sohrab kordrostami

In the real world, all available data are not definitive and are considered based on quality. Estimating the values of the inputs when we change the values of the outputs as desired is one of the important applications of inverse data envelopment analysis. If we want to estimate the level of inputs (outputs) among a group of decision-making units (DMUs), when some or all of its outputs (inputs) are changed so that cost efficiency is maintained or improved, inverse data envelopment analysis is used. In this article, cost efficiency is investigated by increasing desired outputs along with triangular fuzzy data. The problem of inverse data envelopment analysis with fuzzy data is presented for the cost efficiency of the DMU under evaluation. Also, in this connection, the results of the proposed model will be examined in a numerical example.

After assessing the efficiency of the units under evaluation, determining their ranking becomes a critical concern for decision-making. Ranking methodologies rely on specific unit inputs and outputs, but some variables affecting unit performance are not directly observable and are termed contextual variables. Thus, the primary objective of this research is to introduce a ranking methodology that incorporates these variables. To achieve this, a two-step approach is employed. Firstly, the Data Envelopment Analysis (DEA) technique calculates efficiency scores for each Decision-Making Unit (DMU), considering its unique inputs and outputs. Subsequently, the linear regression method is applied to assess the impact of contextual variables on efficiency. Finally, efficiency scores are modified by removing the impact of contextual variables, enabling unit ranking. To illustrate this proposed methodology, we conducted a performance evaluation of provincial gas companies in Iran. The evaluation covered four periods, from 2013 to 2016.

The traditional inverse data envelopment analysis (IDEA) models assess specific performance metrics in relation to changes in others, without taking into consideration the existence of random and undesirable outputs. This study presents a novel inverse DEA model with random and undesirable outputs, enabling the estimation of some random performance measures for changes of other random measures. The proposed chance-constrained inverse DEA model integrates both managerial and natural disposability constraints. By using the introduced approach, the estimation of natural disposable random inputs is presented for changes in random desirable outputs. Also, undesirable outputs are assessed for the perturbation of managerial disposable random inputs while the stochastic efficiency is maintained. The models are solved as  linear problems, with a numerical example provided to illustrate their application. The findings indicate that this approach is effective for evaluating efficiency and performance metrics in scenarios involving random and undesirable outputs.

Semantic Textual Similarity (STS) across languages is a pivotal challenge in natural language processing, with applications ranging from plagiarism detection to machine translation. Despite significant strides in STS, it remains a formidable task in languages with distinct syntactic structures and limited digital resources. Linguistic diversity, especially in word order variation, poses unique challenges, exemplified by languages adhering to Subject-Object-Verb (SOV) or Subject-Verb-Object (SVO) patterns, compounded by complexities like pronoun-dropping. This paper addresses the intricate task of measuring STS in Persian, characterized by SOV word order and distinctive linguistic features.

We propose a novel joint learning approach, harnessing an enhanced self-attention model, to tackle the STS challenge in both SOV and SVO language structures. Our methodology involves establishing a comprehensive multilingual corpus with parallel data for SOV and SVO languages, ensuring a diverse representation of linguistic structures. An improved self-attention model is introduced, featuring weighted relative positional encoding and enriched context representations infused with co-occurrence information through pointwise mutual information (PMI) factors. A joint learning framework leverages shared representations across languages, facilitating effective knowledge transfer and bridging the linguistic gap between SOV and SVO languages.

Our model, trained on Persian-English and Persian-Persian language pairs simultaneously, successfully extracts informative features, explicitly considering differences in word order and pronoun-dropping. During the training, the batch is sampled from STS benchmark with English and Translated Persian Pair texts and fed into customized encoder to get attention matrix and output embeddings. Then, the similarity module predicts the STS score. We use the STS score to compute the Mean Square Error (MSE) loss. Evaluation on Persian-English and Persian-Persian STS-Benchmarks demonstrates impressive performance, achieving Pearson correlation coefficients of 89.51% and 92.47%, respectively. Comparative experiments reveal superior performance against existing models, emphasizing the effectiveness of our proposed approach.

The ablation study further substantiates the robustness of our system, showcasing faster convergence and reduced susceptibility to overfitting. The results underscore the significance of our enhanced model in addressing the complexities of measuring semantic similarity in languages with diverse linguistic structures and limited digital resources. The approach not only advances cross-lingual STS capabilities but also provides insights into handling syntactic variations, such as SOV and SVO word orders, and pronoun-dropping. This research opens avenues for future investigations into enhancing STS in languages with unique structural characteristics.

A significant part of each system is determining system efficiency to conduct future planning-associated operations. Data Envelopment Analysis (DEA) is often employed to measure system efficiency. The paper considers a bi-level structure and proposes a new non-radial method by generalizing Russell's model to measure system efficiency. Furthermore, data from 33 branches of an Iranian state bank in 2021 are investigated to present an example of the application. The results indicate that among 33 branches, only two branches are regarded as efficient at both leader and follower levels.

Since the healthcare system is one of the most important pillars of community health, and considering that providing healthcare services to the people is one of the elements of individual development in any country, attention and supervision of this sector can lead to development and social welfare. To ensure better and higher quality healthcare services, performance evaluation in the health sector plays a crucial role. In order to achieve this, proper and proportional use of existing facilities and assets is inevitable. In this study, by introducing an application in the field of healthcare systems, the educational hospitals of the country have been measured in terms of performance and their managerial ability has been calculated. Additionally, by identifying and introducing the impact of contextual variables on the performance of decision-making units, their efficiency has been assessed. For this purpose, data related to educational hospitals in 31 provinces of the country was collected, and then by identifying contextual variables and with the presence of undesirable factors, the efficiency was evaluated and the managerial ability of each was calculated. To reach this goal, in the first step, technical efficiency with the presence of undesirable factors was calculated using data envelopment analysis technique, and then the logarithm of technical efficiency obtained from the first stage was regressed on a set of contextual variables that affect hospital performance. In the next stage, managerial ability was extracted from the residual of the regression obtained from the previous stage. Finally, a unique ranking based on the managerial ability of each unit was provided. Ultimately, the results obtained were analyzed and examined in order to provide valuable suggestions for managers and more efficient management of the country's hospitals to maintain public health. According to the study, without considering contextual variables, 25 effective units were evaluated, but by applying the effect of contextual variables on the efficiency index, no unit becomes effective, proving the high impact of such indices on the performance of units. Additionally, in the ranking of units based on managerial ability, Lorestan province ranked first and Golestan province ranked last.

Internet of Things describes a situation in which a large number of devices (things) are connected through a number of sensors via Internet, and lack of cooperation of some nodes in providing service to other nodes might interrupt the connection of some things, degrading network efficiency. A multi-phase mechanism based on Game theory and direct/indirect fame has been designed to motivate the selfish and malicious nodes to cooperate in IoT, which begins by deploying nodes in the IoT network. In the first phase, the nodes are grouped into clusters with cluster-heads for data collection. In the second phase, a multiplayer and dynamic game is executed while forwarding their data packet or others’ data packet. Nodes can pick their strategy when data packet forwarding in the third phase (Fuzzy logic reputation). Nodes will determine the neighboring node reputation by using fuzzy system. The amount of reputation of each of the nodes has been realized and finally, with the help of second phase and fuzzy logic, each node is decided to be cooperate or selfish nodes and in case of head clusters and fuzzy logic in some cases, the opportunity node will be reestablished to cooperate in network activities otherwise the node will be isolated. The effectiveness of the proposed solution has been assessed and the parameters of non-cooperative node detection accuracy, positive and negative warning rates, network PDR, and average endto-end latency perform better compared to other previous methods.

Due to the nonlinear and discrete nature of BCC (Banker, Charnes, and Cooper, [11]) models for determining the most efficient decision-making unit, it is practically impossible to evaluate the models' dual and, consequently, optimistic case. Thus, in this paper, the linear model with linear constraints proposed by Akhlaghi et al. [2] is used to investigate the dual equality of the model's robust problem and the optimistic case of the new model's dual under VRS uncertainty. The model proposed in this paper is novel in comparison to previous models because it solves the most efficient decision-making unit only once, without relying on uncertain data to determine its rank. The paper demonstrates how the proposed robust model can also ascertain the most efficient decision-making unit when uncertainty exists. Furthermore, the dual issues raised by robust counterparts in the new linear programming (LP) model are addressed to identify the most efficient decision-making unit. The robust counterpart is demonstrated to be equivalent to a linear program under interval uncertainty, and the dual of the robust counterpart is shown to be equal to the optimistic counterpart of the dual problem. Consequently, this study aims to demonstrate that the dual problem is equivalent to a decision-maker operating under optimal data, whereas the primal robust problem is equivalent to a decision-maker operating through the worst-case possible data scenario.

Semantic Textual Similarity (STS) is considered one of the subfields of natural language processing that has gained extensive research attention in recent years. Measuring the semantic similarity between words, phrases, paragraphs, and documents plays a significant role in natural language processing and computational linguistics. Semantic Textual Similarity finds applications in plagiarism detection, machine translation, information retrieval, and similar areas. STS aims to develop computational methods that can capture the nuanced degrees of resemblance in meaning between words, phrases, sentences, paragraphs, or even entire documents which is a challenging task for languages with low digital resources. This task becomes intricate in languages with pronoun-dropping and Subject-Object-Verb (SOV) word order specifications, such as Persian, due to their distinctive syntactic structures. One of the most important aspects of linguistic diversity lies in word order variation within languages. Some languages adhere to Subject-Object-Verb (SOV) word order, while others follow Subject-Verb-Object (SVO) patterns. These structural disparities, compounded by factors like pronoun-dropping, render the task of measuring cross-lingual STS in such languages exceptionally intricate. In the context of low-resource languages like Persian, this study proposes a customized model based on linguistic properties. Leveraging pronoun-dropping and SOV word order specifications of Persian, we introduce an innovative enhancement: a novel weighted relative positional encoding integrated into the self-attention mechanism. Moreover, we enrich context representations by infusing co-occurrence information through pointwise mutual information (PMI) factors. This paper introduces a cross-lingual model for semantic similarity analysis between Persian and English texts, utilizing parallel corpora. The experiments show that our proposed model achieves better performance than other models. Ablation study also shows that our system can converge faster and is less prone to overfitting. The proposed model is evaluated on Persian-English and Persian-Persian STS-Benchmarks and achieved 88.29% and 91.65% Pearson correlation coefficients on monolingual and cross-lingual STS-B, respectively.

Supply chains often have different technologies. Additionally, organizations with multiple stages can evaluate their operational efficiency by analyzing scale elasticity, which helps determine if they are functioning optimally or if there is room for improvement. This evaluation allows for the identification of potential inefficiencies and opportunities for enhancement. Consequently, this research introduces a two-stage DEA-based approach with undesirable outputs to examine the scale elasticity of supply chains within meta and group frontiers. The measurement of group and meta performance of general systems and stages is conducted for this purpose. Moreover, the study addresses the scale elasticity of supply chains with undesirable outputs by considering the heterogeneity of technologies. To achieve this, the study focuses on the right and left scale elasticity of efficient general systems and each stage. A real-world application from the soft drink industry is provided to illustrate the proposed model. The results show the applicability of the introduced methodology.

The electricity industry plays a pivotal role in a country's economic growth and development. Therefore, it is imperative to assess its performance and identify the strengths and weaknesses of its different sectors, such as production, transmission, and distribution, to enhance economic growth in diverse areas. Given the significance of the transmission sector, this research focuses on analyzing and evaluating the performance of 16 regional electricity companies in Iran from 1390 to 1398, with the aim of comprehending the impact of contextual variables on efficiency. To achieve this, the study will utilize two techniques - Data Envelopment Analysis (DEA) and Ordinary Least Squares (OLS) - to determine the efficiency score and estimate the effect of contextual variables on efficiency, respectively. In the first stage, the DEA technique is employed to calculate the technical efficiency of each company, considering their specific inputs and outputs. In the second stage, the logarithm of the efficiency scores obtained is regressed on contextual variables to establish their effect on efficiency. The residual derived from the regression is referred to as managerial ability. Finally, the companies are ranked based on their modified efficiency after removing the impact of contextual variables.

In this paper, a new method is presented using a combination of deep learning method, specifically recursive neural network, and Markov chain. The aim is to obtain more realistic results with lower cost in predicting COVID-19 patients. For this purpose, the BestFirst algorithm is used for the search section, and the Cfssubseteval algorithm is implemented for evaluating the features in the data preprocessing section. The proposed method is simulated using the real data of COVID-19 patients who were hospitalized in treatment centers of Tehran treatment management affiliated to the Social Security Organization of Iran in 2020. The obtained results were compared with three valid advanced methods. The results showed that the proposed method significantly reduces the amount of memory resource usage and CPU usage time compared to similar methods, and at the same time, the accuracy also increases significantly.

In traditional data envelopment analysis (DEA) models, the relative efficiency of decision making units (DMUs) is usually evaluated in a particular period of time such that the status of each measure from the input or output viewpoint is certain. However, in many applications, the performance of organizations should be measured over multiple periods of time while the status of some factors called “flexible measures” from the perspective of input or output is uncertain. The purpose of this study is to propose approaches to evaluate the efficiency of the multi-period systems where flexible measures are presented. For this reason, oriented and non-oriented DEA-based approaches from the standpoints of individual DMU and aggregate efficiencies are rendered to measure the overall and period efficiency of multi-period systems with flexible measures. Also, efficiency changes between two periods are estimated using the Malmquist productivity index (MPI). A dataset is provided to validate the proposed approach.

Cost efficiency in which cost coefficients are given for some inputs (cost coefficients can be different for disparate decision-making units (DMUs)) is one of the most important concepts in data envelopment analysis (DEA) to analyze the performance. Moreover, in some occasions, the cost performance and changes of input measures should be addressed while the convexity property is violated. Therefore, in this paper, first a DEA model is provided to assess cost efficiency based on the free disposal hull (FDH) model. Then, by considering cost and technical efficiencies achieved, a multi-objective problem called the inverse FDH cost model is presented to determine input values based on output changes while the cost and technical efficiency levels are preserved. The multi-objective problem is computed applying two approaches. Also, a dataset from the literature is presented to show the performance of the proposed method. For this purpose, we used the data of six banks in different countries. We added 2% to the outputs and analyzed the inputs with two models. In the first model, we used cost coefficients for weights, and in the second model, we used the same weights. Contrary to forecasts, some entries have decreased and others have increased. But from the results, we have noticed that the first model is more realistic because most of the solutions have increased in this model.

In the most investigations of sustainability, including environmental, social and economic issues, in addition to the desirable outputs, undesirable outputs are also presented, which is an obstacle to sustainable development. In this regard, the purpose of this paper is providing an approach based on Data Envelopment Analysis (DEA) with different forms of weak disposability of undesirable outputs to move towards sustainability.

Presenting a DEA-based model, the sustainability and performance of each dimension of sustainability are calculated simultaneously, while undesirable outputs are present with different forms of weak disposability. The sustainability performance of provincial gas companies is examined using the proposed technique.

The results show that the proposed method in the performance analysis of sustainability and its dimensions is efficient when undesirable outputs are presented.

Originality/Value:

 DEA provides a variety of disposability to minimize undesirable outputs and moves to optimize. In this study, an integrated approach with different forms of weak disposability is presented to analyze sustainability.

Estimating the revenue efficiency of entities under evaluation is one of the most important evaluations that can give valuable information about organizations provided that the output prices are known . In this research, a new definition of optimal scale size ( OSS ) based on maximizing the average revenue efficiency (ARE) is presented . Also , the ARE is defined in both convex and non - convex sets, which is independent of returns to scale and the assumption that the vector of input-output prices of units is uniform . Next, due to the presence of uncertain data in many real applications, the introduced ARE model is extended to evaluate systems with random inputs and outputs , and approaches are provided to calculate it . Finally , the proposed method is used in an experimental example and the ARE is calculated for a data set of postal areas in Iran .

Due to the changes of performance measures, a vital aspect for decision makers is finding optimal scale sizes of entities. Moreover, there are undesirable measures in many investigations. In the existing data envelopment analysis (DEA) approaches, optimal scale sizes (OSSs), average-cost efficiency (ACE) and average-revenue efficiency (ARE) of decision making units (DMUs) with desirable measures under strong disposability have been estimated while undesirable factors are presented in many real world examinations. Accordingly, in this research, OSSs and ARE of DMUs with undesirable outputs are addressed under managerial disposability. ARE is defined as the composite of scale and output allocative efficiencies under managerial disposability. To illustrate in detail, a two-stage DEA-based approach is rendered to estimate ARE and OSSs in the presence of undesirable outputs. A numerical example and an illustrative case are given to explain the proposed approach in this study.

Due to the complexity of the processes, analyzing the performance of the systems and examining the change in the scale of operations to improve productivity is a significant aspect. Therefore, in this research, a two-stage directional network data envelopment analysis approach is proposed to estimate the overall efficiency and the performance of each stage of two-stage processes, while undesirable outputs are produced in addition to desirable outputs. The proposed technology is based on the properties of the weak disposability of undesirable outputs and variable returns to scale. In addition, after presenting the response function, methods are proposed to calculate the right and left scale elasticities of overall efficient two-stage processes. In order to investigate the validity of the proposed method, a data set from 78 branches of the Iranian bank is used and their efficiency and scale elasticity are examined. The results show that the introduced method is useful and widely applicable to analyze the performance and scale of two-stage network systems in the presence of undesirable outputs.

Data envelopment analysis (DEA) is a nonparametric method for measuring the efficiency of decision-making units (DMUs) with multiple inputs and outputs. This research used the original DEA model and extended it to solve the DEA efficiency measurement problem, specifically for unseparated shared inputs. The consideration of this context aims to establish a new DEA approach to explore bank branch performance in different activities based on the optimal usage of unseparated shared inputs. In this study, in the first stage, the efficiency score is calculated from several activities using graph efficiency, and then, a maximum efficiency score pertaining to each DMU is applied to propose a new model. In the second stage, the efficiency score, which is calculated by the new approach on unseparated shared inputs, is defined as a new constraint based on shared inputs on the CCR model. This approach is implemented on the real data of 25 branches of a private bank in Iran. In fact, the efficiency of each branch is calculated, and enhancement guidelines are presented considering the three activities of production, electronic banking, and intermediation. Presenting one real efficiency score for each DMU, instead of the traditional efficiency score, leads to more robust decisions based on a more transparent performance evaluation in bank branches.

In data envelopment analysis (DEA) models, the weak disposability axiom has been used for undesirable outputs such as environmental pollutants in many cases. Considering this axiom, different models are exhibited to assess the efficiency of units under evaluation. In this paper, we express duality in weak disposability models proposed by Kousmanen (2005) and Amirteimoori et al. (2017). In the following, economical interpretations of dual are represented in the abovementioned technologies. Also, projection points for inefficient units will be introduced; then, proposed methods will also be explained in an application that includes different designs for the production line of a company.

Data envelopment analysis models are able to rank decision-making units (DMUs) based on their efficiency scores‎. ‎In spite of the fact that there exists a unique ranking of inefficient DMUs‎, ‎ranking efficient DMUs is problematic‎. ‎However‎, ‎rather than ranking methods‎, ‎another way to choose one of the efficient units is to determine the most efficient DMU‎. ‎Up to the present‎, ‎many models have been proposed to rank DMUs and determine the most efficient one‎. ‎These models require solving nonlinear or integer programs‎, ‎which are NP-hard and time-consuming‎. ‎Considering efficient DMU's characteristics‎, ‎this paper proposes a procedure to find the most efficient DMU through some simple operations‎. ‎The validity of the proposed approach is verified and tested via some numerical examples.

In recent decades, the topic of performance measurement has been one of the popular topics for large companies and manufacturers managers . The range - adjusted measurement ( RAM ) model in data envelopment analysis (DEA) is a non - radial model used to evaluate the performance of firms. Due to the presence of uncertain data in many investigations, a fully fuzzy range-adjusted measurement model with strong complementary slackness conditions to find efficient two-stage systems in a reference set is presented in this paper and it is used to evaluate airline . Given that we have a multi-objective network model of fully fuzzy range - adjusted measurement with strong complementary slackness conditions, the proposed model is solved using the lexicograph method . We also compare it with the existing fully fuzzy network DEA - range adjusted measurement model. Finally, we apply this model using the data of 14 Iranian airlines.

In many production processes, the efficiency of decision-making units should be investigated in the presence of undesirable inputs and outputs and non-discretionary factors. It is also important for managers to examine the impact of changes on one index on another. Therefore, the purpose of this paper is to develop an approach to estimate the efficiency of systems with undesirable inputs and outputs and non-discretionary measures and also to analyze changes.

First, a novel approach based on data envelopment analysis was presented in order to analyze efficiency in the presence of undesirable input-output indices and non-discretionary factors. In fact, the new weak disposability axiom was used and the efficiency of the systems was investigated in the presence of undesirable input-output index and non-discretionary factors. Then the impact of changes of one non-discretionary index on another index was measured by maintaining efficiency, and the marginal rates of these changes were calculated.

In order to explain the proposed method in practice, an applied example provided based on 31 branches of Islamic Azad University (department of education) was presented and the results were discussed and analyzed. By using the suggested method, the efficiency scores and the impact of modifications were obtained.

Performance scores related to branches indicated that most of them behaved well. Also, it was observed that always increasing (decreasing) one index, while maintaining efficiency, does not end in an increase (decrease) in another index but it may remain unchanged or decreased (increased). Due to the inclusion of undesirable input and output measures and non-discretionary factors, the results of the proposed method were more acceptable than the classical DEA models.

In resource allocation and target setting problems, a central planner decision making from a managerial point of view has a pivotal role, especially in presence of undesirable outputs such as greenhouse gas emissions. In these situations, firms have to incorporate to each other to achieve the goals of the central planner. The existing DEA-based resource allocation models have not considered the influence of managerial effort and technology innovation. In this study, we will use the managerial disposability assumption to reflect the central planner managerial achievement and technology novelty perspective in the process of resource allocation and target setting.

Using a managerial disposability assumption in this paper offers a solution to a correct and acceptable resource allocation and target setting along with improving the performance of units. To analyze the method presented in this paper, the data of 29 famous international airlines representing the global aviation industry have been selected and studied.

The results of this study show that in this model, decision-making units use managerial disposability assumption in the regulation of undesirable outputs based on the perspective of cooperation strategies to improve their environmental performance. In addition, in this approach increasing the inputs, fixing the amount of the desirable outputs, reducing the amount of undesirable outputs will be allowed. This model ensures that the adjusted decision-making units in the next period, will improve their efficiency after resource allocation and target setting, as well as improving the overall efficiency is observed in the results obtained by this method.

The paper presents a new approach of resource allocation and target setting based on data envelopment analysis which considers the impact of managerial effort and technology innovation on resource allocation and target setting problems.

Existing systems for diagnosing heart disease are time consuming, expensive, and prone to error. In this regard, a diagnostic algorithm has been proposed for the causes of heart disease based on a frequent pattern with the B-mine algorithm optimized by association rules. Initially, a data set of disease is used to select a feature, so that it deals with a set of training features. Then, association rules are used to classify educational and experimental sets, and then the factors affecting heart disease are analyzed. The numerical results from the experiments of real and standard datasets of cardiac patients show that the average accuracy of the proposed method is approximately 98%, which has been tested on the Cleveland database that includes 76 features in the case of heart disease dataset, 14 features of which are related to heart disease. This paper also uses four common categories such as decision tree to build the model. The data set studied in this article contains 270 records as well as 14 features. The accuracy of predicting the results of the support vector machine classifications, k nearest neighbor, decision tree and simple Bayesian is 81.11%, 66.67%, 59.72% and 19.85%, respectively, which are relatively satisfactory results.