Search Results (289)

The agri-food sector is undergoing a comprehensive transformation as it transitions towards net zero. To achieve this, fundamental changes and innovations are required, including changes in how food is produced and delivered to customers, new technologies, data and physical infrastructures, and algorithmic advancements. In this paper, we explore the opportunities and challenges of deploying AI-based data infrastructures for sustainability in the agri-food sector by focusing on two case studies: soft-fruit production and brewery operations. We investigate the potential benefits of incorporating Internet of Things (IoT) sensors and AI technologies for improving the use of resources, reducing carbon footprints, and enhancing decision-making. We identify user engagement with new technologies as a key challenge, together with issues in data quality arising from environmental volatility, difficulties in generalising models, including those designed for carbon calculators, and socio-technical barriers to adoption. We highlight and advocate for user engagement, more granular availability of sensor, production, and emissions data, and more transparent carbon footprint calculations. Our proposed future directions include semantic data integration to enhance interoperability, the generation of synthetic data to overcome the lack of real-world farm data, and multi-objective optimisation systems to model the competing interests between yield and sustainability goals. In general, we argue that AI is not a silver bullet for net zero challenges in the agri-food industry, but at the same time, AI solutions, when appropriately designed and deployed, can be a useful tool when operating in synergy with other approaches. Full article

Patient satisfaction is a key measure of the quality of healthcare, directly impacting the success and competitiveness of healthcare providers in an increasingly demanding market. Traditional feedback collection methods often fall short of capturing the full spectrum of patient experiences, leading to skewed satisfaction reports due to patients’ reluctance to criticize services and the inherent limitations of survey designs. To address these issues, advanced Natural Language Processing (NLP) techniques such as aspect-based sentiment analysis are emerging as essential tools. Aspect-based sentiment analysis breaks down the feedback text into specific aspects and evaluates the sentiment for each aspect, offering a more nuanced and actionable understanding of patient opinions. Despite its potential, aspect-based sentiment analysis is under-explored in the healthcare sector, particularly in the Arabic literature. This study addresses this gap by performing an Arabic aspect-based sentiment analysis on patient experience data, introducing the newly constructed Hospital Experiences Arabic Reviews (HEAR) dataset, and conducting a comparative study using Bidirectional Embedding Representations from Transformers (BERT) combined with machine learning classifiers, as well as fine-tuning BERT models, including MARBERT, ArabicBERT, AraBERT, QARiB, and CAMeLBERT. Additionally, the performance of GPT-4 via OpenAI’s ChatGPT is evaluated in this context, making a significant contribution to the comparative study of BERT with traditional classifiers and the assessment of GPT-4 for aspect-based sentiment analysis in healthcare, ultimately offering valuable insights for enhancing patient experiences through the use of AI-driven approaches. The results show that the joint model leveraging MARBERT and SVM achieves the highest accuracy of 92.14%, surpassing other models, including GPT-4, in both aspect category detection and polarity tasks. Full article

This study seeks to understand the key success factors that underpin efficiency, transparency, and user trust in automated decision support systems (DSS) that leverage AI technologies across industries. The aim of this study is to facilitate more accurate decision-making with such AI-based DSS, as well as build trust through the need for visibility and explainability by increasing user acceptance. This study primarily examines the nature of AI-based DSS adoption and the challenges of maintaining system transparency and improving accuracy. The results provide practical guidance for professionals and decision-makers to develop AI-driven decision support systems that are not only effective but also trusted by users. The results are also important to gain insight into how artificial intelligence fits into and combines with decision-making, which can be derived from research when thinking about embedding systems in ethical standards. Full article

Background: Obstructive sleep apnea is a sleep disorder that is linked to many health complications and can even be lethal in its severe form. Overnight polysomnography is the gold standard for diagnosing apnea, which is expensive, time-consuming, and requires manual analysis by a sleep expert. Artificial intelligence (AI)-embedded wearable device as a portable and less intrusive monitoring system is a highly desired alternative to polysomnography. However, AI models often require substantial storage capacity and computational power for edge inference which makes it a challenging task to implement the models in hardware with memory and power constraints. Methods: This study demonstrates the implementation of depth-wise separable convolution (DSC) as a resource-efficient alternative to spatial convolution (SC) for real-time detection of apneic activity. Single lead electrocardiogram (ECG) and oxygen saturation (SpO₂) signals were acquired from the PhysioNet databank. Using each type of convolution, three different models were developed using ECG, SpO₂, and model fusion. For both types of convolutions, the fusion models outperformed the models built on individual signals across all the performance metrics. Results: Although the SC-based fusion model performed the best, the DSC-based fusion model was 9.4, 1.85, and 11.3 times more energy efficient than SC-based ECG, SpO₂, and fusion models, respectively. Furthermore, the accuracy, precision, and specificity yielded by the DSC-based fusion model were comparable to those of the SC-based individual models (~95%, ~94%, and ~94%, respectively). Conclusions: DSC is commonly used in mobile vision tasks, but its potential in clinical applications for 1-D signals remains unexplored. While SC-based models outperform DSC in accuracy, the DSC-based model offers a more energy-efficient solution with acceptable performance, making it suitable for AI-embedded apnea detection systems. Full article

This paper proposes a new benchmark specifically designed for in-sensor digital machine learning computing to meet an ultra-low embedded memory requirement. With the exponential growth of edge devices, efficient local processing is essential to mitigate economic costs, latency, and privacy concerns associated with the centralized cloud processing. Emerging intelligent sensors equipped with computing assets to run neural network inferences and embedded in the same package, which hosts the sensing elements, present new challenges due to their limited memory resources and computational skills. This benchmark evaluates models trained with Quantization Aware Training (QAT) and compares their performance with Post-Training Quantization (PTQ) across three use cases: Human Activity Recognition (HAR) by means of the SHL dataset, Physical Activity Monitoring (PAM) by means of the PAMAP2 dataset, and superficial electromyography (sEMG) regression with the NINAPRO DB8 dataset. The results demonstrate the effectiveness of QAT over PTQ in most scenarios, highlighting the potential for deploying advanced AI models on highly resource-constrained sensors. The INT8 versions of the models always outperformed their FP32, regarding memory and latency reductions, except for the activations for CNN. The CNN model exhibited reduced memory usage and latency with respect to its Dense counterpart, allowing it to meet the stringent 8KiB data RAM and 32 KiB program RAM limits of the ISPU. The TCN model proved to be too large to fit within the memory constraints of the ISPU, primarily due to its greater capacity in terms of number of parameters, designed for processing more complex signals like EMG. This benchmark aims to guide the development of efficient AI solutions for In-Sensor Machine Learning Computing, fostering innovation in the field of Edge AI benchmarking, such as the one conducted by the MLCommons-Tiny working group. Full article

The detection of essays written by AI compared to those authored by students is increasingly becoming a significant issue in educational settings. This research examines various numerical text representation techniques to improve the classification of these essays. Utilizing a diverse dataset, we undertook several preprocessing steps, including data cleaning, tokenization, and lemmatization. Our system analyzes different text representation methods such as Bag of Words, TF-IDF, and fastText embeddings in conjunction with multiple classifiers. Our experiments showed that TF-IDF weights paired with logistic regression reached the highest accuracy of 99.82%. Methods like Bag of Words, TF-IDF, and fastText embeddings achieved accuracies exceeding 96.50% across all tested classifiers. Sentence embeddings, including MiniLM and distilBERT, yielded accuracies from 93.78% to 96.63%, indicating room for further refinement. Conversely, pre-trained fastText embeddings showed reduced performance, with a lowest accuracy of 89.88% in logistic regression. Remarkably, the XGBoost classifier delivered the highest minimum accuracy of 96.24%. Specificity and precision were above 99% for most methods, showcasing high capability in differentiating between student-created and AI-generated texts. This study underscores the vital role of choosing dataset-specific text representations to boost classification accuracy. Full article

Thanks to the rapid advancements in Connected and Automated Vehicles (CAVs) and vehicular communication technologies, the concept of the Internet of Vehicles (IoVs) combined with Artificial Intelligence (AI) and big data promotes the vision of an Intelligent Transportation System (ITS). An ITS is critical in enhancing road safety, traffic efficiency, and the overall driving experience by enabling a comprehensive data exchange platform. However, the open and dynamic nature of IoV networks brings significant performance and security challenges to IoV data acquisition, storage, and usage. To comprehensively tackle these challenges, this paper proposes a Decentralized Digital Watermarking framework for smart Vehicular networks (D2WaVe). D2WaVe consists of two core components: FIAE-GAN, a novel feature-integrated and attention-enhanced robust image watermarking model based on a Generative Adversarial Network (GAN), and BloVA, a Blockchain-based Video frames Authentication scheme. By leveraging an encoder–noise–decoder framework, trained FIAE-GAN watermarking models can achieve the invisibility and robustness of watermarks that can be embedded in video frames to verify the authenticity of video data. BloVA ensures the integrity and auditability of IoV data in the storing and sharing stages. Experimental results based on a proof-of-concept prototype implementation validate the feasibility and effectiveness of our D2WaVe scheme for securing and auditing video data exchange in smart vehicular networks. Full article

Neural networks (NNs) have proven their ability to deal with many computer vision tasks, including image-based remote sensing such as the identification and segmentation of hyperspectral images captured by satellites. Often, NNs run on a ground system upon receiving the data from the satellite. On the one hand, this approach introduces a considerable latency due to the time needed to transmit the satellite-borne images to the ground station. On the other hand, it allows the employment of computationally intensive NNs to analyze the received data. Low-budget missions, e.g., CubeSat missions, have computation capability and power consumption requirements that may prevent the deployment of complex NNs onboard satellites. These factors represent a limitation for applications that may benefit from a low-latency response, e.g., wildfire detection, oil spill identification, etc. To address this problem, in the last few years, some missions have started adopting NN accelerators to reduce the power consumption and the inference time of NNs deployed onboard satellites. Additionally, the harsh space environment, including radiation, poses significant challenges to the reliability and longevity of onboard hardware. In this review, we will show which hardware accelerators, both from industry and academia, have been found suitable for onboard NN acceleration and the main software techniques aimed at reducing the computational requirements of NNs when addressing low-power scenarios. Full article

Spam reviews pose a significant challenge to the integrity of online platforms, misleading consumers and undermining the credibility of genuine feedback. This paper introduces an innovative AI-generated spam review detection framework that leverages Deep Learning algorithms and Natural Language Processing (NLP) techniques to identify and mitigate spam reviews effectively. Our framework utilizes multiple Deep Learning models, including Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, Gated Recurrent Unit (GRU), and Bidirectional LSTM (BiLSTM), to capture intricate patterns in textual data. The system processes and analyzes large volumes of review content to detect deceptive patterns by utilizing advanced NLP and text embedding techniques such as One-Hot Encoding, Word2Vec, and Term Frequency-Inverse Document Frequency (TF-IDF). By combining three embedding techniques with four Deep Learning algorithms, a total of twelve exhaustive experiments were conducted to detect AI-generated spam reviews. The experimental results demonstrate that our approach outperforms the traditional machine learning models, offering a robust solution for ensuring the authenticity of online reviews. Among the models evaluated, those employing Word2Vec embeddings, particularly the BiLSTM_Word2Vec model, exhibited the strongest performance. The BiLSTM model with Word2Vec achieved the highest performance, with an exceptional accuracy of 98.46%, a precision of 0.98, a recall of 0.97, and an F1-score of 0.98, reflecting a near-perfect balance between precision and recall. Its high F2-score (0.9810) and F0.5-score (0.9857) further highlight its effectiveness in accurately detecting AI-generated spam while minimizing false positives, making it the most reliable option for this task. Similarly, the Word2Vec-based LSTM model also performed exceptionally well, with an accuracy of 97.58%, a precision of 0.97, a recall of 0.96, and an F1-score of 0.97. The CNN model with Word2Vec similarly delivered strong results, achieving an accuracy of 97.61%, a precision of 0.97, a recall of 0.96, and an F1-score of 0.97. This study is unique in its focus on detecting spam reviews specifically generated by AI-based tools rather than solely detecting spam reviews or AI-generated text. This research contributes to the field of spam detection by offering a scalable, efficient, and accurate framework that can be integrated into various online platforms, enhancing user trust and the decision-making processes. Full article

In an AI-immersing age, scholars look for new possibilities of employing AI technology to their fields, and how to strengthen security and protect privacy is no exception. In a coverless data hiding domain, the embedding capacity of an image generally depends on the size of a chosen database. Therefore, choosing a suitable database is a critical issue in coverless data hiding. A novel coverless data hiding approach is proposed by applying deep learning models to generate texture-like cover images or code images. These code images are then used to construct steganographic images to transmit covert messages. Effective mapping tables between code images in the database and hash sequences are established during the process. The cover images generated by a two-dimensional fractional Brownian motion (2D FBM) are simply called fractional Brownian images (FBIs). The only parameter, the Hurst exponent, of the 2D FBM determines the patterns of these cover images, and the seeds of a random number generator determine the various appearances of a pattern. Through the 2D FBM, we can easily generate as many FBIs of multifarious sizes, patterns, and appearances as possible whenever and wherever. In the paper, a deep learning model is treated as a secret key selecting qualified FBIs as code images to encode corresponding hash sequences. Both different seeds and different deep learning models can pick out diverse qualified FBIs. The proposed coverless data hiding scheme is effective when the amount of secret data is limited. The experimental results show that our proposed approach is more reliable, efficient, and of higher embedding capacity, compared to other coverless data hiding methods. Full article

Systematic reviews are a powerful tool to summarize the existing evidence in medical literature. However, identifying relevant articles is difficult, and this typically involves structured searches with keyword-based strategies, followed by the painstaking manual selection of relevant evidence. A.I. may help investigators, for example, through topic modeling, i.e., algorithms that can understand the content of a text. We applied BERTopic, a transformer-based topic-modeling algorithm, to two datasets consisting of 6137 and 5309 articles, respectively, used in recently published systematic reviews on peri-implantitis and bone regeneration. We extracted the title of each article, encoded it into embeddings, and input it into BERTopic, which then rapidly identified 14 and 22 topic clusters, respectively, and it automatically created labels describing the content of these groups based on their semantics. For both datasets, BERTopic uncovered a variable number of articles unrelated to the query, which accounted for up to 30% of the dataset—achieving a sensitivity of up to 0.79 and a specificity of at least 0.99. These articles could have been discarded from the screening, reducing the workload of investigators. Our results suggest that adding a topic-modeling step to the screening process could potentially save working hours for researchers involved in systematic reviews of the literature. Full article

The recent emergence of generative AI, particularly prompt-based models, and its embedding in many social domains and practices has revived the notion of co-creation and distributed agency already familiar in art practice and theory. Drawing on Actor-Network Theory (ANT) and its central notion of agency, this article explores the extent to which the collaboration between the artist and AI represents a new form of co-creation and distributed agency. It compares AI art with artistic movements such as Dada, Surrealism, Minimalism and Conceptual Art, which also challenged the notion of the autonomous artist and her agency by incorporating randomness on the one hand and rule-based systems on the other. In contrast, artistic practice with AI can be described as an iterative process of creative feedback loops, oscillating between order and disorder, (calculated) randomness and calculation, enabling a very specific kind of self-reflection and entanglement with the alienation of one’s own perspective. Furthermore, this article argues that most artistic projects that explore and work with AI are, in their own specific way, a demonstration of hybridity and entanglement, as well as the distribution of agency between the human and the non-human, and can thus be described as a network phenomenon. Full article

In this paper, the concept of symmetry is used to design the efficient inference of a fall-detection algorithm for elderly people on embedded processors—that is, there is a symmetric relation between the model’s structure and the memory footprint on the embedded processor. Artificial intelligence (AI) and, more particularly, Long Short-Term Memory (LSTM) neural networks are commonly used in the detection of falls in the elderly population based on acceleration measures. Nevertheless, embedded systems that may be utilized on wearable or wireless sensor networks have a hurdle due to the customarily massive dimensions of those networks. Because of this, the algorithms’ most popular implementation relies on edge or cloud computing, which raises privacy concerns and presents challenges since a lot of data need to be sent via a communication channel. The current work proposes a memory occupancy model for LSTM-type networks to pave the way to more efficient embedded implementations. Also, it offers a sensitivity analysis of the network hyper-parameters through a grid search procedure to refine the LSTM topology network under scrutiny. Lastly, it proposes a new methodology that acts over the quantization granularity for the embedded AI implementation on wearable devices. The extensive simulation results demonstrate the effectiveness and feasibility of the proposed methodology. For the embedded implementation of the LSTM for the fall-detection problem on a wearable platform, one can see that an STM8L low-power processor could support a 40-hidden-cell LSTM network with an accuracy of $96.52\%$. Full article

The rapid advancement of artificial intelligence (AI) technology, combined with the widespread proliferation of Internet of Things (IoT) devices, has significantly expanded the scope of AI applications, from data centers to edge devices. Running AI applications on edge devices requires a careful balance between data processing performance and energy efficiency. This challenge becomes even more critical when the computational load of applications dynamically changes over time, making it difficult to maintain optimal performance and energy efficiency simultaneously. To address these challenges, we propose a novel processing-in-memory (PIM) technology that dynamically optimizes performance and power consumption in response to real-time workload variations in AI applications. Our proposed solution consists of a new PIM architecture and an operational algorithm designed to maximize its effectiveness. The PIM architecture follows a well-established structure known for effectively handling data-centric tasks in AI applications. However, unlike conventional designs, it features a heterogeneous configuration of high-performance PIM (HP-PIM) modules and low-power PIM (LP-PIM) modules. This enables the system to dynamically adjust data processing based on varying computational load, optimizing energy efficiency according to the application’s workload demands. In addition, we present a data placement optimization algorithm to fully leverage the potential of the heterogeneous PIM architecture. This algorithm predicts changes in application workloads and optimally allocates data to the HP-PIM and LP-PIM modules, improving energy efficiency. To validate and evaluate the proposed technology, we implemented the PIM architecture and developed an embedded processor that integrates this architecture. We performed FPGA prototyping of the processor, and functional verification was successfully completed. Experimental results from running applications with varying workload demands on the prototype PIM processor demonstrate that the proposed technology achieves up to 29.54% energy savings. Full article

In the last few decades, major progress has been made in the medical field; in particular, new treatments and advanced health technologies allow for considerable improvements in life expectancy and, more broadly, in quality of life. As a consequence, the number of elderly people is expected to increase in the following years. This trend, along with the need to improve the independence of frail people, has led to the development of unobtrusive solutions to monitor daily activities and provide feedback in case of risky situations and falls. Monitoring devices based on radar sensors represent a possible approach to tackle postural analysis while preserving the person’s privacy and are especially useful in domestic environments. This work presents an innovative solution that combines millimeter-wave radar technology with artificial intelligence (AI) to detect different types of postures: a series of algorithms and neural network methodologies are evaluated using experimental acquisitions with healthy subjects. All methods produce very good results according to the main parameters evaluating performance; the long short-term memory (LSTM) and GRU show the most consistent results while, at the same time, maintaining reduced computational complexity, thus providing a very good candidate to be implemented in a dedicated embedded system designed to monitor postures. Full article