Abstract: Artificial Intelligence (AI) has been widely successful and effective in optimizing tasks across various fields such as smart agriculture, healthcare, and education, positioning it as a key enabler in advancing the Sustainable Development Goals (SDGs) set forth by the United Nations. Understanding the perspectives of industry and academic experts in domains such as AI, data science, biotechnology, natural and physical sciences, can provide valuable multidisciplinary insights into the responsible usage of AI. This study analyses 12 opinion essays written by 13 experts from academia and industry, focusing on the opportunities and challenges provided by AI in the advancement of SDGs. These experts provide their unique perspectives shaped by their nationality, professional role, gender, and domain of expertise. The analysis highlights several strong application areas of AI, including enhancing diagnostic accuracy, monitoring cattle and crops to reduce wastage, ensuring accessibility of education, and enhancing gender equality. However, the experts also caution against potential risks and ethical implications associated with AI, such as the risk of algorithmic bias, concerns with over reliance, and inequitable access to AI enabled tools.

Abstract: Understanding customer purchasing behavior is essential for businesses to optimize marketing strategies and improve customer retention. This study employs machine learningbased clustering techniques to segment customers based on transactional data. By leveraging Recency, Frequency, and Monetary (RFM) analysis, the study compares multiple clustering algorithms to identify distinct customer groups. Experimental results demonstrate that the proposed approach effectively categorizes customers, enabling data-driven decision-making for targeted marketing. These findings highlight the potential of unsupervised learning methods in enhancing business intelligence and customer relationship management.

Abstract: The growing demand for cost-efficient digital transactions has driven the need for scalable and low-cost payment solutions. Traditional blockchain-based transactions suffer from high fees and slow processing times, making decentralized off-chain payment networks a promising alternative. In this paper, we propose SpeedyMurmurs, an AI-enhanced decentralized routing algorithm that significantly reduces payment processing costs and transaction delays. Our approach optimizes payment routing efficiency through embedding-based path discovery, reducing routing overhead by up to two orders of magnitude and cutting transaction processing times by over 50 percent compared to existing blockchain networks. By leveraging machine learning-driven transaction optimization, our system dynamically selects the most cost-effective paths for digital payments while maintaining user privacy and security. Experimental results demonstrate that SpeedyMurmurs reduces transaction fees and computational costs, making decentralized payment systems more financially viable. This research highlights the role of AI-powered routing strategies in minimizing costs and improving efficiency in modern payment networks.

Abstract: Vehicular Ad-hoc Networks (VANETs) face critical challenges regarding intermittent connectivity and latency due to high node mobility, often resulting in a performance trade-off between reactive and proactive routing paradigms. This study aims to resolve these inherent limitations and ensure reliable communication in volatile environments. We propose a novel context-aware framework, the Dynamic Hybrid Routing Protocol (DHRP), which integrates Ad hoc On-Demand Distance Vector (AODV) and Optimized Link State Routing (OLSR). Distinguished by a predictive multi-criteria switching logic and a hysteresis-based stability mechanism, the proposed method employs a synergistic cross-layer framework that adapts transmission power and routing strategy in real time. Validated through extensive simulations using NS-3 and SUMO, experimental results demonstrate that the protocol outperforms traditional baselines and contemporary benchmarks across all key metrics. Specifically, the system maintains a Packet Delivery Ratio (PDR) exceeding 90%, ensures end-to-end delays remain under the safety-critical 40 ms threshold, and achieves energy savings of up to 60%. In conclusion, DHRP successfully resolves the routing performance dichotomy, providing a scalable, energy-efficient foundation for next-generation Intelligent Transportation Systems (ITS) in which reliable safety messaging is paramount.

Abstract: The field of agentic artificial intelligence is transitioning from reasoning-centric architectures toward systems explicitly designed for reliability under uncertainty. Current agent frameworks, while effective in controlled environments, exhibit cognitive rigidity—an inability to proactively correct planning trajectories when confronted with unexpected faults. This paper introduces Adapt-Plan, a foundational hybrid architecture that reformulates planning as a control-theoretic process by elevating the Planning Efficiency Index (PEI) from a post-hoc evaluation metric to a real-time control signal. Through dual-mode planning (strategic and tactical) and Extended Dynamic Memory (EDM) for selective experience consolidation, Adapt-Plan enables agents to detect behavioral drift early and initiate corrective adaptation before catastrophic degradation occurs. Controlled validation across 150 episodes demonstrates PEI=0.91 ± 0.06 and FRR=78% ± 4.2% (95% CI [74%, 82%], p < 0.001, Cohen’s d = 2.18 vs. ReAct), establishing the algorithmic viability of metric-driven adaptation. Comprehensive ablation studies isolate component contributions, revealing that PEI-guided control accounts for 31% of performance gains. These architectural principles were subsequently validated at scale through rigorous certification frameworks, confirming that PEI-driven control generalizes to deployment-grade reliability when augmented with safety protocols. This work establishes the conceptual foundation for reliable agentic AI through the tight integration of architecture, metrics, and control.

Abstract:

The integration of artificial intelligence (AI) in precision agriculture marks a transformative step toward sustainable, efficient, and data-driven farming practices. By merging AI with predictive analytics and autonomous monitoring systems, agriculture is empowered to achieve higher crop yields and maintain robust soil health. AI-driven models process vast datasets from sensors, drones, and IoT devices to predict crop performance, recommend targeted interventions, and enable real-time monitoring of field conditions. This synergy not only allows for early detection of threats such as pests or nutrient deficiencies but also ensures optimized resource utilization, reducing environmental impact. The adoption of these intelligent systems paves the way for a resilient agricultural landscape that can adapt to the challenges posed by climate variability and the growing global food demand, ultimately fostering productivity and long-term ecological sustainability.

Abstract: To overcome the low accuracy of conventional methods for estimating liquid volume and food nutrient content in bowl-type tableware, as well as the tool dependence and time-consuming nature of manual measurements, this study proposes an integrated approach that combines geometric reconstruction with deep learning–based segmentation. After a one-time camera cali-bration, only a frontal and a top-down image of a bowl are required. The pipeline automatically extracts key geometric information, including rim diameter, base diameter, bowl height, and the inner-wall profile, to complete geometric modeling and capacity computation. The estimated parameters are stored in a reusable bowl database, enabling repeated predictions of liquid vol-ume and food nutrient content at different fill heights. We further propose Bowl Thick Net to predict bowl wall thickness with millimeter-level accuracy. In addition, we developed a Geome-try-aware Feature Pyramid Network (GFPN) module and integrated it into an improved Mask R-CNN framework to enable precise segmentation of bowl contours. By integrating the contour mask with the predicted bowl wall thickness, precise geometric parameters for capacity estima-tion can be obtained. Liquid volume is then predicted using the geometric relationship of the liq-uid or food surface, while food nutrient content is estimated by coupling predicted food weight with a nutritional composition database. Experiments demonstrate an arithmetic mean error of −3.03% for bowl capacity estimation, a mean liquid-volume prediction error of 9.24%, and a mean nutrient-content (by weight) prediction error of 11.49% across eight food categories.

Abstract:

We study a purely local inverse problem for non-reversible Randers metrics \( F = \|\cdot\|_g + \beta \) defined on smooth oriented surfaces. Using only the lengths of sufficiently small closed curves around a point \( p \), we prove that the exterior derivative \( d\beta(p) \) can be uniquely and stably recovered. Moreover, we establish that \( d\beta(p) \) is the only second-order local invariant retrievable from such local length measurements. Our approach is entirely metric-based, independent of geodesic flows or boundary data, and naturally extends to general curved surfaces.

Abstract: We present the Atemporal Tablet Framework (ATF), a complete geometric ontology that derives spacetime, quantum mechanics, and gravity from a single mathematical structure. The universe is modeled as a fiber bundle T ->(π) M where T is a static higher-dimensional manifold and M is emergent 3+1D spacetime. Temporal dynamics arise from projection operators Πt : T -> M extremizing a projective action SΠ. Quantum states are epistemic distributions over fibers, with the Born rule emerging naturally via measure disintegration. Measurement corresponds to topological phase-locking without wavefunction collapse. Einstein’s equations arise as equations of motion for Πt, while quantum fields emerge as fiber vibrations. The framework makes specific testable predictions: sidereal anisotropy in qubit decoherence ε = 1.23 × 10^-8 ± 3 × 10^-9 (derived from holographic scaling) and modified dispersion relations at scale EP / sqrt(ε). We prove a reconstruction theorem establishing that spacetime observations can determine the underlying geometry, and demonstrate that Standard Model particle content emerges naturally from Fx ≅ CP3 × S5 / Γ fiber geometry. ATF provides a mathematically rigorous, experimentally falsifiable foundation for quantum gravity that resolves long-standing interpretational issues while making concrete predictions testable with current technology.

Abstract: Fruits are an integral part of our diet. Various types of proteins and vitamins are obtained through fruits. Apple is a major fruit that is consumed globally. This is a multipurpose fruit that is used in the preparation of various food products and also in medicines. Therefore, it is important to analyze its future prices. India is the largest producer of apples, thus it is very important to analyze the Apple prices of Indian agricultural markets. Machine learning and deep learning models have not been previously applied to this Indian dataset. Various time series models like Long Short-Term Memory (LSTM), SARIMA, and ETS are developed, but the performance of LSTM is much better compared to the other models, with the lowest error rates (MAE of 554.08, RMSE of 752.10, 191, and MAPE of 6.63 percent). Thus, the proposed study provides the solution to a real-life problem, which ultimately can be used for agriculture policy making and smart market strategies.

Abstract:

This paper describes a new auto-associative network called a Unit-Merge Network. It is so-called because novel compound keys are used to link 2 nodes in 1 layer, with 1 node in the next layer. Unit nodes at the base store integer values that can represent binary words. The word size is critical and specific to the dataset and it also provides a first level of consistency over the input patterns. A second cohesion network then links the unit nodes list, through novel compound keys that create layers of decreasing dimension, until the top layer contains only 1 node for any pattern. Thus, a pattern can be found using a search and compare technique through the memory network. The Unit-Merge network is compared to a Hopfield network and a Sparse Distributed Memory (SDM). It is shown that the memory requirements are not unreasonable and that it has a much larger capacity than a discrete Hopfield network, for example. It can store sparse data, deal with noisy input and a complexity of O(log n) compares favourably with these networks. This is demonstrated with test results for 4 benchmark datasets. Apart from the unit size, the rest of the configuration is automatic, and its simplistic design could make it an attractive option for some applications.

Abstract: The increasing adoption of contactless transactions has introduced new security challenges, particularly in protecting data from unauthorized access and fraudulent activities. This study presents a novel authentication framework designed to enhance the security of near-field communication (NFC) transactions while maintaining efficiency for real-time processing. By leveraging a lightweight cryptographic approach, the proposed system mitigates threats such as relay attacks and unauthorized skimming. Experimental analysis demonstrates improvements in authentication reliability and resistance to common attack vectors. This research contributes to the development of more secure and scalable contactless payment solutions.

Abstract: With the growing reliance on peer-to-peer (P2P) networks for digital transactions, traditional electronic payment systems require enhancements to ensure security, efficiency, and trust. This study introduces an innovative digital payment framework enabling currency-based exchanges between consumers and vendors within a peer-to-peer environment. The outlined approach is inspired by Millicent’s scrip methodology and leverages digital envelope encryption to bolster protection. Unlike conventional payment methods that heavily rely on financial institutions, the protocol minimizes their involvement, restricting their role to trust establishment and transaction finalization. The system introduces a distributed allocation model, where merchants locally authorize payments, reducing transaction overhead and enhancing scalability. Additionally, the protocol is optimized for repeated payments, making it particularly efficient for recurring transactions between the same buyer and merchant. By integrating cryptographic techniques and decentralizing payment authorization, this protocol presents a secure, efficient, and scalable solution for digital payments in P2P environments.

Abstract: Near Field Communication (NFC) technology is increasingly being integrated into mobile devices, enabling applications such as contactless payments and public transportation access. This paper investigates the security architecture of NFC systems, focusing on mobile device implementations and the vulnerabilities they introduce. Various configurations for NFC’s Secure Element (SE), such as SD cards, multiple UICC slots, and shared SIM resources, are discussed, highlighting potential security challenges related to relay attacks, malware distribution, differential power analysis, and denial-of-service attacks. In particular, relay attacks and malware distribution are identified as significant threats that could compromise user security during transactions. The paper further explores countermeasures like two-factor authentication, distance-bounding protocols, and defensive cryptographic techniques to mitigate these risks. Additionally, it emphasizes the complexities introduced by trust issues between Mobile Network Operators (MNOs) and thirdparty providers in sharing secure resources. Finally, the research suggests that while NFC itself is relatively secure, applications built on top of this infrastructure are more prone to security risks. As NFC technology continues to evolve, ensuring robust security for its applications, particularly in the financial and healthcare sectors, will be critical to its widespread adoption.

Abstract: Identity and Access Management (IAM) increasingly relies on adaptive controls—step-up challenges, recovery verification, device and behavior signals, and continuous authorization—to reduce account takeover and misuse. At the same time, IAM systems must prepare for post-quantum cryptography (PQC) transitions that affect credentials, signing, and verification paths. These shifts create a practical governance problem: when an identity action is allowed, challenged, denied, or escalated (e.g., passwordless enrollment, recovery credential release, privileged step-up, or machine key rotation), teams must be able to explain why the decision happened, what evidence was considered, and how the decision can be independently verified later. This paper introduces Decision Receipts (DR): a verifiable, privacy-aware record produced at decision time that captures (i) policy context and versioning, (ii) normalized evidence descriptors (not raw personal data), (iii) action outcomes and reason codes, and (iv) cryptographic signatures supporting long-term auditability under PQC. We propose an open receipt schema, canonicalization rules, and verifier workflows using widely deployed identity standards (OAuth 2.0, OpenID Connect, JWT) and modern signing containers (JWS/COSE), with optional anchoring into transparency logs for tamper-evidence. The approach is intentionally IP-safe and adoptable as an audit overlay independent of any specific orchestrator implementation.

Abstract: There are added some matters for the estimation of \( H(n,m) \) in the appendix.

Abstract: This work addresses correlation bias and causal effect confounding in advertising recommendation systems and presents a causal learning–based recommendation framework. We first examine the limitations of conventional recommendation algorithms in complex advertising environments, where confounding variables and exposure bias often prevent models from capturing users’ true preferences. To tackle these issues, we design a unified embedding architecture that jointly represents user, advertisement, and contextual features, and incorporates a structural causal graph to explicitly model dependencies among variables. During model training, causal consistency regularization and inverse propensity weighting are integrated to mitigate the impact of biased exposure mechanisms and non-uniform sampling. A joint optimization objective is further formulated to couple click-through rate prediction with causal consistency estimation, enabling robust causal effect learning without sacrificing predictive accuracy. Extensive experiments on large-scale advertising datasets demonstrate that the proposed approach consistently outperforms several representative baselines in terms of Precision@10, Recall@10, NDCG@10, and MAP, while exhibiting strong robustness under multi-dimensional sensitivity analysis. Overall, this study highlights the practical value of causal modeling and consistency-aware learning in advertising recommendation and offers a computationally grounded approach for improving both interpretability and fairness in recommendation systems.

Abstract: The rapid evolution of data-driven fields demands educational paradigms that transition from static analysis to dynamic interaction with live information. This paper presents a novel technical framework, the Dual-Agent Curator-Tutor (DACT), which integrates Artificial Intelligence as a concurrent Real-Time Data Curator and Interactive Tutor within Immersive Analytics (IA) learning environments. The DACT framework features two synergistic AI agents: a Curation Agent that dynamically ingests, filters, and contextualizes live data streams (e.g., IoT, financial feeds) for pedagogical alignment, and a Tutoring Agent that provides adaptive, scaffolded instruction based on multimodal analysis of learner behavior within an immersive visualization space (VR/AR). This creates a closed-loop ecosystem where the data landscape and instructional guidance co-adapt in real-time to the learner’s actions. We detail a modular architecture implementing this model, utilizing perturbation-based learning for adaptive curation—inspired by recent optimization techniques—and a rule-based pedagogical engine. We propose a rigorous quantitative evaluation methodology involving controlled experiments to measure gains in analytical proficiency, cognitive load reduction, and behavioral patterns. The paper argues that this seamless integration of automated data management and personalized tutoring within an immersive context represents a transformative advancement for experiential learning, effectively leveraging technology to offload cognitive overhead and elevate higher-order analytical reasoning skills.

Abstract: Traditional kernel fuzzers rely on coarse-grained coverage metrics that cannot reflect complex microarchitectural behaviors. We present a hardware-assisted fuzzing framework that leverages branch buffer telemetry from modern CPUs (LBR, BTB sampling) to refine fuzzing feedback. A model-based inference algorithm aggregates branch-data patterns to estimate microarchitectural novelty and guides seed prioritization. Experiments on Intel Ice Lake and AMD Zen 3 systems demonstrate 27% improvement in unique path coverage, with 11 newly identified concurrency bugs across filesystem and scheduler subsystems. Compared with coverage-only fuzzing, our method reduces time-to-crash by 46% while keeping overhead below 12%. This work shows microarchitectural-level signals can significantly boost kernel fuzzing’s effectiveness.

Abstract: This study develops an adaptive workflow allocation mechanism for anti-money laundering (AML) operations, aiming to improve the accuracy and efficiency of suspicious-transaction review. A multi-agent simulation platform was constructed to model transaction flows, alert generation, and analyst decision behaviors. The system integrates model-confidence estimation, analyst-fatigue prediction, and real-time workload signals to dynamically route alerts. Experiments were conducted using 27.3 million historical transactions and 186,000 alerts from a large commercial financial dataset. Compared with fixed allocation rules, the adaptive mechanism increased alert-escalation precision from 0.32 to 0.46 and recall from 0.70 to 0.78, while reducing average handling time by 19.4%. The proportion of high-risk alerts processed within the target time window improved by 23.8%. These results demonstrate that workflow optimization can meaningfully enhance AML performance beyond model-level improvements.