Decentralized clinical decision support (CDS), using the Web browser as a local application platform, fully decouples the CDS from vendor-specific EMR, removes reliance on server infrastructure, and does not require c...
ISBN:
(纸本)9783031343438;9783031343445
Decentralized clinical decision support (CDS), using the Web browser as a local application platform, fully decouples the CDS from vendor-specific EMR, removes reliance on server infrastructure, and does not require custom software. Using GLEAN, a clinical workflow can be loaded within aWeb browser to provide decentralized and specialized CDS at a point-of-care. To that end, GLEAN workflows include all knowledge needed for their local execution;the standards-based and secure data sharing with EMR, if needed;and detection of multimorbidity conflicts. This specialized CDS will execute all decision logic locally in the Web browser;using SMART-on-FHIR, locally entered data can be securely submitted to a FHIR-compliant EMR, and remote data can be retrieved. In such a decentralized setting, clinicians can securely collaborate on multimorbidity patients: (1) by sharing workflow traces, i.e., progression of their local workflows, other clinicians can keep appraised of their decision making;and (2) by leveraging medical online knowledge sources, conflicts (e.g., drug-drug, drug-interaction) between multimorbidity decisions can be detected and resolved.
An efficient and accurate system capable of predicting contaminant source location is important for environmental monitoring and security systems. However, one of the main challenges in developing such a system is the...
ISBN:
(纸本)9783031407246;9783031407253
An efficient and accurate system capable of predicting contaminant source location is important for environmental monitoring and security systems. However, one of the main challenges in developing such a system is the high computational cost associated with modeling the physical aspects of atmospheric dispersion. The present work addresses this issue by proposing a faster solution that uses a Multi Layer Perceptron (MLP) Neural Network to predict sensor readings based on the source location coordinates. The MLP is trained using synthetic data generated by solving the advection diffusion equation numerically, and collected at predetermined sensor locations and time. The proposed method generates accurate predictions in a fraction of the time required for the conventional method. This approach can help in detecting and identifying contaminants quickly and efficiently, and it has potential applications in environmental monitoring and security systems, making it an important tool for protecting public health and safety.
By thinking, information processing and decision-making in threes, the idea, theory and methods of three-way decision have been successfully applied to various domains. However, the current three-way decision has two ...
ISBN:
(纸本)9783031509582;9783031509599
By thinking, information processing and decision-making in threes, the idea, theory and methods of three-way decision have been successfully applied to various domains. However, the current three-way decision has two following limitations. On the one hand, the narrow three-way decision associated with rough sets either has trouble processing continuous data or fails to represent knowledge by equivalence classes. On the other hand, the inputs of generalized three-way decision are individual objects rather than equivalence classes, which reduces the decision efficiency. To this end, we try to integrate efficient granular-ball computing into three-way decision. Firstly, we propose a novel model, i.e., granular-ball three-way decision to improve the efficiency and robustness of three-way decision. Secondly, sequential three-way decision based on granular-ball is presented to investigate the appropriate multi-granularity structures and represent the same object at different granularities. Finally, we analyze the advantages of granular-balls to strengthen the real-world applications of three-way decision.
Recently, the research into language models fine-tuned to follow prompts has made notable advances. These are commonly used in the form of chatbots. One special case of chatbots is that of Task-Oriented Dialogue (TOD)...
ISBN:
(数字)9783031426087
ISBN:
(纸本)9783031426070;9783031426087
Recently, the research into language models fine-tuned to follow prompts has made notable advances. These are commonly used in the form of chatbots. One special case of chatbots is that of Task-Oriented Dialogue (TOD) systems that aim to help the user achieve specific tasks using external services. High quality training data for these systems is costly to come by. We thus evaluate if the new prompt-following models can generate annotated synthetic dialogues and if these can be used to train a TOD system. To this end we generate data based on descriptions of a dialogues goal. We train a state-of-the-art TOD system to compare it in a low resource setting with and without synthetic dialogues. The evaluation shows that using prompt-following language models to generate synthetic dialogues could help training better TOD systems.
Software-Defined Networking (SDN) has received a lot of interest in recent years because of its benefits over network controllability. Nonetheless, the deployment of SDN in legacy networks is likely to take months or ...
ISBN:
(纸本)9783031414558;9783031414565
Software-Defined Networking (SDN) has received a lot of interest in recent years because of its benefits over network controllability. Nonetheless, the deployment of SDN in legacy networks is likely to take months or years due to funding constraints. Traffic scheduling that involve flow splitting provides the flexibility for traffic flow. It is able to minimize the maximum link capacity of a network and to reduce the traffic congestion in the network. The majority of the studies focus on how to balance the flows coming out of the conventional nodes and how to partition the flows that gather at the SDN nodes so that the maximum link usage of the entire network can be reduced. Energy efficiency of a network are important to save cost and energy. During traffic scheduling, the energy consumption of a traffic flow should be considered. As a result, in hybrid SDN, we offer a heuristic approach for energy and congestion awareness traffic scheduling with flow splitting. We first define the aforementioned issue in an Integer Linear Programming (ILP) model, and then we assess the suggested ILP model and heuristic algorithm in terms of solution quality and processing time. The findings indicate that with polynomial time complexity, our suggested approach retains its overall soundness.
In this era of the Internet of Things (IoT), a large number of sensor devices collect and generate various sensing data over time. It is very essential to mine fresh information by analyzing large amounts of data, pre...
ISBN:
(数字)9783031402890
ISBN:
(纸本)9783031402883;9783031402890
In this era of the Internet of Things (IoT), a large number of sensor devices collect and generate various sensing data over time. It is very essential to mine fresh information by analyzing large amounts of data, predict the future, and make correct decisions. Therefore, a growing number of data-intensive computing frameworks have been proposed, such as Hadoop, Spark, Flink, etc. Rather than reading and writing files to disks, Spark processes data with a memory-based computing framework to improve the performance, which has attracted more attention from researchers. However, due to a wealth of operators provided by Spark, a certain application can be implemented in various ways, which also show big differences in performance. Therefore, tuning a Spark application is a very error-prone and time-consuming process, and requires developers to have a deep understanding of Spark's operating principles and characteristics. In this paper, we summarize a series of rules such as operator reordering and operator replacement to design and implement a Spark program optimizer, called SPOAHA, based on the artificial Hummingbird algorithm. Experimental results show that without changing the semantics of the original program, the optimized program dramatically reduces the amount of data involved in the shuffling period, and speeds up the execution time by up to 2.7x.
Reinforcement learning is playing an increasingly important role in the field of recommender systems. In this paper, we enhance the performance of a reinforcement learning-based recommender system by incorporating a k...
ISBN:
(纸本)9783031368189;9783031368196
Reinforcement learning is playing an increasingly important role in the field of recommender systems. In this paper, we enhance the performance of a reinforcement learning-based recommender system by incorporating a knowledge graph, which serves as the embedding method for translating entities and relationships into vectors. To recommend diverse items, we define an explorative reward function for the Markov decision process that determines the recommendation. We also describe an action space pruning strategy that narrows down the reasoning space, and present a target for policy gradient optimization. Our experimental results show that the proposed method improves the recommendation performance and provides rational explanations.
Flocks navigate for large distances, moving in a coherent path through space, under mutual influence of flock members. Such influences may include repulsion, orientation, and attraction. Certain applications give rise...
ISBN:
(纸本)9783031212024;9783031212031
Flocks navigate for large distances, moving in a coherent path through space, under mutual influence of flock members. Such influences may include repulsion, orientation, and attraction. Certain applications give rise to the need to control the movements of flocks, e.g., circumventing critical zones. Researchers have investigated the problem of seeding flocks with a percentage of externally controlled agents to achieve effective flock control. Recent studies of flock control include orthogonal directions of (a) selecting influencing or leader agents and (b) orienting the leader agents. We build on these studies and evaluate combinations of selecting and orienting choices for fast convergence of the flock to follow desired travel directions with both adaptive and non-adaptive selection and orientation algorithms. We evaluate the effectiveness of combined flock control strategies under different physical world models. We explore the case of non-looping (non-toroidal) environments and attempt to overcome their challenges. (This is a continuation of work presented here [3]).
MAZE environments are popular test environments for reinforcement learning techniques as they are characterised by a sequence of discrete decisions of a learner, i.e. a multi-step learning problem where the reward is ...
ISBN:
(纸本)9783031425073;9783031425080
MAZE environments are popular test environments for reinforcement learning techniques as they are characterised by a sequence of discrete decisions of a learner, i.e. a multi-step learning problem where the reward is just available after reaching a goal. Further, MAZE scenarios can easily be altered to change the complexity of the learning problem. However, this is usually done in simulations and is seldom based on real hardware. In this paper, we investigate how far the simulation-based results can be transferred to an example platform, the ROBOTIS Turtle-Bot3 system. We explain the concept of setting up a discrete environment that is perceivable by the robot, investigate the behaviour of a Q-learner in simulations and describe how far the simulation-based behaviour can be reproduced with the robot.
For negotiation dialogue tasks, instead of adopting stationary strategies, a more advanced opponent may demonstrate sophisticated behaviors by employing reasoning strategies to predict its opponent's actions. To a...
ISBN:
(数字)9783031255496
ISBN:
(纸本)9783031255496;9783031255489
For negotiation dialogue tasks, instead of adopting stationary strategies, a more advanced opponent may demonstrate sophisticated behaviors by employing reasoning strategies to predict its opponent's actions. To address this challenge, this work proposes a novel dialogue agent, which leverages the predictive power of Bayesian policy reuse and the recursive reasoning ability of theory of mind, allowing efficiently detecting the policy of opponents using either stationary or higher-level reasoning strategies and learning a best-response policy when faced with previously unseen strategies. Finally, we present the results of the proposed agent against state-of-the-art baselines on the CRAIGSLISTBARGAIN dataset and show that the agent outperforms existing agents and its efficacy of detecting new unseen strategies (This is an extended version of the paper [3] presented at the 20th IEEE International Conference on Ubiquitous intelligence and Computing 2022.)
暂无评论