Khorramshahr University of Marine Science and Technology

The present study was conducted with the aim of developing a structural–interpretive model (ISM) for customer confusion management based on the Internet of Things (IoT) in the banking industry. This research is applied in nature and qualitative in approach, employing thematic analysis and the structural–interpretive modeling method to explain the relationships among the factors influencing customer confusion management. The statistical population consisted of 17 managers and experts from Bank Mellat branches in Tehran, who were selected through purposive and snowball sampling methods. Data were collected using semi-structured interviews and analyzed with MAXQDA 2018 software. The results of the thematic analysis led to the identification of 77 indicators categorized into 14 main components and six key dimensions. These dimensions include sources of confusion, consequences of customer confusion, internet infrastructures, customer characteristics affecting the increase in confusion, organizational learning management approaches for IoT, and organizational factors influencing confusion management. The findings of the interpretive structural model indicated that the organizational learning management approach for IoT has the highest level of influence, while the sources and consequences of confusion occupy the lowest level. The results further revealed that employing IoT in the banking industry—through enhancing technological infrastructures, automating service processes, and implementing targeted customer education—can effectively reduce confusion and improve customer experience. Therefore, strengthening organizational learning, developing modern technologies, and increasing informational transparency are regarded as fundamental requirements for the effective management of customer confusion in banks.

Water scarcity is widely regarded as a fundamental contributor to underdevelopment and poverty, significantly influencing individuals' and societies' perceptions of deprivation.This study seeks to elucidate the relationship between the "feeling of relative deprivation" and "hydropolitical relations," particularly in the context of the Gauvshomar Dam construction process in Lorestan Province.The primary research question addresses how the absence of dam construction impacts the development of a sense of relative deprivation among local populations.Employing a descriptive-analytical research methodology, this study adopts both quantitative and qualitative approaches, utilizing documentary and library research alongside field observations.A questionnaire serves as the primary measurement tool, with SPSS employed for data analysis.The research indicators encompass economic, social, cultural, environmental-physical, and political-security dimensions.Findings indicate that the lack of dam construction has exacerbated feelings of relative deprivation across these dimensions, contributing to a general sense of dissatisfaction among the residents of the study area.

Visual robots, leveraging robotic and computer vision techniques, play a crucial role in waste management and recycling due to their speed and accuracy. However, neural networks employed for trash recognition often struggle with classifying rotated images, undermining the reliability of recycling robots. In this paper, the application of rotation-invariant image classification models in recycling robots is explored. A transformer-based model, called RIFormer, is utilized to address this limitation. RIFormer is compared with state-of-the-art models like MobileNetV2, ResNet18, and Vision Transformer (ViT) to evaluate its effectiveness in recognizing rotated images at various angles, thereby enhancing the accuracy and reliability of recycling robots. The models were trained on the Augmented TrashNet Dataset, containing 10,000 samples, over 100 epochs. Experimental results demonstrate that RIFormer achieves a superior accuracy of $97 \%$, significantly outperforming other models, which fail to exceed $92 \%$. Additionally, RIFormer exhibits robustness against rotations, unlike its counterparts. This demonstrates that rotation-invariant models like RIFormer are more suitable for practical robotic applications.

A 60-day research was conducted to evaluate the influence of dietary fish oil (FO) and selenium nanoparticles (SeNPs) on performance of Penaeus vannamei juveniles (2.4 ± 0.0 g) reared in seawater (SW) or hypersaline (HS) water conditions. A 2 × 2 × 2 factorial research was designed with two FO levels, including 3% and 6%, two SeNPs dosages, including 0.4 and 0.8 mg/kg and two water salinity, including SW (35 g/L) and HS (50 g/L). Eight experimental groups were designed as follow: (1) FO3Se0.4SW (3% FO + 0.4 mg/kg SeNPs reared in SW), (2) FO3Se0.8SW (3% FO + 0.8 mg/kg SeNPs reared in SW), (3) FO6Se0.4SW (6% FO + 0.4 mg/kg SeNPs reared in SW), (4) FO6Se0.8SW (6% FO + 0.8 mg/kg SeNPs reared in SW), (5) FO3Se0.4HS (3% FO + 0.4 mg/kg SeNPs reared in HS), (6) FO3Se0.8HS (3% FO + 0.8 mg/kg SeNPs reared in HS), (7) FO6Se0.4HS (6% FO + 0.4 mg/kg SeNPs reared in HS), and (8) FO6Se0.8HS (6% FO + 0.8 mg/kg SeNPs reared in HS). Four hundred and eighty P. vannamei were randomly distributed into 24 250-L cylindrical fiberglass tanks. Each experimental group was replicated in three tanks (20 shrimp/tank). Shrimps were fed at 5% of their initial biomass three times daily. Water temperature and dissolved oxygen levels were maintained at 31.5 ± 1.2°C and 5.5 ± 1.0 mg/L, respectively. Shrimp in FO6Se0.4SW and FO6Se0.8SW had higher weight gain (WG) compared to those in FO3Se0.8SW and FO3Se0.4HS (p < 0.05). Rearing shrimp in HS water increased whole-body protein and ash contents but reduced moisture level. The amount of docosahexaenoic acid (DHA) in FO3Se0.4SW, FO6Se0.4SW, FO6Se0.8SW, and FO6Se0.4HS was relatively higher than the other treatments. The antioxidant activities, including glutathione peroxidase (GPx), superoxide dismutase (SOD), catalase (CAT), and glutathione (GSH) level decreased and malondialdehyde (MDA) content in hepatopancreas increased in shrimp reared in HS water. Finally, the transcription levels of insulin-like growth hormones and immune-related genes, including lectin (lec), peneidine 3 (pen-3), prophenoloxidase (propo), and lysozyme (lyz) increased by increasing dietary FO level of 6%. Additionally, these levels were modulated by the interaction of water salinity and dietary SeNPs and FO levels. Based on the findings of the current study, increasing the dietary FO level to 6% and incorporating a moderate level of SeNPs (0.4 mg/kg) were found to enhance P. vannamei tolerance to HS water.

This study was designed to examine and prioritize the effective factors on the success of Internet marketing of export companies by utilizing the TOPSIS multi-criteria decision-making technique. The study’s statistical population consisted of the managers of 52 export trading companies in Mazandaran province, Iran. A questionnaire was used to collect the required data and its validity questionnaire was confirmed by relevant experts and professors. Also, the reliability values of the questionnaire variables were obtained using Cronbach's alpha for internal factors, external factors, international factors, and internet marketing success as 0.84, 0.85, 0.83, and 0.86, respectively. The structural equation model method was used for the analysis of the research data and testing the hypotheses using SPSS and Lisrel software, while the TOPSIS technique was applied to prioritize the identified factors. As indicated by the findings of this research, three internal, external, and international factors can affect the success of Internet marketing. Furthermore, ranking by the TOPSIS method revealed that external factors are rated the first place in this rating, followed by internal factors and international factors in the second and third places.