Data Envelopment Analysis (DEA) is an extremely flexible and useful methodology, which provides a relative efficiency score for each decision-making unit.Classic DEA models only evaluate the performance efficiency of units and they cannot provide any information about the progress and regress of units in two periods. Also, they suppose that all inputs and outputs are positive real numbers. But in the real world, due to the existence of uncertainty, this assumption may not always be true. So, the DEA models with fuzzy data (FDEA models) can more realistically represent real-world problems than the conventional DEA models. In this paper, we assume all inputs, outputs and efficiency measures are triangular fuzzy numbers. So we present a method for obtaining the fuzzy Malmquist productivity index on fuzzy data, and finally, we can determine progress and regression for units with fuzzy data. Manuscript profile

مشابهت یابی معنایی متون (STS)یک وظیفه چالش‌برانگیز در زبان‌های با منابع دیجیتالی محدود است، دشواری‌های اصلی ناشی از کمبود مجموعه‌های آموزشی دسته‌بندی‌شده و مشکلات مرتبط با آموزش مدل‌های مؤثر است. در اینجا یک رویکرد یادگیری مشترک با استفاده از مدل خودتوجه بهبودیافته برای مقابله با چالش STS در ساختارهای زبانی (فاعل، مفعول، فعل) SOV و (فاعل، فعل، مفعول) SVO معرفی شده است. ابتدا یک مجموعه داده چندزبانه جامع با داده‌های موازی برای زبان‌های SOV و SVO را ایجاد کرده و تنوع زبانی گسترده‌ای را تضمین می‌کنیم. ما یک مدل خودتوجه بهبودیافته با رمزگذاری نسبی موقعیت وزن‌دار جدید غنی‌شده با تزریق اطلاعات هم‌رخدادی از طریق عوامل اطلاعات مشترک نقطه‌ای (PMI) معرفی می‌کنیم. علاوه بر این، ما از یک چارچوب یادگیری مشترک استفاده می‌کنیم که نمونه های مشترک بین زبان‌ها را به منظور بهبود STS بین زبانی استفاده می‌کند. با آموزش همزمان در چندین جفت زبان، مدل ما توانایی انتقال دانش را به دست می‌آورد و به طور مؤثر پل ارتباطی بین زبان‌های با ساختارهای متفاوت SOV و SVO ایجاد می کند. مدل پیشنهادی ما بر روی مجموعه داده‌های STS- Benchmarks فارسی-انگلیسی و فارسی-فارسی ارزیابی شد و به ترتیب به ضریب همبستگی پیرسون 88.29٪ و 91.65٪ دست‌یافت. آزمایش‌های انجام‌شده نشان می‌دهند که مدل پیشنهادی ما در مقایسه با مدل‌های دیگر عملکرد بهتری داشته است. مطالعه کاهشی نیز نشان می‌دهد که سیستم ما قادر به همگرایی سریعتر است و کمتر مستعد بیش برازش است Manuscript profile

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. Manuscript profile

In recent years, increasing carbon emissions and relatively unfavorable climate change have led to paying attention to the concepts of sustainability as well as the imposition of strict government regulations on manufacturers and service providers. This has caused all parts of society, containing consumers, governments, and companies, to pay greater attention to low-carbon manufacturing in the supply chain (SC). To this end, this paper has designed a sustainable, multi-echelon, multi-product, multi-period, and multi-objective closed-loop supply chain (CLSC) network, with different distribution and collection channels and Radio Frequency Identification (RFID) technology, in addition, in order to produce low-carbon products, low-carbon products and subsidizing policies, and also deals with pricing strategy. This model, at the same time, maximizes profits and the social responsibility of the SC network, while it minimizes the overall delay in delivery time and environmental pollution. To cope with the parameters' uncertainty, a Robust scenario-based Stochastic programming (RSSP) approach has been used, and to solve and validate, the small-size model the Augmented Epsilon Constraint (AEC) method, and to solve large-sized ones, the third edition of the Non-dominated Sorting Genetic Algorithm (NSGA-III) and Multi-Objective Grey Wolf Optimizer Algorithm (MOGWO) are used. According to the computational results, the suggested model can provide efficient decisions and the MOGWO algorithm yields 14.5% improvement in execution time compared to the NSGA-III algorithm. Also suggested model can be a great tool for managers and professionals with a wide range of strategic applications. Manuscript profile