODIM method, the distortion of decision results caused by the risk attitude and psychological state of decision makers is corrected as far as possible. In my opinion, probabilistic linguistic term set (PLTS) ensures that the model can better cope with the real environment and the complexity and ambiguity of decision makers’ thinking. Finally, the successful application of this new probabilistic linguistic TODIM method based on PT (PT-PL-TODIM method) in the selection of ICSSS proves that the model is practical and is also of great value to the decision-making research related to ICS. Moreover, the comparative analysis between the proposed model and the existing model effectively confirms the reliability of the proposed method. In the future, it is hoped that this method can be successfully applied to more decision-making fields.

Short-term wind speed forecasting based on long short-term memory and improved BP neural network

Abstract: Accurate and reasonable wind speed prediction system has a significant impact on the utilization of wind energy. A novel combination forecasting model based on Long Short-Term Memory (LSTM) network and BP neural network is designed in this paper. This model combines the principle of deep learning algorithm and the improved BP neural network to deal with nonlinear wind speed prediction. Before the prediction, singular spectrum analysis (ssa) and complete ensemble empirical model decomposition adaptive noise (CEEMDAN) are selected as the data pretreatment part to de-noise the original wind speed data and decompose it into multiple components. This part is conducive to improving the signal-to-noise ratio (SNR) of wind speed data and simplifying the characteristics of wind speed data. Then, in order to reduce the error accumulation and computation redundancy, fuzzy entropy (FE) is used to calculate the time complexity of each component, according to the Spearman correlation, the inherent mode function (IMF) components are recombined to form a new subsequence. Experimental results show that the error accumulation can be reduced by 48.65% for dataset 1 and 29.53% for dataset 2, and the operation time can be reduced by about 50% for two datasets. To avoid the limitation of a single model, introducing the LSTM and improved BPNN which improved by sparrow search algorithm (SSA) two different prediction models are used to predict the sub sequences with high complexity and the low complexity subsequences, respectively. Finally, the predicted values of the models are superimposed to get the final values. In order to verify the validity of the proposed model, the final predictions, compared with six different prediction models, show that this model can achieve the best performance and obtain higher prediction accuracy. Such as the performance evaluation indexes (RMSE = 0.051, MAPE = 0.929%) are smallest obtained from dataset1 by one-step prediction, and (RMSE = 0.086, MAPE = 0.966%) are smallest obtained from dataset2 by one-step prediction. In addition, the Pearson correlation between the predicted value and the true wind speed value obtained by the prediction model applied to the two data sets is the highest 99.17% and 98.73%, respectively.

Multi-criteria food waste treatment method selection using single-valued neutrosophic-CRITIC-MULTIMOORA framework

Abstract: Proper management and treatment of food waste have become a key concern due to its significant environmental, social, and economic ramifications. The selection of the most appropriate food waste treatment method among a set of alternative methods can be regarded as a multi-criteria decision-making problem because of the association of numerous qualitative and quantitative attributes. In this paper, we offer a novel integrated framework by combining criteria interaction through inter-criteria correlation (CRITIC) and multi-objective optimization based on ratio analysis with the full multiplicative form (MULTIMOORA) methods with single-valued neutrosophic sets (SVNSs) for assessing the multi-criteria food waste treatment methods selection. In this methodology, the CRITIC technique is applied for computing the attribute weights, and the MULTIMOORA model is employed for estimating the ranking of the options within SVNSs context. To examine the introduced methodology’s efficiency and achievability, a case study of food waste treatment method (FWTM) assessment is discussed in the SVNSs setting. Further, comparative study and sensitivity investigation are offered to certify the presented framework for prioritizing FWTMs. The final results indicate that the proposed approach achieves better solutions than the extant models.

Entropy-Based Weights on Decision Makers in Group Decision-Makingsetting with Hybrid Preference Representations

Abstract: Abstract The weights of decision makers play an important role in group decision-making problems. Entropy is a very important measure in information science. This work models an approach to determine the weights of decision makers by using an entropy measure. A new normalized projection as a separation measure, along with TOPSIS (technique for order preference by similarity to ideal solution) technique, is used for current decision model. The attribute values in current model are characterized by exact values and intervals. A comparison and experimental analysis show the applicability, feasibility, effectiveness and advantages of the proposed method.