Saffron is a rare and valuable crop that is only cultivated in specific regions with suitable topographical conditions. To improve saffron cultivation, it is crucial to monitor and precisely control the crop’s agronomic variables over at least one growth cycle to create a fully automated environment. To this end, agronomic variables in the Punjab region of India were analyzed and set points were calculated using third-order polynomial equations through the application of image processing techniques. The relationship between canopy cover, growth percentage, and agronomic variables was also investigated for optimal yield and quality. The addition of adulterants, such as turmeric and artificial colorants, to saffron is a major concern due to the potential for quality compromise and fraud by supply chain vendors. Hence, there is a need for devising an easy, reliable, and user-friendly mechanism to help in the detection of adulterants added to the saffron stigmas. This paper proposes an automated IoT-based saffron cultivation environment using sensors for determining set points of agronomical variables. In addition, a sensor-based chamber has been proposed to provide quality and adulteration checks of saffron and to eliminate product counterfeiting. The AquaCrop simulator was employed to evaluate the proposed framework’s performance. The results of the simulation show improved biomass, yield, and harvest index compared with the existing solutions in precision agriculture. Given the high value and demand for saffron, ensuring its purity and quality is essential to sustain its cultivation and the economic viability of the market.

As K-wave has been strengthened via recent K-contents, K-wave items such as cosmetics and electronic devices have also gained attention globally. For small-sized export sellers who purchased the items and exported them to different countries, it is significant to discover which K-wave items are trending in specific countries. To do so, we proposed an ensemble recommender system by producing the dense vector, which is generated by a variant of Bidirectional Encoder Representations from Transformers (BERT), and balancing the vector with a sparse vector in order to ensure the efficient execution speed and recommendation accuracy. Based on the data we have collected specifically for potential K-items, our experiment showed that the proposed model outperforms the various baselines, which are used for content-based filtering.

Domain adaptation aims to effectively learn from two domains with different distributions, solving labeling problems; however, traditional methods assume that the source and target data are in-distribution data that share the same labels. In practice, Out-Of-Distribution (OOD) data which do not share labels with the existing data may also be collected during the target data collection process. These OOD data introduce noise and confusion, leading to decreased performance during adaptation. To address this issue, we propose RObust Domain Adaptation from Out-Of-Distribution data (RODA-OOD), a novel method based on data-centric AI principles that focuses on improving data quality rather than refining model architecture. RODA-OOD utilizes the characteristics of deep learning models that prioritize learning in-distribution data, which are easier to train on compared to OOD data. By dynamically adjusting the threshold for OOD detection, the proposed method effectively filters out OOD data, allowing the model to focus on relevant target data. RODA-OOD was compared with competitor and original domain adaptation algorithms based on target data accuracy. The results show that RODA-OOD demonstrates the most robust performance against OOD data, achieving a 21.3% increase in accuracy compared to existing domain adaptation methods. Thus, RODA-OOD can provide a solution to the OOD issue in unsupervised domain adaptation.

With the increasing complexity and integration density of System-in-Package (SiP) technologies, the demand for selective electromagnetic interference (EMI) shielding is growing. Conventional sputtering processes, while effective for conformal EMI shielding, lack selectivity and often require additional masking or post-processing steps. In this study, we propose a novel, laser-based approach for the selective removal of EMI shielding layers without physical masking. Numerical simulations were conducted to investigate the thermal and mechanical behavior of multilayer EMI shielding structures under two irradiation modes: full-area and laser scanning. The results showed that the laser scanning method induced higher interfacial shear stress, reaching up to 38.6 MPa, compared to full-area irradiation (12.5 MPa), effectively promoting delamination while maintaining the integrity of the underlying epoxy mold compound (EMC). Experimental validation using a nanosecond pulsed fiber laser confirmed that complete removal of the EMI shielding layer could be achieved at optimized laser powers (~6 W) without damaging the EMC, whereas excessive power (8 W) caused material degradation. The laser scanning speed was 50 mm/s, and the total laser irradiation time of the package was 0.14 s, which was very fast. This study demonstrates the feasibility of a non-contact, damage-free, and selective EMI shielding removal technique, offering a promising solution for next-generation semiconductor packaging.

The rise of online learning has led to the development of various knowledge tracing (KT) methods. However, existing methods have overlooked the problem of increasing computational cost when utilizing large graphs and long learning sequences. To address this issue, we introduce Dual Graph Attention-based Knowledge Tracing (DGAKT), a graph neural network model designed to leverage high-order information from subgraphs representing student-exercise-KC relationships. DGAKT incorporates a subgraph-based approach to enhance computational efficiency. By processing only relevant subgraphs for each target interaction, DGAKT significantly reduces memory and computational requirements compared to full global graph models. Extensive experimental results demonstrate that DGAKT not only outperforms existing KT models but also sets a new standard in resource efficiency, addressing a critical need that has been largely overlooked by prior KT approaches.