Results 1 - 10 of 908
Results 1 - 10 of 908. Search took: 0.023 seconds
|Sort by: date | relevance|
[en] The objective of this report is to demonstrate the use of a methology supporting decisions on protective measures following severe nuclear accidents. A multicriteria decision analysis approach is recommended where value tradeoffs are postponed until the very last stage of the decision process. Use of efficient frontiers is made to exclude all technically inferior solutions and present the decision maker with all nondominated solutions. A choice among these solutions implies a value trade-off among the multiple criteria. An interactive computer packge has been developed where the decision maker can choose a point on the efficient frontier in the consequence space and immediately see the alternative in the decision space resulting in the chosen consequences. The methodology is demonstrated through an application on the choice among possible protective measures in contaminated areas of the former USSR after the Chernobyl accident. Two distinct cases are considered: First a decision is to be made only on the basis of the level of soil contamination with Cs-137 and the total cost of the chosen protective policy; Next the decision is based on the geographic dimension of the contamination ant the total cost. Three alternative countermeasure actions are considered for population segments living on soil contaminated at a certain level or in a specific geographic region: (a) relocation of the population; (b) improvement of the living conditions; and, (c) no countermeasures at all. This is final deliverable of the CEC-CIS Joint Study Project 2, Task 5: Decision-Aiding-System for Establishing Intervention Levels, performed under Contracts COSU-CT91-0007 and COSU-CT92-0021 with the Commission of European Communities through CEPN
[en] We describe the implementation of Boosted Decision Trees in the hls4ml library, which allows the translation of a trained model into FPGA firmware through an automated conversion process. Thanks to its fully on-chip implementation, hls4ml performs inference of Boosted Decision Tree models with extremely low latency. With a typical latency less than 100 ns, this solution is suitable for FPGA-based real-time processing, such as in the Level-1 Trigger system of a collider experiment. These developments open up prospects for physicists to deploy BDTs in FPGAs for identifying the origin of jets, better reconstructing the energies of muons, and enabling better selection of rare signal processes.
[en] In this paper two machine learning algorithms, decision trees (DT) and Hamming clustering (HC), are compared in building approximate reliability expression (RE). The main idea is to employ a classification technique, trained on a restricted subset of data, to produce an estimate of the RE, which provides reasonably accurate values of the reliability. The experiments show that although both methods yield excellent predictions, the HC procedure achieves better results with respect to the DT algorithm
[en] We present a simple introduction to the decision tree algorithm using some examples from nuclear physics. We show how to improve the accuracy of the classical liquid drop nuclear mass model by performing feature engineering with a decision tree. Finally, we apply the method to the Duflo–Zuker model showing that, despite their simplicity, decision trees are capable of improving the description of nuclear masses using a limited number of free parameters. (guide)
[en] This paper studies propositional proof systems in which lines are sequents of decision trees or branching programs, deterministic or non-deterministic. Decision trees (DTs) are represented by a natural term syntax, inducing the system LDT, and non-determinism is modelled by including disjunction, , as primitive (system LNDT). Branching programs generalise DTs to dag-like structures and are duly handled by extension variables in our setting, as is common in proof complexity (systems eLDT and eLNDT). Deterministic and non-deterministic branching programs are natural nonuniform analogues of log-space (L) and nondeterministic log-space (NL), respectively. Thus eLDT and eLNDT serve as natural systems of reasoning corresponding to L and NL, respectively. The main results of the paper are simulation and non-simulation results for tree-like and dag-like proofs in LDT, LNDT, eLDT and eLNDT. We also compare them with Frege systems, constant-depth Frege systems and extended Frege systems.
[en] Real networks exhibit nontrivial topological features, such as heavy-tailed degree distribution, high clustering, and small-worldness. Researchers have developed several generative models for synthesizing artificial networks that are structurally similar to real networks. An important research problem is to identify the generative model that best fits to a target network. In this paper, we investigate this problem and our goal is to select the model that is able to generate graphs similar to a given network instance. By the means of generating synthetic networks with seven outstanding generative models, we have utilized machine learning methods to develop a decision tree for model selection. Our proposed method, which is named “Generative Model Selection for Complex Networks,” outperforms existing methods with respect to accuracy, scalability, and size-independence
[en] In case of radiological contamination of foodstuffs, the introduction of a countermeasure has to be justified by balancing its advantages and drawbacks, as recommended by ICRP. Also, to provide authorities with information about the decision context, it has been decided to develop a Decision Support System (DSS). A knowledge-based approach is used for the DSS. Indeed, it allows: . better modelling thanks to, for instance, object oriented programming and rules, . ability to introduce more knowledge thanks to an easier consistency and validity control of the knowledge base, . handling of uncertainties (incomplete, uncertain or evolving knowledge). The present state of the system is presented. DACFOOD is a decision aiding system for contamined foodstuffs, based on a knowledge-based approach. A demonstration model has been developed in a post-Chernobyl CEC research program. It evaluates the sanitary situation, the alternative actions through costs and sanitary effects, and gives information on the decisional background
[en] The nodular thyroid disease frequently involves delicate diagnostic problems, in particular concerning the diagnosis of thyroid cancer. Therefore, in case of a palpable thyroid nodule the physician must first of all perform the proper procedures (ultrasonography, scintigraphy, fine-needle aspiration biopsy, etc.) for a selective approach to diagnosis, as a necessary step before an eventual surgical treatment. (author)
[en] Background and purpose: The absolute number of new cancer patients that will require at least one course of radiotherapy in each country of Europe was estimated. Material and methods: The incidence and relative frequency of cancer types from the year 2012 European Cancer Observatory estimates were used in combination with the population-based stage at diagnosis from five cancer registries. These data were applied to the decision trees of the evidence-based indications to calculate the Optimal Utilization Proportion (OUP) by tumour site. Results: In the minimum scenario, the OUP ranged from 47.0% in the Russian Federation to 53.2% in Belgium with no clear geographical pattern of the variability among countries. The impact of stage at diagnosis on the OUP by country was rather limited. Within the 24 countries where data on actual use of radiotherapy were available, a gap between optimal and actual use has been observed in most of the countries. Conclusions: The actual utilization of radiotherapy is significantly lower than the optimal use predicted from the evidence based estimates in the literature. This discrepancy poses a major challenge for policy makers when planning the resources at the national level to improve the provision in European countries
[en] Configuration Management ensures facility continues to meet goals (safety, production etc) through its lifetime. It consists of processes and tools that document as-designed/as-built facility, operational modalities, current licensing basis, as well as all other licensing and operational requirements. It also controls changes to the above.