New era of artificial intelligence and machine learning-based detection, diagnosis, and therapeutics in Parkinson's disease

Parkinson's disease (PD) is characterized by the loss of neuronal cells, which leads to synaptic dysfunction and cognitive defects. Despite the advancements in treatment strategies, the management of PD is still a challenging event. Early prediction and diagnosis of PD are of utmost importance for effective management of PD. In addition, the classification of patients with PD as compared to normal healthy individuals also imposes drawbacks in the early diagnosis of PD. To address these challenges, artificial intelligence (AI) and machine learning (ML) models have been implicated in the diagnosis, prediction, and treatment of PD. Recent times have also demonstrated the implication of AI and ML models in the classification of PD based on neuroimaging methods, speech recording, gait abnormalities, and others. Herein, we have briefly discussed the role of AI and ML in the diagnosis, treatment, and identification of novel biomarkers in the progression of PD. We have also highlighted the role of AI and ML in PD management through altered lipidomics and gut-brain axis. We briefly explain the role of early PD detection through AI and ML algorithms based on speech recordings, handwriting patterns, gait abnormalities, and neuroimaging techniques. Further, the review discuss the potential role of the metaverse, the Internet of Things, and electronic health records in the effective management of PD to improve the quality of life. Lastly, we also focused on the implementation of AI and ML-algorithms in neurosurgical process and drug discovery.

Literature review of digital twin in healthcare

This article aims to make a bibliometric literature review using systematic scientific mapping and content analysis of digital twins in healthcare to know the evolution, domain, keywords, content type, and kind and purpose of digital twin's implementation in healthcare, so a consolidation and future improvement of existing knowledge can be made and gaps for new studies can be identified. The increase in publications of digital twins in healthcare is quite recent and it is still concentrated in the domain of technology sources. The subject is majorly concentrated in patient's digital twin group and in precision medicine and aspects, issues and/or policies subgroups, although the publications keywords mirror it only at the group side. Digital twins in healthcare are probably stepping out of the infancy phase. On the other hand, digital twins in hospital group and the device and facilities management subgroups are more mature with all knowledge gathered from the manufacturing sector. There is an absence of some publication's types in general, device and care subgroup and no whole body or hospital digital twin was reported. Based on the presented arguments, guidelines for future research were presented: advance in the creation of general frameworks, in subgroups not as much explored, and in groups and subgroups already explored, but that need more advancement to achieve the main goals of a whole human or hospital digital twin with the main issues resolved.

Exploring microbial diversity in hot springs of Surajkund, India through 16S rRNA analysis and thermozyme characterization from endogenous isolates

Hot springs are a valuable source of biologically significant chemicals due to their high microbial diversity. To investigate the possibilities for industrial uses of these bacteria, researchers collected water and sediment samples from variety of hot springs. Our investigation employed both culture-dependent and culture-independent techniques, including 16S-based marker gene analysis of the microbiota from the hot springs of Surajkund, Jharkhand. In addition, we cultivated thermophilic isolates and screened for their ability to produce amylase, xylanase, and cellulase. After the optimized production of amylase the enzyme was partially purified and characterized using UPLC, DLS-ZP, and TGA. The retention time for the amylase was observed to be around 0.5 min. We confirmed the stability of the amylase at higher temperatures through observation of a steady thermo gravimetric profile at 400 °C. One of the thermophilic isolates obtained from the kund, demonstrated the potential to degrade lignocellulosic agricultural waste.

CADUCEO: A Platform to Support Federated Healthcare Facilities through Artificial Intelligence

Data-driven algorithms have proven to be effective for a variety of medical tasks, including disease categorization and prediction, personalized medicine design, and imaging diagnostics. Although their performance is frequently on par with that of clinicians, their widespread use is constrained by a number of obstacles, including the requirement for high-quality data that are typical of the population, the difficulty of explaining how they operate, and ethical and regulatory concerns. The use of data augmentation and synthetic data generation methodologies, such as federated learning and explainable artificial intelligence ones, could provide a viable solution to the current issues, facilitating the widespread application of artificial intelligence algorithms in the clinical application domain and reducing the time needed for prevention, diagnosis, and prognosis by up to 70%. To this end, a novel AI-based functional framework is conceived and presented in this paper.

Uncovering Protein Networks in Cardiovascular Proteomics

Biological networks have been widely used in many different diseases to identify potential biomarkers and design drug targets. In the present review, we describe the main computational techniques for reconstructing and analyzing different types of protein networks and summarize the previous applications of such techniques in cardiovascular diseases. Existing tools are critically compared, discussing when each method is preferred such as the use of co-expression networks for functional annotation of protein clusters and the use of directed networks for inferring regulatory associations. Finally, we are presenting examples of reconstructing protein networks of different types (regulatory, co-expression, and protein-protein interaction networks). We demonstrate the necessity to reconstruct networks separately for each cardiovascular tissue type and disease entity and provide illustrative examples of the importance of taking into consideration relevant post-translational modifications. Finally, we demonstrate and discuss how the findings of protein networks could be interpreted using single-cell RNA-sequencing data.

Comparing sustainable product hashtags: Insights from a historical twitter dataset

This data article describes the process of data collection and analysis of Twitter conversations about sustainable products. The dataset contains the IDs of tweets tagged with the hashtags #sustainableproducts, #ecoproducts, #ecofriendlyproducts, and #greenproducts. The time period spans 10 years and includes a total of over 140 thousand tweets from around the world. The article describes the process of obtaining the data using Twarc and the Twitter developer's academic researcher API and describes the preprocessing techniques used to identify keywords, hashtags, topics, and sentiments expressed in the conversations. The analysis identifies key attributes of each sustainable product category as well as commonalities and differences within and across categories. The data have the potential to be reused in future research related to sustainable consumption and production, including further analysis of the sentiments and attitudes expressed in the Twitter conversations and comparison with other social media platforms or survey data. In addition, the data can serve as a basis for marketing strategies and product design by enterprises or organizations seeking to promote sustainable products.

Innovative Hybrid Approach for Masked Face Recognition Using Pretrained Mask Detection and Segmentation, Robust PCA, and KNN Classifier

Face masks are widely used in various industries and jobs, such as healthcare, food service, construction, manufacturing, retail, hospitality, transportation, education, and public safety. Masked face recognition is essential to accurately identify and authenticate individuals wearing masks. Masked face recognition has emerged as a vital technology to address this problem and enable accurate identification and authentication in masked scenarios. In this paper, we propose a novel method that utilizes a combination of deep-learning-based mask detection, landmark and oval face detection, and robust principal component analysis (RPCA) for masked face recognition. Specifically, we use pretrained ssd-MobileNetV2 for detecting the presence and location of masks on a face and employ landmark and oval face detection to identify key facial features. The proposed method also utilizes RPCA to separate occluded and non-occluded components of an image, making it more reliable in identifying faces with masks. To optimize the performance of our proposed method, we use particle swarm optimization (PSO) to optimize both the KNN features and the number of k for KNN. Experimental results demonstrate that our proposed method outperforms existing methods in terms of accuracy and robustness to occlusion. Our proposed method achieves a recognition rate of 97%, which is significantly higher than the state-of-the-art methods. Our proposed method represents a significant improvement over existing methods for masked face recognition, providing high accuracy and robustness to occlusion.

Ten quick tips for bioinformatics analyses using an Apache Spark distributed computing environment

Some scientific studies involve huge amounts of bioinformatics data that cannot be analyzed on personal computers usually employed by researchers for day-to-day activities but rather necessitate effective computational infrastructures that can work in a distributed way. For this purpose, distributed computing systems have become useful tools to analyze large amounts of bioinformatics data and to generate relevant results on virtual environments, where software can be executed for hours or even days without affecting the personal computer or laptop of a researcher. Even if distributed computing resources have become pivotal in multiple bioinformatics laboratories, often researchers and students use them in the wrong ways, making mistakes that can cause the distributed computers to underperform or that can even generate wrong outcomes. In this context, we present here ten quick tips for the usage of Apache Spark distributed computing systems for bioinformatics analyses: ten simple guidelines that, if taken into account, can help users avoid common mistakes and can help them run their bioinformatics analyses smoothly. Even if we designed our recommendations for beginners and students, they should be followed by experts too. We think our quick tips can help anyone make use of Apache Spark distributed computing systems more efficiently and ultimately help generate better, more reliable scientific results.

Gender prediction based on University students' complex thinking competency: An analysis from machine learning approaches

This article aims to study machine learning models to determine their performance in classifying students by gender based on their perception of complex thinking competency. Data were collected from a convenience sample of 605 students from a private university in Mexico with the eComplexity instrument. In this study, we consider the following data analyses: 1) predict students' gender based on their perception of complex thinking competency and sub-competencies from a 25 items questionnaire, 2) analyze models' performance during training and testing stages, and 3) study the models' prediction bias through a confusion matrix analysis. Our results confirm the hypothesis that the four machine learning models (Random Forest, Support Vector Machines, Multi-layer Perception, and One-Dimensional Convolutional Neural Network) can find sufficient differences in the eComplexity data to classify correctly up to 96.94% and 82.14% of the students' gender in the training and testing stage, respectively. The confusion matrix analysis revealed partiality in gender prediction among all machine learning models, even though we have applied an oversampling method to reduce the imbalance dataset. It showed that the most frequent error was to predict Male students as Female class. This paper provides empirical support for analyzing perception data through machine learning models in survey research. This work proposed a novel educational practice based on developing complex thinking competency and machine learning models to facilitate educational itineraries adapted to the training needs of each group to reduce social gaps existing due to gender.

The Rab5 effector FERRY links early endosomes with mRNA localization

Localized translation is vital to polarized cells and requires precise and robust distribution of different mRNAs and ribosomes across the cell. However, the underlying molecular mechanisms are poorly understood and important players are lacking. Here, we discovered a Rab5 effector, the five-subunit endosomal Rab5 and RNA/ribosome intermediary (FERRY) complex, that recruits mRNAs and ribosomes to early endosomes through direct mRNA-interaction. FERRY displays preferential binding to certain groups of transcripts, including mRNAs encoding mitochondrial proteins. Deletion of FERRY subunits reduces the endosomal localization of transcripts in cells and has a significant impact on mRNA levels. Clinical studies show that genetic disruption of FERRY causes severe brain damage. We found that, in neurons, FERRY co-localizes with mRNA on early endosomes, and mRNA loaded FERRY-positive endosomes are in close proximity of mitochondria. FERRY thus transforms endosomes into mRNA carriers and plays a key role in regulating mRNA distribution and transport.

Unmasking the potential: evaluating image inpainting techniques for masked face reconstruction

The performance of most Face Recognizers tends to degrade when dealing with masked faces, making face recognition challenging. Image inpainting, a technique traditionally used for restoring old or damaged images, removing objects, or retouching photos, could potentially aid in reconstructing masked faces. In this paper, we compared three state-of-the-art image inpainting models-PatchMatch, a traditional algorithm, and two deep learning GAN-based models, Edge Connect and Free form image inpainting-to assess their performance in regenerating masked faces. The evaluation was conducted using own created synthetic datasets MaskedFace-CelebA and MaskedFace-CelebA-HQ, along with a synthetic masked dataset created for paired comparisons of masked images with ground truth for face verification. The computed results for Image Quality Assessment (IQA) between ground truth and reconstructed facial images indicated that the Gated Convolution model performed better than the other two models. To further validate the results, the reconstructed and ground truth images were also subject to VGG16 classifier, a widely used benchmark model for image recognition. The classifier outcomes supported the quantitative and qualitative assessment based on IQA.

Inter-Device Reliability of a Three-Dimensional Markerless Motion Capture System Quantifying Elementary Movement Patterns in Humans

With advancements in technology able to quantify wide-ranging features of human movement, the aim of the present study was to investigate the inter-device technological reliability of a three-dimensional markerless motion capture system (3D-MCS), quantifying different movement tasks. A total of 20 healthy individuals performed a test battery consisting of 29 different movements, from which 214 different metrics were derived. Two 3D-MCS located in close proximity were utilized to quantify movement characteristics. Independent sample t-tests with selected reliability statistics (i.e., intraclass correlation coefficient (ICC), effect sizes, and mean absolute differences) were used to evaluate the agreement between the two systems. The study results suggested that 95.7% of all metrics analyzed revealed negligible or small between-device effect sizes. Further, 91.6% of all metrics analyzed showed moderate or better agreement when looking at the ICC values, while 32.2% of all metrics showed excellent agreement. For metrics measuring joint angles (198 metrics), the mean difference between systems was 2.9 degrees, while for metrics investigating distance measures (16 metrics; e.g., center of mass depth), the mean difference between systems was 0.62 cm. Caution is advised when trying to generalize the study findings beyond the specific technology and software used in this investigation. Given the technological reliability reported in this study, as well as the logistical and time-related limitations associated with marker-based motion capture systems, it may be suggested that 3D-MCS present practitioners with an opportunity to reliably and efficiently measure the movement characteristics of patients and athletes. This has implications for monitoring the health/performance of a broad range of populations.

Data Preprocessing and Augmentation Improved Visual Field Prediction of Recurrent Neural Network with Multi-Central Datasets

INTRODUCTION

The purpose of this study was to determine whether data preprocessing and augmentation could improve visual field (VF) prediction of recurrent neural network (RNN) with multi-central datasets.

METHODS

This retrospective study collected data from five glaucoma services between June 2004 and January 2021. From an initial dataset of 331,691 VFs, we considered reliable VF tests with fixed intervals. Since the VF monitoring interval is very variable, we applied data augmentation using multiple sets of data for patients with more than eight VFs. We obtained 5,430 VFs from 463 patients and 13,747 VFs from 1,076 patients by setting the fixed test interval to 365 ± 60 days (D = 365) and 180 ± 60 days (D = 180), respectively. Five consecutive VFs were provided to the constructed RNN as input and the 6th VF was compared with the output of the RNN. The performance of the periodic RNN (D = 365) was compared to that of an aperiodic RNN. The performance of the RNN with 6 long- and short-term memory (LSTM) cells (D = 180) was compared with that of the RNN with 5-LSTM cells. To compare the prediction performance, the root mean square error (RMSE) and mean absolute error (MAE) of the total deviation value (TDV) were calculated as accuracy metrics.

RESULTS

The performance of the periodic model (D = 365) improved significantly over aperiodic model. Overall prediction error (MAE) was 2.56 ± 0.46 dB versus 3.26 ± 0.41 dB (periodic vs. aperiodic) (p < 0.001). A higher perimetric frequency was better for predicting future VF. The overall prediction error (RMSE) was 3.15 ± 2.29 dB versus 3.42 ± 2.25 dB (D = 180 vs. D = 365). Increasing the number of input VFs improved the performance of VF prediction in D = 180 periodic model (3.15 ± 2.29 dB vs. 3.18 ± 2.34 dB, p < 0.001). The 6-LSTM in the D = 180 periodic model was more robust to worsening of VF reliability and disease severity. The prediction accuracy worsened as the false-negative rate increased and the mean deviation decreased.

CONCLUSION

Data preprocessing with augmentation improved the VF prediction of the RNN model using multi-center datasets. The periodic RNN model predicted the future VF significantly better than the aperiodic RNN model.

Capturing the dynamics of microbial interactions through individual-specific networks

Longitudinal analysis of multivariate individual-specific microbiome profiles over time or across conditions remains dauntin. Most statistical tools and methods that are available to study microbiomes are based on cross-sectional data. Over the past few years, several attempts have been made to model the dynamics of bacterial species over time or across conditions. However, the field needs novel views on handling microbial interactions in temporal analyses. This study proposes a novel data analysis framework, MNDA, that combines representation learning and individual-specific microbial co-occurrence networks to uncover taxon neighborhood dynamics. As a use case, we consider a cohort of newborns with microbiomes available at 6 and 9 months after birth, and extraneous data available on the mode of delivery and diet changes between the considered time points. Our results show that prediction models for these extraneous outcomes based on an MNDA measure of local neighborhood dynamics for each taxon outperform traditional prediction models solely based on individual-specific microbial abundances. Furthermore, our results show that unsupervised similarity analysis of newborns in the study, again using the notion of a taxon's dynamic neighborhood derived from time-matched individual-specific microbial networks, can reveal different subpopulations of individuals, compared to standard microbiome-based clustering, with potential relevance to clinical practice. This study highlights the complementarity of microbial interactions and abundances in downstream analyses and opens new avenues to personalized prediction or stratified medicine with temporal microbiome data.

Recovering fluorescence spectra hidden by scattering signal: In search of the best smoother

Interpolation of the scattering areas in fluorescence excitation-emission matrices is a useful preprocessing method in fluorescence spectroscopy and data modelling. Commonly used row-by-row interpolation using piecewise cubic Hermite interpolating polynomials smoother (PCHIP), however, frequently leads to artifacts because it does not make any use of the information in the other dimension. We have suggested the way of constructing the penalty matrices for Whittaker smoothing that removed one of the main sources of difference between the axis of multiparametric signal - the grid step size - thus making it possible to reduce the number of parameters to optimize. We have compared Whittaker smoother with various surface interpolation methods, including LOESS, Kriging, multilevel B-spline approximation, and PCHIP for the purpose of data preprocessing before PARAFAC modelling of fluorescence signal on a model dataset. The two leaders by signal reconstruction and reconstruction of PARAFAC loadings are LOESS and Whittaker smoothing; the latter is additionally shown to have fundamentally interpretable parameters, which are easier to optimise for the whole dataset. Moreover, Whittaker keeps the shape of the signal and is resistant to variations in data structure and noise level that is very important in numerous applications. We also tested smoothers performance for Åsmund Rinnan fluorescence dataset and the high performance of Whittaker was proved. We can recommend the Whittaker smoothing as a perfect tool for interpolation of scattering areas in florescence spectra of seawaters with low signal-to-noise ratio.

A Novel Feature Selection Strategy Based on Salp Swarm Algorithm for Plant Disease Detection

Deep learning has been widely used for plant disease recognition in smart agriculture and has proven to be a powerful tool for image classification and pattern recognition. However, it has limited interpretability for deep features. With the transfer of expert knowledge, handcrafted features provide a new way for personalized diagnosis of plant diseases. However, irrelevant and redundant features lead to high dimensionality. In this study, we proposed a swarm intelligence algorithm for feature selection [salp swarm algorithm for feature selection (SSAFS)] in image-based plant disease detection. SSAFS is employed to determine the ideal combination of handcrafted features to maximize classification success while minimizing the number of features. To verify the effectiveness of the developed SSAFS algorithm, we conducted experimental studies using SSAFS and 5 metaheuristic algorithms. Several evaluation metrics were used to evaluate and analyze the performance of these methods on 4 datasets from the UCI machine learning repository and 6 plant phenomics datasets from PlantVillage. Experimental results and statistical analyses validated the outstanding performance of SSAFS compared to existing state-of-the-art algorithms, confirming the superiority of SSAFS in exploring the feature space and identifying the most valuable features for diseased plant image classification. This computational tool will allow us to explore an optimal combination of handcrafted features to improve plant disease recognition accuracy and processing time.

Untargeted white-box adversarial attack to break into deep leaning based COVID-19 monitoring face mask detection system

The face mask detection system has been a valuable tool to combat COVID-19 by preventing its rapid transmission. This article demonstrated that the present deep learning-based face mask detection systems are vulnerable to adversarial attacks. We proposed a framework for a robust face mask detection system that is resistant to adversarial attacks. We first developed a face mask detection system by fine-tuning the MobileNetv2 model and training it on the custom-built dataset. The model performed exceptionally well, achieving 95.83% of accuracy on test data. Then, the model's performance is assessed using adversarial images calculated by the fast gradient sign method (FGSM). The FGSM attack reduced the model's classification accuracy from 95.83% to 14.53%, indicating that the adversarial attack on the proposed model severely damaged its performance. Finally, we illustrated that the proposed robust framework enhanced the model's resistance to adversarial attacks. Although there was a notable drop in the accuracy of the robust model on unseen clean data from 95.83% to 92.79%, the model performed exceptionally well, improving the accuracy from 14.53% to 92% on adversarial data. We expect our research to heighten awareness of adversarial attacks on COVID-19 monitoring systems and inspire others to protect healthcare systems from similar attacks.

Forecasting real exchange rate (REER) using artificial intelligence and time series models

Forecasting is an attractive topic in every field of study because no one knows the exact nature of the underlying phenomena, but it can be guessed using mathematical functions. As the world progresses towards technology and betterment, algorithms are updated to understand the nature of ongoing phenomena. Machine learning (ML) algorithms are an updated phenomenon used in every task aspect. Real exchange rate data is assumed to be one of the significant components of the business market, which plays a pivotal role in learning market trends. In this work, machine learning models, i.e., the Multi-layer perceptron model (MLP), Extreme learning machine (ELM) model and classical time series models are used, Autoregressive integrated moving average (ARIMA) and Exponential Smoothing (ES) model to model and predict the real exchange rate data set (REER). The data under consideration is from January 2019 to June 2022 and comprises 864 observations. This study split the data set into training and testing and applied all stated models. This study selects a model that meets the Key Performance Indicators (KPI) criteria. This model was selected as the best candidate model to predict the behaviour of the real exchange rate data set.

extendGAN+: Transferable Data Augmentation Framework Using WGAN-GP for Data-Driven Indoor Localisation Model

For indoor localisation, a challenge in data-driven localisation is to ensure sufficient data to train the prediction model to produce a good accuracy. However, for WiFi-based data collection, human effort is still required to capture a large amount of data as the representation Received Signal Strength (RSS) could easily be affected by obstacles and other factors. In this paper, we propose an extendGAN+ pipeline that leverages up-sampling with the Dirichlet distribution to improve location prediction accuracy with small sample sizes, applies transferred WGAN-GP for synthetic data generation, and ensures data quality with a filtering module. The results highlight the effectiveness of the proposed data augmentation method not only by localisation performance but also showcase the variety of RSS patterns it could produce. Benchmarking against the baseline methods such as fingerprint, random forest, and its base dataset with localisation models, extendGAN+ shows improvements of up to 23.47%, 25.35%, and 18.88% respectively. Furthermore, compared to existing GAN+ methods, it reduces training time by a factor of four due to transfer learning and improves performance by 10.13%.

A Practice-Distributed Thunder-Localization System with Crowd-Sourced Smart IoT Devices

Lightning localization is of great significance to weather forecasting, forest fire prevention, aviation, military, and other aspects. Traditional lightning localization requires the deployment of base stations and expensive measurement equipment. With the development of IoT technology and the continuous expansion of application scenarios, IoT devices can be interconnected through sensors and other technical means to ultimately achieve the goal of automatic intelligent computing. Therefore, this paper proposes a low-cost distributed thunder-localization system based on IoT smart devices, namely ThunderLoc. The main idea of ThunderLoc is to collect dual-microphone data from IoT smart devices, such as smartphones or smart speakers, through crowdsourcing, turning the localization problem into a search problem in Hamming space. We studied the dual microphones integrated with smartphones and used the sign of Time Difference Of Arrival (TDOA) as measurement information. Through a simple generalized cross-correlation method, the TDOA of thunderclaps on the same smartphone can be estimated. After quantifying the TDOA measurement from the smartphone node, thunder localization was performed by minimizing the Hamming distance between the binary sequence and the binary vector measured in a database. The ThunderLoc system was evaluated through extensive simulations and experiments (a testbed with 30 smartphone nodes). The extensive experimental results demonstrate that ThunderLoc outperforms the main existing schemes in terms of effectively locating position and good robustness.

The impact of physicochemical features of carbon electrodes on the capacitive performance of supercapacitors: a machine learning approach

Hybrid electric vehicles and portable electronic systems use supercapacitors for energy storage owing to their fast charging/discharging rates, long life cycle, and low maintenance. Specific capacitance is regarded as one of the most important performance-related characteristics of a supercapacitor's electrode. In the current study, Machine Learning (ML) algorithms were used to determine the impact of various physicochemical properties of carbon-based materials on the capacitive performance of electric double-layer capacitors. Published experimental datasets from 147 references (4899 data entries) were extracted and then used to train and test the ML models, to determine the relative importance of electrode material features on specific capacitance. These features include current density, pore volume, pore size, presence of defects, potential window, specific surface area, oxygen, and nitrogen content of the carbon-based electrode material. Additionally, categorical variables as the testing method, electrolyte, and carbon structure of the electrodes are considered as well. Among five applied regression models, an extreme gradient boosting model was found to best correlate those features with the capacitive performance, highlighting that the specific surface area, the presence of nitrogen doping, and the potential window are the most significant descriptors for the specific capacitance. These findings are summarized in a modular and open-source application for estimating the capacitance of supercapacitors given, as only inputs, the features of their carbon-based electrodes, the electrolyte and testing method. In perspective, this work introduces a new wide dataset of carbon electrodes for supercapacitors extracted from the experimental literature, also giving an instance of how electrochemical technology can benefit from ML models.

Survey on crop pest detection using deep learning and machine learning approaches

The most important elements in the realm of commercial food standards are effective pest management and control. Crop pests can make a huge impact on crop quality and productivity. It is critical to seek and develop new tools to diagnose the pest disease before it caused major crop loss. Crop abnormalities, pests, or dietetic deficiencies have usually been diagnosed by human experts. Anyhow, this was both costly and time-consuming. To resolve these issues, some approaches for crop pest detection have to be focused on. A clear overview of recent research in the area of crop pests and pathogens identification using techniques in Machine Learning Techniques like Random Forest (RF), Support Vector Machine (SVM), and Decision Tree (DT), Naive Bayes (NB), and also some Deep Learning methods like Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Deep convolutional neural network (DCNN), Deep Belief Network (DBN) was presented. The outlined strategy increases crop productivity while providing the highest level of crop protection. By offering the greatest amount of crop protection, the described strategy improves crop efficiency. This survey provides knowledge of some modern approaches for keeping an eye on agricultural fields for pest detection and contains a definition of plant pest detection to identify and categorise citrus plant pests, rice, and cotton as well as numerous ways of detecting them. These methods enable automatic monitoring of vast domains, therefore lowering human error and effort.

The use of climate information in humanitarian relief efforts: a literature review

Purpose

This paper aims to provide a systematic literature review of the state-of-the-art applications of climate information in humanitarian relief efforts, to further the knowledge of how climate science can be better integrated into the decision-making process of humanitarian supply chains.

Design/methodology/approach

A systematic literature review was conducted using a combination of key search terms developed from both climate science and humanitarian logistics literature. Articles from four major databases were retrieved, reduced and analyzed.

Findings

The study illustrates the status of application of climate information in humanitarian work, and identifies usability, collaboration and coordination as three key themes.

Originality/value

By delivering an overview of the current applications and challenges of climate information, this literature review proposes a three-phase conceptual framework.

MultiCens: Multilayer network centrality measures to uncover molecular mediators of tissue-tissue communication

With the evolution of multicellularity, communication among cells in different tissues and organs became pivotal to life. Molecular basis of such communication has long been studied, but genome-wide screens for genes and other biomolecules mediating tissue-tissue signaling are lacking. To systematically identify inter-tissue mediators, we present a novel computational approach MultiCens (Multilayer/Multi-tissue network Centrality measures). Unlike single-layer network methods, MultiCens can distinguish within- vs. across-layer connectivity to quantify the "influence" of any gene in a tissue on a query set of genes of interest in another tissue. MultiCens enjoys theoretical guarantees on convergence and decomposability, and performs well on synthetic benchmarks. On human multi-tissue datasets, MultiCens predicts known and novel genes linked to hormones. MultiCens further reveals shifts in gene network architecture among four brain regions in Alzheimer's disease. MultiCens-prioritized hypotheses from these two diverse applications, and potential future ones like "Multi-tissue-expanded Gene Ontology" analysis, can enable whole-body yet molecular-level systems investigations in humans.

Network modeling helps to tackle the complexity of drug-disease systems

From the (patho)physiological point of view, diseases can be considered as emergent properties of living systems stemming from the complexity of these systems. Complex systems display some typical features, including the presence of emergent behavior and the organization in successive hierarchic levels. Drug treatments increase this complexity scenario, and from some years the use of network models has been introduced to describe drug-disease systems and to make predictions about them with regard to several aspects related to drug discovery. Here, we review some recent examples thereof with the aim to illustrate how network science tools can be very effective in addressing both tasks. We will examine the use of bipartite networks that lead to the important concept of "disease module", as well as the introduction of more articulated models, like multi-scale and multiplex networks, able to describe disease systems at increasing levels of organization. Examples of predictive models will then be discussed, considering both those that exploit approaches purely based on graph theory and those that integrate machine learning methods. A short account of both kinds of methodological applications will be provided. Finally, the point will be made on the present situation of modeling complex drug-disease systems highlighting some open issues. This article is categorized under: Neurological Diseases > Computational Models Infectious Diseases > Computational Models Cardiovascular Diseases > Computational Models.

Deep learning for face mask detection: a survey

The Coronavirus Disease (Covid-19) was declared as a pandemic by WHO (World Health Organization) on 11 March 2020, and it is still currently going on, thereby impacting tremendously the whole world. As of September 2021, more than 220 million cases and 4.56 million deaths have been confirmed, which is a vast number and a significant threat to humanity. Although, As of 6 September 2021, a total of 5,352,927,296 vaccine doses have been administered, still many people worldwide are not fully vaccinated yet. As stated by WHO, "Masks" should be used as one of the measures to restrain the transmission of this virus. So, to reduce the infection, one has to cover their face, and to detect whether a person's face is covered with a mask or not, a "Face mask detection system" is needed. Face Mask Detection falls under the category of "Object Detection," which is one of the sub-domains of Computer Vision and Image Processing. Object Detection consists of both "Image Classification" and "Image Localization". Deep learning is a subset of Machine learning which, in turn, is a subset of Artificial intelligence that is widely being used to detect face masks; even some people are using hybrid approaches to make the most use of it and to build an efficient "Face mask detection system". In this paper, the main aim is to review all the research that has been done till now on this topic, various datasets and Techniques used, and their performances followed by limitations and improvements. As a result, the purpose of this study is to give a broader perspective to a researcher to identify patterns and trends in Face mask detection (Object Detection) within the framework of covid-19.

A new Apache Spark-based framework for big data streaming forecasting in IoT networks

Analyzing time-dependent data acquired in a continuous flow is a major challenge for various fields, such as big data and machine learning. Being able to analyze a large volume of data from various sources, such as sensors, networks, and the internet, is essential for improving the efficiency of our society's production processes. Additionally, this vast amount of data is collected dynamically in a continuous stream. The goal of this research is to provide a comprehensive framework for forecasting big data streams from Internet of Things networks and serve as a guide for designing and deploying other third-party solutions. Hence, a new framework for time series forecasting in a big data streaming scenario, using data collected from Internet of Things networks, is presented. This framework comprises of five main modules: Internet of Things network design and deployment, big data streaming architecture, stream data modeling method, big data forecasting method, and a comprehensive real-world application scenario, consisting of a physical Internet of Things network feeding the big data streaming architecture, being the linear regression the algorithm used for illustrative purposes. Comparison with other frameworks reveals that this is the first framework that incorporates and integrates all the aforementioned modules.

Feed-Forward Deep Neural Network (FFDNN)-Based Deep Features for Static Malware Detection

The portable executable header (PEH) information is commonly used as a feature for malware detection systems to train and validate machine learning (ML) or deep learning (DL) classifiers. We propose to extract the deep features from the PEH information through hidden layers of a feed-forward deep neural network (FFDNN). The extraction of deep features of hidden layers represents the dataset with a better generalization for malware detection. While feeding the deep feature of one hidden layer to the succeeding layer, the Gaussian error linear unit (GeLU) activation function is applied. The FFDNN is trained with the GeLU activation function using the deep features of individual layers as well as concatenated deep features of all hidden layers. Similarly, the ML classifiers are also trained and validated in with individual layer deep features and concatenated features. Three highly effective ML classifiers, random forest (RF), support vector machine (SVM), and k-nearest neighbour (k-NN) have been investigated. The performance of the proposed model is demonstrated using a statically significant large dataset. The obtained results are interesting and encouraging in terms of classification accuracy. The classification accuracy reaches 99.15% with the internal discriminative deep feature for the proposed FFDNN-ML classifier with the GeLU activation function.

Sputum smears quality inspection using an ensemble feature extraction approach

The diagnosis of tuberculosis (TB) is extremely important. Sputum smear microscopy is thought to be the best method available in terms of accessibility and ease of use in resource-constrained countries. In this paper, research was conducted to evaluate the effectiveness of tuberculosis diagnosis by examining, among other things, the underlying causes of sputum smear quality for Ethiopian states such as Tigray, Amahira, and Oromia. However, because it is done manually, it has its limitations. This study proposes a model for sputum smear quality inspection using an ensemble feature extraction approach. The dataset used was recorded and labeled by experts in a regional lab in Bahir Dar, near Felege Hiwot Hospital after being collected from Gabi Hospital, Felege Hiwot Hospital, Adit Clinic and Gondar Hospital, as well as Kidanemihret Clinic in Gondar. We used a controlled environment to reduce environmental influences and eliminate variation. All the data was collected using a smartphone (the standard 15) with a jpg file extension and a pixel resolution of 1,728 × 3,840. Prior to feature extraction, bicubic resizing, and ROI extraction using thresholding was performed. In addition, sequential Gaussian and Gabor filters were used for noise reduction, augmentation, and CLAHE was used for enhancement. For feature extraction, GLCM from the gray label and CNN from the color image were both chosen. Ultimately, when CNN, SVM, and KNN classifiers were used to test both CNN and GLCM features, KNN outperformed them all with scores of 87, 93, and 94% for GLCM, CNN, and a hybrid of CNN and GLCM, respectively. CNN with GLCM outperformed other methods by 0.7 and 0.1% for GLCM and CNN feature extractors using the same classifier, respectively. In addition, the KNN classifier with the combination of CNN and GLCM as feature extractors performed better than existing methods by 1.48%.

Mask Usage Recognition using Vision Transformer with Transfer Learning and Data Augmentation

The COVID-19 pandemic has disrupted various levels of society. The use of masks is essential in preventing the spread of COVID-19 by identifying an image of a person using a mask. Although only 23.1% of people use masks correctly, Artificial Neural Networks (ANN) can help classify the use of good masks to help slow the spread of the Covid-19 virus. However, it requires a large dataset to train an ANN that can classify the use of masks correctly. MaskedFace-Net is a suitable dataset consisting of 137016 digital images with 4 class labels, namely Mask, Mask Chin, Mask Mouth Chin, and Mask Nose Mouth. Mask classification training utilizes Vision Transformers (ViT) architecture with transfer learning method using pre-trained weights on ImageNet-21k, with random augmentation. In addition, the hyper-parameters of training of 20 epochs, an Stochastic Gradient Descent (SGD) optimizer with a learning rate of 0.03, a batch size of 64, a Gaussian Cumulative Distribution (GeLU) activation function, and a Cross-Entropy loss function are used to be applied on the training of three architectures of ViT, namely Base-16, Large-16, and Huge-14. Furthermore, comparisons of with and without augmentation and transfer learning are conducted. This study found that the best classification is transfer learning and augmentation using ViT Huge-14. Using this method on MaskedFace-Net dataset, the research reaches an accuracy of 0.9601 on training data, 0.9412 on validation data, and 0.9534 on test data. This research shows that training the ViT model with data augmentation and transfer learning improves classification of the mask usage, even better than convolutional-based Residual Network (ResNet).

Enhancing cybersecurity situation awareness through visualization: A USB data exfiltration case study

Employees who have legitimate access to an organization's data may occasionally put sensitive corporate data at risk, either carelessly or maliciously. Ideally, potential breaches should be detected as soon as they occur, but in practice there may be delays, because human analysts are not able to recognize data exfiltration behaviors quickly enough with the tools available to them. Visualization may improve cybersecurity situation awareness. In this paper, we present a dashboard application for investigating file activity, as a way to improve situation awareness. We developed this dashboard for a wide range of stakeholders within a large financial services company. Cybersecurity experts/analysts, data owners, team leaders/managers, high level administrators, and other investigators all provided input to its design. The use of a co-design approach helped to create trust between users and the new visualization tools, which were built to be compatible with existing work processes. We discuss the user-centered design process that informed the development of the dashboard, and the functionality of its three inter-operable monitoring dashboards. In this case three dashboards were developed covering high-level overview, file volume/type comparison, and individual activity, but the appropriate number and type of dashboards to use will likely vary according to the nature of the detection task). We also present two use cases with usability results and preliminary usage data. The results presented examined the amount of use that the dashboards received as well as measures obtained using the Technology Acceptance Model (TAM). We also report user comments about the dashboards and how to improve them.

Deep Learning Approach for SDN-Enabled Intrusion Detection System in IoT Networks

Owing to the prevalence of the Internet of things (IoT) devices connected to the Internet, the number of IoT-based attacks has been growing yearly. The existing solutions may not effectively mitigate IoT attacks. In particular, the advanced network-based attack detection solutions using traditional Intrusion detection systems are challenging when the network environment supports traditional as well as IoT protocols and uses a centralized network architecture such as a software defined network (SDN). In this paper, we propose a long short-term memory (LSTM) based approach to detect network attacks using SDN supported intrusion detection system in IoT networks. We present an extensive performance evaluation of the machine learning (ML) and deep learning (DL) model in two SDNIoT-focused datasets. We also propose an LSTM-based architecture for the effective multiclass classification of network attacks in IoT networks. Our evaluation of the proposed model shows that our model effectively identifies the attacks and classifies the attack types with an accuracy of 0.971. In addition, various visualization methods are shown to understand the dataset’s characteristics and visualize the embedding features.

Predictors of the rate of cognitive decline in older adults using machine learning

BACKGROUND

The longitudinal rates of cognitive decline among aging populations are heterogeneous. Few studies have investigated the possibility of implementing prognostic models to predict cognitive changes with the combination of categorical and continuous data from multiple domains.

OBJECTIVE

Implement a multivariate robust model to predict longitudinal cognitive changes over 12 years among older adults and to identify the most significant predictors of cognitive changes using machine learning techniques.

METHOD

In total, data of 2733 participants aged 50-85 years from the English Longitudinal Study of Ageing are included. Two categories of cognitive changes were determined including minor cognitive decliners (2361 participants, 86.4%) and major cognitive decliners (372 participants, 13.6%) over 12 years from wave 2 (2004-2005) to wave 8 (2016-2017). Machine learning methods were used to implement the predictive models and to identify the predictors of cognitive decline using 43 baseline features from seven domains including sociodemographic, social engagement, health, physical functioning, psychological, health-related behaviors, and baseline cognitive tests.

RESULTS

The model predicted future major cognitive decliners from those with the minor cognitive decline with a relatively high performance. The overall AUC, sensitivity, and specificity of prediction were 72.84%, 78.23%, and 67.41%, respectively. Furthermore, the top 7 ranked features with an important role in predicting major vs minor cognitive decliners included age, employment status, socioeconomic status, self-rated memory changes, immediate word recall, the feeling of loneliness, and vigorous physical activity. In contrast, the five least important baseline features consisted of smoking, instrumental activities of daily living, eye disease, life satisfaction, and cardiovascular disease.

CONCLUSION

The present study indicated the possibility of identifying individuals at high risk of future major cognitive decline as well as potential risk/protective factors of cognitive decline among older adults. The findings could assist in improving the effective interventions to delay cognitive decline among aging populations.

Innovation signals: leveraging machine learning to separate noise from news

The early detection of and an adequate response to meaningful signals of change have a defining impact on the competitive vitality and the competitive advantage of companies. For this strategically important task, companies apply corporate foresight, aiming to enable superior company performance. With the growing dynamics of global markets, the amount of data to be analyzed for this purpose is constantly increasing. As a result, these analyses are often performed with an unreasonably high investment of financial and human resources, or are even not performed at all. To address this challenge, this paper presents a machine-learning-based approach to help companies identify early signals of change with a higher level of automation than before. For this, we combine a newly-proposed quantitative approach with the existing qualitative approaches by Cooper (stage-gate model) and by Rohrbeck (corporate foresight process). After a search field of interest has been defined, the related data is collected from web news sites, early signals are identified and selected automatically, and domain experts then assess these signals with respect to their relevance and novelty. Once it has been set up, the approach can be executed iteratively at regular time intervals in order to continuously scan for new signals of change. By means of three case studies supported by domain experts we demonstrate the effectiveness of our approach. After presenting our findings and discussing possible limitations of the approach, we suggest future research opportunities to further advance this field.

A comprehensive and systematic literature review on the big data management techniques in the internet of things

The Internet of Things (IoT) is a communication paradigm and a collection of heterogeneous interconnected devices. It produces large-scale distributed, and diverse data called big data. Big Data Management (BDM) in IoT is used for knowledge discovery and intelligent decision-making and is one of the most significant research challenges today. There are several mechanisms and technologies for BDM in IoT. This paper aims to study the important mechanisms in this area systematically. This paper studies articles published between 2016 and August 2022. Initially, 751 articles were identified, but a paper selection process reduced the number of articles to 110 significant studies. Four categories to study BDM mechanisms in IoT include BDM processes, BDM architectures/frameworks, quality attributes, and big data analytics types. Also, this paper represents a detailed comparison of the mechanisms in each category. Finally, the development challenges and open issues of BDM in IoT are discussed. As a result, predictive analysis and classification methods are used in many articles. On the other hand, some quality attributes such as confidentiality, accessibility, and sustainability are less considered. Also, none of the articles use key-value databases for data storage. This study can help researchers develop more effective BDM in IoT methods in a complex environment.

RRFMDS: Rapid Real-Time Face Mask Detection System for Effective COVID-19 Monitoring

The primary mode of COVID-19 transmission is through respiratory droplets that are produced when an infected person talks, coughs, or sneezes. To avoid the fast spread of the virus, the WHO has instructed people to use face masks in crowded and public areas. This paper proposes the rapid real-time face mask detection system or RRFMDS, an automated computer-aided system to detect a violation of a face mask in real-time video. In the proposed system, single-shot multi-box detector is utilized for face detection, while fine-tuned MobileNetV2 is used for face mask classification. The system is lightweight (low resource requirement) and can be merged with pre-installed CCTV cameras to detect face mask violation. The system is trained on a custom dataset which consists of 14,535 images, of which 5000 belong to incorrect masks, 4789 to with masks, and 4746 to without masks. The primary purpose of creating such a dataset was to develop a face mask detection system that can detect almost all types of face masks with different orientations. The system can detect all three classes (incorrect masks, with mask and without mask faces) with an average accuracy of 99.15% and 97.81%, respectively, on training and testing data. The system, on average, takes 0.14201142 s to process a single frame, including detecting the faces from the video, processing a frame and classification.

Systematic analysis of healthcare big data analytics for efficient care and disease diagnosing

Big data has revolutionized the world by providing tremendous opportunities for a variety of applications. It contains a gigantic amount of data, especially a plethora of data types that has been significantly useful in diverse research domains. In healthcare domain, the researchers use computational devices to extract enriched relevant information from this data and develop smart applications to solve real-life problems in a timely fashion. Electronic health (eHealth) and mobile health (mHealth) facilities alongwith the availability of new computational models have enabled the doctors and researchers to extract relevant information and visualize the healthcare big data in a new spectrum. Digital transformation of healthcare systems by using of information system, medical technology, handheld and smart wearable devices has posed many challenges to researchers and caretakers in the form of storage, minimizing treatment cost, and processing time (to extract enriched information, and minimize error rates to make optimum decisions). In this research work, the existing literature is analysed and assessed, to identify gaps that result in affecting the overall performance of the available healthcare applications. Also, it aims to suggest enhanced solutions to address these gaps. In this comprehensive systematic research work, the existing literature reported during 2011 to 2021, is thoroughly analysed for identifying the efforts made to facilitate the doctors and practitioners for diagnosing diseases using healthcare big data analytics. A set of rresearch questions are formulated to analyse the relevant articles for identifying the key features and optimum management solutions, and laterally use these analyses to achieve effective outcomes. The results of this systematic mapping conclude that despite of hard efforts made in the domains of healthcare big data analytics, the newer hybrid machine learning based systems and cloud computing-based models should be adapted to reduce treatment cost, simulation time and achieve improved quality of care. This systematic mapping will also result in enhancing the capabilities of doctors, practitioners, researchers, and policymakers to use this study as evidence for future research.

Research on Indoor 3D Positioning Algorithm Based on WiFi Fingerprint

Indoor 3D positioning is useful in multistory buildings, such as shopping malls, libraries, and airports. This study focuses on indoor 3D positioning using wireless access points (AP) in an environment without adding additional hardware facilities in large-scale complex places. The integration of a deep learning algorithm into indoor 3D positioning is studied, and a 3D dynamic positioning model based on temporal fingerprints is proposed. In contrast to the traditional positioning models with a single input, the proposed method uses a sliding time window to build a temporal fingerprint chip as the input of the positioning model to provide abundant information for positioning. Temporal information can be used to distinguish locations with similar fingerprint vectors and to improve the accuracy and robustness of positioning. Moreover, deep learning has been applied for the automatic extraction of spatiotemporal features. A temporal convolutional network (TCN) feature extractor is proposed in this paper, which adopts a causal convolution mechanism, dilated convolution mechanism, and residual connection mechanism and is not limited by the size of the convolution kernel. It is capable of learning hidden information and spatiotemporal relationships from the input features and the extracted spatiotemporal features are connected with a deep neural network (DNN) regressor to fit the complex nonlinear mapping relationship between the features and position coordinates to estimate the 3D position coordinates of the target. Finally, an open-source public dataset was used to verify the performance of the localization algorithm. Experimental results demonstrated the effectiveness of the proposed positioning model and a comparison between the proposed model and existing models proved that the proposed model can provide more accurate three-dimensional position coordinates.

konnect2prot: a web application to explore the protein properties in a functional protein-protein interaction network

MOTIVATION

The regulation of proteins governs the biological processes and functions and, therefore, the organisms' phenotype. So there is an unmet need for a systematic tool for identifying the proteins that play a crucial role in information processing in a protein-protein interaction (PPI) network. However, the current protein databases and web servers still lag behind to provide an end-to-end pipeline that can leverage the topological understanding of a context-specific PPI network to identify the influential spreaders. Addressing this, we developed a web application, 'konnect2prot' (k2p), which can generate context-specific directional PPI network from the input proteins and detect their biological and topological importance in the network.

RESULTS

We pooled together a large amount of ontological knowledge, parsed it down into a functional network, and gained insight into the molecular underpinnings of the disease development by creating a one-stop junction for PPI data. k2p contains both local and global information about a protein, such as protein class, disease mutations, ligands and PDB structure, enriched processes and pathways, multi-disease interactome and hubs and bottlenecks in the directional network. It also identifies spreaders in the network and maps them to disease hallmarks to determine whether they can affect the disease state or not.

AVAILABILITY AND IMPLEMENTATION

konnect2prot is freely accessible using the link https://konnect2prot.thsti.in. The code repository is https://github.com/samrat-lab/k2p_bioinfo-2022.

Big Data Clustering Using Chemical Reaction Optimization Technique: A Computational Symmetry Paradigm for Location-Aware Decision Support in Geospatial Query Processing

The emergence of geospatial big data has opened up new avenues for identifying urban environments. Although both geographic information systems (GIS) and expert systems (ES) have been useful in resolving geographical decision issues, they are not without their own shortcomings. The combination of GIS and ES has gained popularity due to the necessity of boosting the effectiveness of these tools in resolving very difficult spatial decision-making problems. The clustering method generates the functional effects necessary to apply spatial analysis techniques. In a symmetric clustering system, two or more nodes run applications and monitor each other simultaneously. This system is more efficient than an asymmetric system since it utilizes all available hardware and does not maintain a node in a hot standby state. However, it is still a major issue to figure out how to expand and speed up clustering algorithms without sacrificing efficiency. The work presented in this paper introduces an optimized hierarchical distributed k-medoid symmetric clustering algorithm for big data spatial query processing. To increase the k-medoid method’s efficiency and create more precise clusters, a hybrid approach combining the k-medoid and Chemical Reaction Optimization (CRO) techniques is presented. CRO is used in this approach to broaden the scope of the optimal medoid and improve clustering by obtaining more accurate data. The suggested paradigm solves the current technique’s issue of predicting the accurate clusters’ number. The suggested approach includes two phases: in the first phase, the local clusters are built using Apache Spark’s parallelism paradigm based on their portion of the whole dataset. In the second phase, the local clusters are merged to create condensed and reliable final clusters. The suggested approach condenses the data provided during aggregation and creates the ideal clusters’ number automatically based on the dataset’s structures. The suggested approach is robust and delivers high-quality results for spatial query analysis, as shown by experimental results. The proposed model reduces average query latency by 23%.

Demographic history and conservation genomics of caribou (Rangifer tarandus) in Québec

The loss of genetic diversity is a challenge many species are facing, with genomics being a potential tool to inform and prioritize decision-making. Most caribou (Rangifer tarandus) populations have experienced significant recent declines throughout Québec, Canada, and are considered of concern, threatened or endangered. Here, we calculated the ancestral and contemporary patterns of genomic diversity of five representative caribou populations and applied a comparative population genomics framework to assess the interplay between demographic events and genomic diversity. We first calculated a caribou specific mutation rate, μ, by extracting orthologous genes from related ungulates and estimating the rate of synonymous mutations. Whole genome re-sequencing was then completed on 67 caribou: from these data we calculated nucleotide diversity, θ π and estimated the coalescent or ancestral effective population size (N e), which ranged from 12,030 to 15,513. When compared to the census size, N C, the endangered Gaspésie Mountain caribou population had the highest ancestral N e:N C ratio which is consistent with recent work suggesting high ancestral N e:N C is of conservation concern. In contrast, values of contemporary N e, estimated from linkage-disequilibrium, ranged from 11 to 162, with Gaspésie having among the highest contemporary N e:N C ratio. Importantly, classic conservation genetics theory would predict this population to be of less concern based on this ratio. Interestingly, F varied only slightly between populations, and despite evidence of bottlenecks across the province, runs of homozygosity were not abundant in the genome. Tajima's D estimates mirrored the demographic models and current conservation status. Our study highlights how genomic patterns are nuanced and potentially misleading if viewed only through a contemporary lens; we argue a holistic conservation genomics view should integrate ancestral N e and Tajima's D into management decisions.

Can language models automate data wrangling?

The automation of data science and other data manipulation processes depend on the integration and formatting of ‘messy’ data. Data wrangling is an umbrella term for these tedious and time-consuming tasks. Tasks such as transforming dates, units or names expressed in different formats have been challenging for machine learning because (1) users expect to solve them with short cues or few examples, and (2) the problems depend heavily on domain knowledge. Interestingly, large language models today (1) can infer from very few examples or even a short clue in natural language, and (2) can integrate vast amounts of domain knowledge. It is then an important research question to analyse whether language models are a promising approach for data wrangling, especially as their capabilities continue growing. In this paper we apply different variants of the language model Generative Pre-trained Transformer (GPT) to five batteries covering a wide range of data wrangling problems. We compare the effect of prompts and few-shot regimes on their results and how they compare with specialised data wrangling systems and other tools. Our major finding is that they appear as a powerful tool for a wide range of data wrangling tasks. We provide some guidelines about how they can be integrated into data processing pipelines, provided the users can take advantage of their flexibility and the diversity of tasks to be addressed. However, reliability is still an important issue to overcome.

A farmer-centric agricultural decision support system for market dynamics in a volatile agricultural supply chain

Purpose

In a volatile agricultural postharvest market, producers require more personalized information about market dynamics for informed decisions on the marketed surplus. However, this adaptive strategy fails to benefit them if the selection of a computational price predictive model to disseminate information on the market outlook is not efficient, and the associated risk of perishability, and storage cost factor are not assumed against the seemingly favourable market behaviour. Consequently, the decision of whether to store or sell at the time of crop harvest is a perennial dilemma to solve. With the intent of addressing this challenge for agricultural producers, the study is focused on designing an agricultural decision support system (ADSS) to suggest a favourable marketing strategy to crop producers.

Design/methodology/approach

The present study is guided by an eclectic theoretical perspective from supply chain literature that included agency theory, transaction cost theory, organizational information processing theory and opportunity cost theory in revenue risk management. The paper models a structured iterative algorithmic framework that leverages the forecasting capacity of different time series and machine learning models, considering the effect of influencing factors on agricultural price movement for better forecasting predictability against market variability or dynamics. It also attempts to formulate an integrated risk management framework for effective sales planning decisions that factors in the associated costs of storage, rental and physical loss until the surplus is held for expected returns.

Findings

Empirical demonstration of the model was simulated on the dynamic markets of tomatoes, onions and potatoes in a north Indian region. The study results endorse that farmer-centric post-harvest information intelligence assists crop producers in the strategic sales planning of their produce, and also vigorously promotes that the effectiveness of decision making is contingent upon the selection of the best predictive model for every future market event.

Practical implications

As a policy implication, the proposed ADSS addresses the pressing need for a robust marketing support system for the socio-economic welfare of farming communities grappling with distress sales, and low remunerative returns.

Originality/value

Based on the extant literature studied, there is no such study that pays personalized attention to agricultural producers, enabling them to make a profitable sales decision against the volatile post-harvest market scenario. The present research is an attempt to fill that gap with the scope of addressing crop producer's ubiquitous dilemma of whether to sell or store at the time of harvesting. Besides, an eclectic and iterative style of predictive modelling has also a limited implication in the agricultural supply chain based on the literature; however, it is found to be a more efficient practice to function in a dynamic market outlook.