Lightweight Internet of Things (IoT) devices—microcontroller-class nodes with less than 512KB RAM, sub-100MHz clocks, and low-power radios (BLE, Zigbee, LoRa, NB-IoT)—are now widely deployed in settings where traditional security stacks are infeasible. This survey examines how Artificial Intelligence (AI) can harden such constrained platforms against device-, network-, and application-layer threats, including spoofing, routing manipulation, DDoS, malware, and Advanced Persistent Threats (APTs). We (i) formalize alightweight envelopethat bounds feasible defenses in terms of RAM, CPU, bandwidth, and energy; (ii) consolidate protocol-side risks across BLE, Zigbee, and LoRaWAN; and (iii) review deployable AI techniques through adeployment-firstlens that separates training (edge, cloud, federated learning) from on-device inference. Distinct from prior surveys, we provide resource-annotated comparisons that report accuracyalongsidemodel size, peak RAM, latency, and estimated energy per inference, showing how pruning, post-training quantization, distillation, and feature narrowing shift feasibility on MCU targets. Covered methods include compact classifiers (linear models, trees, SVM), quantized TinyCNN/TinyRNN and graph-based intrusion detection, reinforcement learning for adaptive rate limiting and channel selection, and privacy-preserving federated learning with update compression. We conclude with a pragmatic agenda—energy-adaptive inference, LPWAN-aware scheduling and federated learning, robustness to poisoning and evasion, and reproducible benchmarks that couple accuracy with size/latency/energy on real hardware—aimed at making AI-based security practical at scale for lightweight IoT deployments.

Fathers' perinatal mental health is a major public health issue, yet few interventions have been developed targeting this group. Fathers face many barriers in accessing perinatal mental health support, including stigma around caregiving and mental health, and thus require careful consideration when designing interventions. This study aimed to examine the feasibility, acceptability, and usability of a mobile app-based intervention for paternal perinatal depression, anxiety, and stress. Following a design science approach, five meta design principles and 15 specific principles were created to guide the intervention design, and a prototype app titled Rover was created. The prototype was evaluated by 43 fathers and 10 mental health clinicians. Participants in both groups rated the app highly for its functionality, clinical content, aesthetics, and digital therapeutic alliance. Qualitative feedback indicated that fathers held particularly favourable views regarding the mood tracking, mindfulness, and goal tracking features. Both groups expressed a preference for more support for the personalisation of content, including more dynamic interactions with the chatbot support feature. To our knowledge, this is the first app-based mental health intervention designed specifically for fathers, with study results providing guidance to the field on developing digital health initiatives for this population.

With the recent popularity of applying large-scale deep neural network-based models for natural language processing (NLP), attention to develop methods for text data augmentation is at its peak, since the limited size of training data tends to significantly affect the accuracy of these models. To this end, we propose a novel text data augmentation technique called Intelligent Masking with Optimal Substitutions Text Data Augmentation (IMOSA). IMOSA, developed for labelled sentences, can identify the most favourable sentences and locate the appropriate word combinations in a particular sentence to replace and generate synthetic sentences with a meaning closer to the original sentence, while also significantly increasing the diversity of the dataset. We demonstrate that the proposed technique notably improves the performance of classifiers based on attention-based transformer models through the extensive experiments for five different text classification tasks which are performed under the low data regime in a context-aware NLP setting. The analysis clearly shows that IMOSA effectively generates more sentences using favourable original examples and completely ignores undesirable examples. Furthermore, the experiments carried out confirm IMOSA’s ability to add diversity to the augmented dataset using multiple distinct masking patterns against the same original sentence, which remarkably adds variety to the training dataset. IMOSA consistently outperforms the two key masked language model-based text data augmentation techniques, and demonstrates a robust performance against the critical challenging NLP tasks.

An accurate reconstruction of Gene Regulatory Networks (GRNs) from time series gene expression data is crucial for discovering complex biological interactions. Among many different approaches for inferring GRNs, there are several methods which produce high false positive interactions, and are unstable, requiring fine tuning for many of their parameters. In this paper, we consider the GRN inference problem as a regression problem, and propose a simple ensemble regression-based feature selection model which is a combination of cross-validated Lasso and cross-validated Ridge algorithms for reconstructing GRNs. Due to the novelty of the proposed ensemble model, it is able to eliminate overfitting, multi co-linearity issues, and irrelevant genes within one computational approach. While observing the type of gene-gene regulatory interactions the regression model also identifies the direction of these interactions. A new coefficient of determination (R2)-based approach identifies the best model to fit the data among LassoCV and RidgeCV, and evaluates the model importance in term of gene-wise maximum in-degree which decides the maximum number of regulatory genes including self-regulations that can be selected from a given method. Then, an evaluated gene score-based majority voting technique aggregates the selected gene lists from each method. In our experiments, the performance of the proposed ensemble approach was evaluated using gene expression datasets from three small-scale real gene networks. Our proposed model outperformed other state-of-the-art methods, producing high true positives, reducing false positives, and obtaining high Structural Accuracy, while maintaining model stability and efficiency.

Reverse engineering of Gene Regulatory Networks (GRNs) from experimentally obtained high-throughput data is an active and promising area of research. Among several modelling techniques, the S-System model, a set of tightly coupled differential equations, mimics the complexities and dynamics of biochemical systems, and thus provides realistic GRN representation. While it offers mathematical flexibility and biological relevance, the high number of learning parameters can lead to a computational burden. In our earlier work, we addressed this issue by judicious use of prior knowledge. However, another major cause of computational load is the need for numerical integration of the differential equations for the estimation of S-system model parameters. In this paper, we propose a method to obtain initial model parameter values from the steady state of the system, thereby computing simpler and less complex algebraic equations compared to the regular differential equations of S-systems. These network parameters are input as prior knowledge for the optimization of the dynamic S-System using differential equations. The proposed framework includes a novel fitness evaluation for steady-state S-System models, a novel evolutionary parameter learning framework, and a technique to incorporate the candidate solutions in dynamic S-System modelling. Our proposed methodology reached optimal model parameter values quickly, requiring only one-third of the fitness function evaluations, compared to our previously reported DRNI (Dynamically regulated network initialization) method for S-System modelling.

Eating disorders (EDs) are characterised by abnormal eating habits and obsessive thought about food, weight, shape, and body image. EDs are experienced by a significant portion of our population. Social media is identified as a possible source of influence for EDs, and there is growing evidence of a large amount of ED-related discussions on the Web via social media platforms, such as Twitter. With this growing trend, automatic content analysis for EDs is becoming increasingly important. To date, there does not exist any comprehensive benchmark ED lexicon to identify ED-related conversations that would, in turn, facilitate these content analysis tasks. In this paper, we propose a novel method for generating a lexicon base for ED language, calledEDBase. The method starts with collecting over 3.7 million ED-focused tweets. In order to semantically represent potential ED terminology in a vector space, an ED word embedding model (EDModel) is trained. Then we develop a novel multi-seeded hierarchical density-based algorithm with contrasting corpora for ED lexicon expansion. TheEDModelis queried by the proposed lexicon expansion algorithm to expand the seed terms to a comprehensive lexicon base. OurEDBaseconsists of a (further expandable) list of 3794 high-quality ED terms, quantified by an ED score, and linked to their parent terms. The proposed method significantly outperforms all existing alternative baseline methods and models by over 25% in terms of precision and 1500 in terms of true positives. This research is expected to be impactful in the health data science and healthcare community.

The inference of Gene Regulatory Networks (GRNs) from time series gene expression data is an effective approach for unveiling important underlying gene-gene relationships and dynamics. While various computational models exist for accurate inference of GRNs, many are computationally inefficient, and do not focus on simultaneous inference of both network topology and dynamics. In this paper, we introduce a simple, Boolean network model-based solution for efficient inference of GRNs. First, the microarray expression data are discretized using the average gene expression value as a threshold. This step permits an experimental approach of defining the maximum indegree of a network. Next, regulatory genes, including the self-regulations for each target gene, are inferred using estimated multivariate mutual information-based Min-Redundancy Max-Relevance Criterion, and further accurate inference is performed by a swapping operation. Subsequently, we introduce a new method, combining Boolean network regulation modelling and Pearson correlation coefficient to identify the interaction types (inhibition or activation) of the regulatory genes. This method is utilized for the efficient determination of the optimal regulatory rule, consisting AND, OR, and NOT operators, by defining the accurate application of the NOT operation in conjunction and disjunction Boolean functions. The proposed approach is evaluated using two real gene expression datasets for an Escherichia coli gene regulatory network and a fission yeast cell cycle network. Although the Structural Accuracy is approximately the same as existing methods (MIBNI, REVEAL, Best-Fit, BIBN, and CST), the proposed method outperforms all these methods with respect to efficiency and Dynamic Accuracy.

Reverse engineering of gene regulatory networks through temporal gene expression data is an active area of research. Among the plethora of modelling techniques under investigation is the decoupled S-system model, which attempts to capture the non-linearity of biological systems in detail. For the model, number of parameters to be estimated are significantly high even when the network is of small or medium scale. Thus, the inference process poses a significant computational burden. In this paper, we propose: (1) a novel population initialization technique, Dynamically Regulated Prediction Initialization (DRPI), which utilises prior knowledge of biological gene expression data to create a feedback loop to produce dynamically regulated high-quality individuals for initial population; (2) an adaptive fitness function; and (3) a method for the maintenance of population diversity. The aim of this work is to reduce the computational complexity of the inference algorithm, to speed up the entire process of reverse engineering. The performance of the proposed algorithm was evaluated against a benchmark dataset and compared with other methods from earlier work. The experimental results show that we succeeded in achieving higher accuracy results in lesser fitness evaluations, considerably reducing the computational burden of the inference process.

Manual and high-quality annotation of social media data has enabled companies and researchers to develop improved implementations using natural language processing. However, human text-annotation is expensive and time-consuming. Crowd-sourcing platforms such as Amazon's Mechanical Turk (MTurk) can be leveraged for the creation of large training corpora for text classification tasks using social media data. Nevertheless, the quality of annotations can vary significantly, based on the interpretations and motivations of annotators completing the tasks. Further, the labelling cost of data through MTurk will increase if target messages are small and having a significant amount of noise (e.g. promotional messages on Twitter). In this work, we propose a new annotation framework to create high-quality human-annotated datasets for text classification from social media data. We present a zero-shot text classification based pre-annotation technique reducing the adverse effects arising due to the highly skewed distribution of data across target classes. The proposed framework significantly reduces the cost and time while maintaining the quality of the annotations. Being generic, it can be applied to annotating text data from any discipline. Our experiment with a Twitter data annotation using the proposed annotation framework shows a cost reduction of 80% with no compromise to quality.