Data Collection Algorithm Research Articles

UAV-Assisted Sleep Scheduling Algorithm for Energy-Efficient Data Collection in Agricultural Internet of Things

The rapid development of the agricultural Internet of Things (IoT) is inseparable from the support of wireless sensor networks (WSNs) in recent years. To further facilitate the adaptation of WSNs to agricultural applications, reducing the energy consumption of sensor nodes in agricultural environments has become a crucial problem. Sleep scheduling in randomly deployed WSNs is an effective method to reduce the energy consumption of sensor nodes, which can extend the network lifetime while ensuring network coverage. However, most of the existing sleep scheduling algorithms require frequent information exchange (such as broadcasting to find neighboring nodes and collecting nodes’ residual energy), which will inevitably lead to massive energy consumption. To address this problem, in this article, an unmanned aerial vehicle (UAV)-assisted sleep scheduling algorithm (UAVSS) is proposed, which avoids excessive information exchange among nodes and ensures sufficient network coverage with the least number of sensor nodes. The UAV traverses all the sensor nodes along the shortest path to help information exchange and gather sensing data. Furthermore, the UAV path is planned by using the latest monarch butterfly optimization (MBO) algorithm. Simulation results indicate that the proposed UAVSS can prolong the network lifetime while ensuring the required area coverage.

Learning Action Translator for Meta Reinforcement Learning on Sparse-Reward Tasks

Meta reinforcement learning (meta-RL) aims to learn a policy solving a set of training tasks simultaneously and quickly adapting to new tasks. It requires massive amounts of data drawn from training tasks to infer the common structure shared among tasks. Without heavy reward engineering, the sparse rewards in long-horizon tasks exacerbate the problem of sample efficiency in meta-RL. Another challenge in meta-RL is the discrepancy of difficulty level among tasks, which might cause one easy task dominating learning of the shared policy and thus preclude policy adaptation to new tasks. This work introduces a novel objective function to learn an action translator among training tasks. We theoretically verify that the value of the transferred policy with the action translator can be close to the value of the source policy and our objective function (approximately) upper bounds the value difference. We propose to combine the action translator with context-based meta-RL algorithms for better data collection and moreefficient exploration during meta-training. Our approach em-pirically improves the sample efficiency and performance ofmeta-RL algorithms on sparse-reward tasks.

Balanced Multi-Channel Data Collection in Wireless Sensor Networks

Data collection is an essential task in Wireless Sensor Networks (WSNs). In data collection process, the sensor nodes transmit their readings to a common base station called Sink. To avoid a collision, it is necessary to use the appropriate scheduling algorithms for data transmission. On the other hand, multi-channel design is considered as a promising technique to reduce network interference and latency of data collection. This technique allows parallel transmissions on different frequency channels, thus time latency will be reduced. In this paper, we present a new scheduling method for multi-channel WSNs called Balanced Multi Channel Data Collection (Balanced MC-DC) Algorithm. The proposed protocol is based on using both Non-Overlapping Channels (NOC) and Partially Overlapping Channels (POC). It uses a new approach that optimizes the processes of tree construction, channel allocation, transmission scheduling and balancing simultaneously. Extensive simulations confirm the superiority of the proposed algorithm over the existing algorithms in wireless sensor networks.

Energy- and Delay-Efficient Algorithm for Large-Scale Data Collection in Mobile-Sink WSNs

This paper deals with the problem of large-scale data collection in wireless sensor networks equipped with multiple multi-interface mobile sinks. Employing mobile sinks for data collection leads to a trade-off between energy consumption and delay. We propose a new algorithm called “Data Gathering with Multi-interface Mobile Sink (DGM <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> S)” to reduce both energy consumption and delay in the network. DGM <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> S considers delays due to sink mobility and sink stopping for receiving the aggregated data from Cluster Heads (CHs). The proposed scheme partitions the network into smaller sub-areas, where in each sub-area, an appropriate trajectory is determined for the corresponding sink. In each partition, DGM <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> S uses an Energy- and Delay-efficient Polling point Selection (EDPS) strategy to determine data of which CHs should be collected at different points of the corresponding sink tour. EDPS considers path loss and shadowing phenomena, as well as the considerations for delay and energy consumption. As another advantage, DGM <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> S employs a scheduling mechanism to make the best use of the multi-interface technology in collecting data of CHs at each polling point. Extensive simulations are conducted to illustrate the advantages of the proposed DGM <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> S algorithm in terms of energy consumption, delay, network’s lifetime, and time complexity.

Optimum Data Transmission Allocation in Multimodal Communication of Underwater Sensor Networks

Multimodal communication (MC) plays a crucial role in the Underwater Sensor Networks (UWSNs). The appropriate resource allocation of MC can significantly optimize the UWSNs performance. This letter investigates the optimum data transmission allocation in the underwater MC. Considering the fundamental MC tradeoff, the MC energy efficiency model is established and formulated as a mixed integer nonlinear programming problem with the provable convex optimization properties. Then this letter designs two schemes for the various data amounts to optimize the optical-acoustic transmission ratio and an optimized data collection algorithm to save the MC energy. Simulation results validate the effectiveness of this letter.

Power‐efficient optimized clustering method with intelligent fog computing for wireless sensor networks

AbstractOne of the most essential characteristics that is taken into consideration while dealing with wireless sensor networks (WSNs) is to optimize energy consumption during the transmission of data packets. Routing algorithms must provide optimal solutions to reduce the amount of energy consumed during the process. In wireless sensor networks, the end devices layer comprises of clustering of sensor nodes using ant lion optimization (ALO) which is followed by fog layer consisting of k‐means clustering technique. When nature inspired optimization solution is implemented, the accuracy majorly depends upon complexity and number of parameters. The algorithm proposed majorly focuses on the bottom layer and not on the fog layer. The results of the proposed algorithm were proved to significantly better than the conventional algorithms used for the very same purpose. The proposed algorithm focuses on ALO only on the bottom layer and not on the fog layer. To evaluate the performance of the proposed algorithm, the results are compared with the findings of several traditional algorithms, such as energy‐efficient cross‐layer‐sensing clustering method, Distributed and Morphological Operation‐based Data Collection Algorithm, and trust‐based secure routing. Comparative results showed that the proposed algorithm presented significant improvement.

End-to-end data collection strategy using mobile sink in wireless sensor networks

Several data collection algorithms, which are based on the combination of using mobile sinks and multiple-hop forwarding, have been proposed to prolong the network lifetime of wireless sensor networks. However, most approaches treat the collection point selection and touring path planning as two independent problems, which leads to a sub-optimal solution for data collection. This article proposed an ant colony optimization based end-to-end data collection strategy to perform the collection point selection and the touring path planning simultaneously. The proposed algorithm first constructs a data-forwarding tree, and then heuristically selects collection points and plans a touring path at the same time. The performance evaluation shows that the end-to-end strategy can improve the network lifetime of wireless sensor network compared to other approaches, especially in the unbalanced distribution scenario of sensors. The end-to-end strategy is also capable of being integrated with other methods.

A Multi-Channel Interference Based Source Location Privacy Protection Scheme in Underwater Acoustic Sensor Networks

The Internet of Underwater Things (IoUT) gradually becomes the future direction of ocean. Many devices, which are called the sources, are deployed in the sea to fulfill scientific, civilian, and military needs. If the locations of these sources are leaked, it will cause a serious economic and military crisis. Thus, in this context, the source location privacy protection becomes critical. In this paper, we focus on the source location privacy protection of underwater acoustic sensor networks (UASNs), where a multi-channel interference based source location privacy protection scheme (MCISLP) is proposed. The MCISLP mainly contains four steps. First, to enhance the sense of autonomous underwater vehicles (AUV) for nodes, a noise reduction method is proposed. Then, based on the previous step, an AUV movement prediction method for nodes is presented to ensure the success of packet transmission. After that, for the information exchange among nodes, given the interference of multiple underwater channels, this interference is then utilized to protect the source location privacy. Further, to reduce the channel interference on the source packet, the time slot control is added to weaken the influence. Last, the ant colony algorithm is adopted for the data collection of AUV. The simulation results show that the MCISLP can increase the network safety time by up to 57 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> (compared with data collection algorithms), and balance the energy consumption of nodes (for about 20 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> energy saving), with a minor compromise of energy consumption of AUV and delay.

Recent Studies Utilizing Artificial Intelligence Techniques for Solving Data Collection, Aggregation and Dissemination Challenges in Wireless Sensor Networks: A Review

The growing importance and widespread adoption of Wireless Sensor Network (WSN) technologies have helped the enhancement of smart environments in numerous sectors such as manufacturing, smart cities, transportation and Internet of Things by providing pervasive real-time applications. In this survey, we analyze the existing research trends with respect to Artificial Intelligence (AI) methods in WSN and the possible use of these methods for WSN enhancement. The main goal of data collection, aggregation and dissemination algorithms is to gather and aggregate data in an energy efficient manner so that network lifetime is enhanced. In this paper, we highlight data collection, aggregation and dissemination challenges in WSN and present a comprehensive discussion on the recent studies that utilized various AI methods to meet specific objectives of WSN, during the span of 2010 to 2021. We compare and contrast different algorithms on the basis of optimization criteria, simulation/real deployment, centralized/distributed kind, mobility and performance parameters. We conclude with possible future research directions. This would guide the reader towards an understanding of up-to-date applications of AI methods with respect to data collection, aggregation and dissemination challenges in WSN. Then, we provide a general evaluation and comparison of different AI methods used in WSNs, which will be a guide for the research community in identifying the mostly adapted methods and the benefits of using various AI methods for solving the challenges related to WSNs. Finally, we conclude the paper stating the open research issues and new possibilities for future studies.

Augmenting Social Science Research with Multimodal Data Collection: The EZ-MMLA Toolkit.

While the majority of social scientists still rely on traditional research instruments (e.g., surveys, self-reports, qualitative observations), multimodal sensing is becoming an emerging methodology for capturing human behaviors. Sensing technology has the potential to complement and enrich traditional measures by providing high frequency data on people’s behavior, cognition and affects. However, there is currently no easy-to-use toolkit for recording multimodal data streams. Existing methodologies rely on the use of physical sensors and custom-written code for accessing sensor data. In this paper, we present the EZ-MMLA toolkit. This toolkit was implemented as a website and provides easy access to multimodal data collection algorithms. One can collect a variety of data modalities: data on users’ attention (eye-tracking), physiological states (heart rate), body posture (skeletal data), gestures (from hand motion), emotions (from facial expressions and speech) and lower-level computer vision algorithms (e.g., fiducial/color tracking). This toolkit can run from any browser and does not require dedicated hardware or programming experience. We compare this toolkit with traditional methods and describe a case study where the EZ-MMLA toolkit was used by aspiring educational researchers in a classroom context. We conclude by discussing future work and other applications of this toolkit, potential limitations and implications.

Resonant Coupling Parameter Estimation with Superconducting Qubits

Today's quantum computers are comprised of tens of qubits interacting with each other and the environment in increasingly complex networks. In order to achieve the best possible performance when operating such systems, it is necessary to have accurate knowledge of all parameters in the quantum computer Hamiltonian. In this article, we demonstrate theoretically and experimentally a method to efficiently learn the parameters of resonant interactions for quantum computers consisting of frequency-tunable superconducting qubits. Such interactions include, for example, those to other qubits, resonators, two-level state defects, or other unwanted modes. Our method is based on a significantly improved swap spectroscopy calibration and consists of an offline data collection algorithm, followed by an online Bayesian learning algorithm. The purpose of the offline algorithm is to detect and roughly estimate resonant interactions from a state of zero knowledge. It produces a square-root reduction in the number of measurements. The online algorithm subsequently refines the estimate of the parameters to comparable accuracy as traditional swap spectroscopy calibration, but in constant time. We perform an experiment implementing our technique with a superconducting qubit. By combining both algorithms, we observe a reduction of the calibration time by one order of magnitude. We believe the method investigated will improve present medium-scale superconducting quantum computers and will also scale up to larger systems. Finally, the two algorithms presented here can be readily adopted by communities working on different physical implementations of quantum computing architectures.

New Results on Small and Dim Infrared Target Detection

Real-time small infrared (IR) target detection is critical to the performance of the situational awareness system in high-altitude aircraft. However, current IR target detection systems are generally hardware-unfriendly and have difficulty in achieving a robust performance in datasets with clouds occupying a large proportion of the image background. In this paper, we present new results by using an efficient method that extracts the candidate targets in the pre-processing stage and fuses the local scale, blob-based contrast map and gradient map in the detection stage. We also developed mid-wave infrared (MWIR) and long-wave infrared (LWIR) cameras for data collection experiments and algorithm evaluations. Experimental results using both publicly available datasets and image sequences acquired by our cameras clearly demonstrated that the proposed method achieves high detection accuracy with the mean AUC being at least 22.3% higher than comparable methods, and the computational cost beating the other methods by a large margin.

A greedy-model-based reinforcement learning algorithm for Beyond-5G cooperative data collection

Data collection is an essential part of Beyond-5G and Internet of Things applications. In urban area, heterogeneous access points such as Wi-Fi routers and base stations can meet the required communication coverage and bandwidth in data collection processes. However, in remote area, without communication infrastructures, it is hard to guarantee the communication quality of a large-scale data aggregation network. An existing approach is to use an unmanned aerial vehicle (UAV) to act as a mobile sink to perform data collection and increase the coverage of intelligent wireless sensing and communications. The efficiency and the reliability of such a UAV-assisted data collection system can be significantly enhanced with an intelligent cooperative strategy for the sensors deployed in the field to communicate with the UAV. Furthermore, an energy-efficient trajectory planning algorithm is crucial to address the physical limitations of the UAV in this application. In this paper, a data collection process is modeled as a Markov decision process (MDP). The paper begins with proposing two heuristic greedy algorithms, namely distance-greedy (DG) algorithm and rate-greedy (RG) algorithm, which are designed based on prior knowledge of the system and can guarantee the completion of the data collection process in a remote area without the help of fixed communication infrastructures. Based on the outcomes, a multi-agent greedy-model-based reinforcement learning (MG-RL) algorithm is proposed, which specifically designs the environmental state and the reward scheme, and introduces multiple UAVs with different parameters to explore environments in parallel to accelerate the training. In conclusion, the two proposed greedy algorithms have lower complexity of implementation while the proposed MG-RL algorithm yields practical UAVs’ flight trajectories and shortens the time for completing a data collection task.

An Efficient Mobile Data Gathering Method with Tree Clustering Algorithm in Wireless Sensor Networks Balanced and Unbalanced Topologies

Mobile Data Collector device (MDC) is adopted to reduce the energy consumption in Wireless Sensor Networks. This device travels the network in order to gather the collected data from sensor nodes. This paper presents a new Tree Clustering algorithm with Mobile Data Collector in Wireless Sensor Networks, which establishes the shortest travelling path passing throw a subset of Cluster Heads (CH). To select CHs, we adopt a competitive scheme, and the best sensor nodes are elected according to the number of packets forwarded between sensor nodes, the number of hops to the tree’s root, the residual energy, and the distance between the node and the closest CH. In simulation results, we adopt the balanced and unbalanced topologies and prove the efficiently of our proposed algorithm considering the network lifetime, the fairness index and the energy consumption in comparison with the existing mobile data collection algorithms.

CORL 2: Exploring Cooperative Opportunities, Reducing Latency and Prolonging Network Lifetime for Data Collection Using Mobile Sinks in Wireless Sensor Networks

Data collection is one of the most important issues in Wireless Sensor Networks (WSNs). Many data collection algorithms have been proposed for collecting data using multiple mobile sinks. However, the anchor selection strategy, which significantly impacts the network lifetime, still can be improved. In addition, the collaboration between multiple mobile sinks is still rare, resulting in the long latency of each round of data collection. This paper proposes an Exploring Cooperative Opportunities, Reducing Latency and Prolong Network Lifetime algorithm for Data Collection, called CORL <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> , aiming to improve the network lifetime and reduce the latency of data collection. Initially, the proposed CORL <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> partitions the monitoring region into several equal-sized subregions and assigns each mobile sink to each subregion. Then CORL <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> further partitions each subregion into several equal-sized grids and constructs a path passing through all grids in that subregion. To reduce the energy consumption for packet forwarding, the proposed CORL <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> further partitions each grid into several small grids and adjusts the path such that it can pass through high-density small grids, aiming to maximize the number of anchors and prolong the network lifetime. Then the cooperative strategy is applied to cooperatively reduce the maximal latency of data collection in all subregions and balance the data collection workload between neighboring mobile sinks. Performance evaluations reveal that the proposed CORL <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> outperforms existing data collection algorithms in terms of latency, energy consumptions, the workload balancing improving index(WBII), and network lifetime for data collection in the given WSNs.

Sleep-Scheduling-Based Hierarchical Data Collection Algorithm for Gliders in Underwater Acoustic Sensor Networks

In recent years, underwater acoustic sensor networks (UASNs) have been widely investigated for ocean environmental monitoring, offshore exploration, and marine military. The core function of UASNs is to collect data for related operations. A number of factors make the monitoring challenging; ocean thermoclines may affect the communication of the underwater nodes and gliders, reducing their communication range at varying depth; moreover, the node movement caused by Ekman drifting effect can significantly interfere with the data transmissions. Thus, these factors are regarded essential towards characterizing the ocean environment. To address these challenges, a sleep-scheduling-based hierarchical data collection algorithm (SSHDCA) for underwater gliders is designed. The UASN is split into multiple virtual cubes, where the nodes in different virtual cubes sleep and work alternately to save energy. Then, the SSHDCA divides the network into a dynamic layer and a static layer. In the dynamic layer, a virtual-cube-based multi-hop method is leveraged to transmit data packets to the central area. In the static layer, an improved density-based clustering technique is applied to assign each node to an appropriate cluster, while the underwater gliders collect data from the cluster heads. Further, to reduce energy consumption, the SSHDCA compresses key and non-key data, reducing the size of the packets. Simulation results have shown that the proposed algorithm is effective in reducing the path length of the gliders and the average energy consumption of the nodes, while increasing the remaining operational life of the whole network.

Multi-AUV Collaborative Data Collection Algorithm Based on Q-Learning in Underwater Acoustic Sensor Networks

Intelligent data collection is a key component in underwater acoustic sensor networks, and it plays an important role in seabed environment monitoring, marine resource detection and marine disaster early warning. Owing to the particularities of the underwater environment, such as reduced infrastructure and noisy communication channels, the data collected by underwater nodes are more efficiently transmitted to a control center on the surface by way of an autonomous underwater vehicle (AUV). However, with the increasing complexity of the underwater tasks, using a single AUV for data collection cannot meet the requirements of low latency and low power consumption. To solve this problem, a multi-AUV collaborative data collection algorithm that reduces the load of data collection task on a single AUV is proposed. The algorithm is divided into two stages: multi-AUV task allocation and Q-learning-based AUV path planning. The data transmission of the clusters is regarded as a set of different tasks, which are assigned to the AUVs for completion. Subsequently, path planning is performed to guide the AUVs, so that the tasks are completed promptly and at a reduced cost. Simulation results show that the proposed algorithm can leverage the energy consumption of a network and extend its lifetime. The performance of the proposed algorithm in energy consumption is increased by about 10%, and the delay of data collection is also significantly reduced.

Current and Prospective Radiation Detection Systems, Screening Infrastructure and Interpretive Algorithms for the Non-Intrusive Screening of Shipping Container Cargo: A Review

The non-intrusive screening of shipping containers at national borders serves as a prominent and vital component in deterring and detecting the illicit transportation of radioactive and/or nuclear materials which could be used for malicious and highly damaging purposes. Screening systems for this purpose must be designed to efficiently detect and identify material that could be used to fabricate radiological dispersal or improvised nuclear explosive devices, while having minimal impact on the flow of cargo and also being affordable for widespread implementation. As part of current screening systems, shipping containers, offloaded from increasingly large cargo ships, are driven through radiation portal monitors comprising plastic scintillators for gamma detection and separate, typically 3He-based, neutron detectors. Such polyvinyl-toluene plastic-based scintillators enable screening systems to meet detection sensitivity standards owing to their economical manufacturing in large sizes, producing high-geometric-efficiency detectors. However, their poor energy resolution fundamentally limits the screening system to making binary “source” or “no source” decisions. To surpass the current capabilities, future generations of shipping container screening systems should be capable of rapid radionuclide identification, activity estimation and source localisation, without inhibiting container transportation. This review considers the physical properties of screening systems (including detector materials, sizes and positions) as well as the data collection and processing algorithms they employ to identify illicit radioactive or nuclear materials. The future aim is to surpass the current capabilities by developing advanced screening systems capable of characterising radioactive or nuclear materials that may be concealed within shipping containers.

Design and Implementation of an Energy-Efficient Weather Station for Wind Data Collection

Agriculture faces critical challenges caused by changing climatic factors and weather patterns with random distribution. This has increased the need for accurate local weather predictions and weather data collection to support precision agriculture. The demand for uninterrupted weather stations is overwhelming, and the Internet of Things (IoT) has the potential to address this demand. One major challenge of energy constraint in remotely deployed IoT devices can be resolved using weather stations that are energy neutral. This paper focuses on optimizing the energy consumption of a weather station by optimizing the data collected and sent from the sensor deployed in remote locations. An asynchronous optimization algorithm for wind data collection has been successfully developed, using the development lifecyle specifically designed for weather stations and focused on achieving energy neutrality. The developed IoT weather station was deployed in the field, and it has the potential to reduce the power consumption of the weather station by more than 60%.

General practice after COVID-19: lessons learned.

<h3>Abstract</h3> <h3>Background</h3> Computerized decision support systems are becoming increasingly prevalent with advances in data collection and machine learning algorithms. However, they are scarcely used for empiric antibiotic therapy. Here we accurately predict the antibiotic resistance profiles of bacterial infections of hospitalized patients using machine learning algorithms applied to patients’ electronic medical records. <h3>Methods</h3> The data included antibiotic resistance results of bacterial cultures from hospitalized patients, alongside their electronic medical records. Five antibiotics were examined: Ceftazidime (n=2942), Gentamicin (n=4360), Imipenem (n=2235), Ofloxacin (n=3117) and Sulfamethoxazole-Trimethoprim (n=3544). We applied lasso logistic regression, neural networks, gradient boosted trees, and an ensemble combining all three algorithms, to predict antibiotic resistance. Variable influence was gauged by permutation tests and Shapely Additive Explanations analysis. <h3>Results</h3> The ensemble model outperformed the separate models and produced accurate predictions on a test set data. When no knowledge regarding the infecting bacterial species was assumed, the ensemble model yielded area under the receiver-operating-characteristic (auROC) scores of 0.73-0.79, for different antibiotics. Including information regarding the bacterial species improved the auROCs to 0.8-0.88. The effects of different variables on the predictions were assessed and found consistent with previously identified risk factors for antibiotic resistance. <h3>Conclusions</h3> Our study demonstrates the potential of machine learning models to accurately predict antibiotic resistance of bacterial infections of hospitalized patients. Moreover, we show that rapid information regarding the infecting bacterial species can improve predictions substantially. The implementation of such systems should be seriously considered by clinicians to aid correct empiric therapy and to potentially reduce antibiotic misuse. <h3>40-word summary</h3> Machine learning models were applied to large and diverse datasets of medical records of hospitalized patients, to predict antibiotic resistance profiles of bacterial infections. The models achieved high accuracy predictions and interpretable results regarding the drivers of antibiotic resistance.