سیستم پشتیبانی از تصمیم برای خطرات و اقدامات متقابل ایمنی جاده ای اروپا
سال انتشار: 2019 - تعداد صفحات فایل pdf انگلیسی: 8 - تعداد صفحات فایل doc فارسی: 32
سیستم پشتیبانی از تصمیم درباره ایمنی جاده ای اروپا (roadsafety-dss.eu) یک سیستم نوآورانه است که شواهد و مدارک دسترس پذیری را درباره گستره وسیعی از خطرات جاده ای و اقدامات متقابل امکانپذیر فراهم می کند. این مقاله پایه و اساس علمی سیستم پشتیبانی از تصمیم را توصیف می کند. ساختار موجود در سیستم پشتیبانی از تصمیم شامل (1) یک طبقه بندی که به شناسایی عوامل خطر و اقدامات متقابل آن می پردازد و آنها را به همدیگر مرتبط می کند، (2) یک مجموعه ای از مطالعات، و (3) خلاصه هایی که تاثیرات تخمین زده شده در منابع علمی را برای هر عامل و سنجه خطر خلاصه بندی می کنند و (4) یک ابزار ارزیابی کارآمدی اقتصادی (محاسبه گر E3) می شود. سیستم پشتیبانی از تصمیم در یک ابزار نوین مبتنی بر وب با فصل مشترک بسیار انسانی اجرا می شود که به کاربران اجازه می دهد تا مرور اجمالی سریعی داشته باشند یا نتایج هر مطالعه را برطبق نیازهای مخصوص آنها عمیق تر بررسی کنند.
کلیدواژه ها: اقدامات متقابل ایمنی جاده | خطرات جاده ای | سودمندی | سیستم آنلاین | مرور | هزینه – سود
|مقاله ترجمه شده|
Automatic hourly solar forecasting using machine learning models
پیش بینی خودکار خورشیدی ساعتی با استفاده از مدل های یادگیری ماشین-2019
Owing to its recent advance, machine learning has spawned a large collection of solar forecasting works. In particular, machine learning is currently one of the most popular approaches for hourly solar forecasting. Nevertheless, there is evidently a myth on forecast accuracy—virtually all research papers claim superiority over others. Apparently, the “best” model can only be selected with hindsight, i.e., after empirical evaluation. For any new forecasting project, it is irrational for solar forecasters to bet on a single model from the start. In this article, the hourly forecasting performance of 68 machine learning algorithms is evaluated for 3 sky conditions, 7 locations, and 5 climate zones in the continental United States. To ensure a fair comparison, no hybrid model is considered, and only off-the-shelf implementations of these algorithms are used. Moreover, all models are trained using the automatic tuning algorithm available in the R caret package. It is found that tree-based methods consistently perform well in terms of two-year overall results, however, they rarely stand out during daily evaluation. Although no universal model can be found, some preferred ones for each sky and climate condition are advised.
Keywords: Automatic machine learning | Solar forecasting | R caret package
Predicting breast cancer metastasis by using serum biomarkers and clinicopathological data with machine learning technologies
پیش بینی متاستاز سرطان پستان با استفاده از نشانگرهای سرمی و داده های کلینیکوپاتولوژیکی با فن آوری های یادگیری ماشین-2019
Background: Approximately 10%–15% of patients with breast cancer die of cancer metastasis or recurrence, and early diagnosis of it can improve prognosis. Breast cancer outcomes may be prognosticated on the basis of surface markers of tumor cells and serum tests. However, evaluation of a combination of clinicopathological features may offer a more comprehensive overview for breast cancer prognosis. Materials and methods: We evaluated serum human epidermal growth factor receptor 2 (sHER2) as part of a combination of clinicopathological features used to predict breast cancer metastasis using machine learning algorithms, namely random forest, support vector machine, logistic regression, and Bayesian classification algorithms. The sample cohort comprised 302 patients who were diagnosed with and treated for breast cancer and received at least one sHER2 test at Chang Gung Memorial Hospital at Linkou between 2003 and 2016. Results: The random-forest-based model was determined to be the optimal model to predict breast cancer metastasis at least 3 months in advance; the correspondingarea under the receiver operating characteristic curve value was 0. 75 (p < 0. 001). Conclusion: The random-forest-based model presented in this study may be helpful as part of a follow-up intervention decision support system and may lead to early detection of recurrence, early treatment, and more favorable outcomes.
Keywords: Breast cancer | Machine learning | Prediction model | Cancer prognosis
Patient Clustering Improves Efficiency of Federated Machine Learning to Predict Mortality and Hospital Stay Time Using Distributed Electronic Medical Records
وشه بندی بیمار باعث افزایش کارآیی یادگیری ماشین فدرال برای پیش بینی مرگ و میر و مدت زمان ماندن بیمارستان با استفاده از سوابق پزشکی الکترونیکی توزیع شده-2019
Electronic medical records (EMRs) support the development of machine learning algorithms for predicting disease incidence, patient response to treatment, and other healthcare events. But so far most algorithms have been centralized, taking little account of the decentralized, non-identically independently distributed (non-IID), and privacy-sensitive characteristics of EMRs that can complicate data collection, sharing and learning. To address this challenge, we introduced a community-based federated machine learning (CBFL) algorithm and evaluated it on non-IID ICU EMRs. Our algorithm clustered the distributed data into clinically meaningful communities that captured similar diagnoses and geographical locations, and learnt one model for each community. Throughout the learning process, the data was kept local at hospitals, while locally-computed results were aggregated on a server. Evaluation results show that CBFL outperformed the baseline federated machine learning (FL) algorithm in terms of Area Under the Receiver Operating Characteristic Curve (ROC AUC), Area Under the Precision-Recall Curve (PR AUC), and communication cost between hospitals and the server. Furthermore, communities’ performance difference could be explained by how dissimilar one community was to others.
Keywords: distributed clustering | autoencoder | federated machine learning | non-IID | critical care
“I have the right to feel safe”: Evaluation of a school-based child sexual abuse prevention program in Ecuador
"من حق دارم احساس امنیت کنم": ارزیابی برنامه پیشگیری از سوءاستفاده جنسی کودکان مبتنی بر مدرسه در اکوادور-2019
Background: Child sexual abuse (CSA) is a complex public health problem that has lifelong implications for children’s wellbeing. Interventions may provide children strategies to protect themselves against CSA, but few have been studied in Latin America. Objective: Evaluate the immediate and medium-term impact of a 10-week educational program on children’s knowledge of CSA self-protection strategies in Ecuador. Participants and settings: Children aged 7–12 years from six public elementary schools in Ecuador were cluster-randomized to either receive the intervention between October and November 2016 (Group 1, k=4) or between March and April 2017 (Group 2, k=2). Methods: To assess CSA knowledge, a random sample of students completed a questionnaire at three time points: 1) initial: before any group received the intervention, 2) intermediate: immediately after Group 1 completed the program but before Group 2 started it, and 3) final: after Group 2 completed the program. We evaluated changes in scores using mixed linear regression models with school as a clustering variable and adjusted degrees of freedom (df=4). Results: Pre-post effect estimates at program completion adjusted for age, sex and clustering by school were 6.5% (95% CI: 2.9, 10.0) and 6.8% (95% CI 3.0, 10.7) for Groups 1 and 2, respectively. Scores did not change among children who had not yet received the intervention at intermediate evaluation (0.94%, 95%CI: −6.0, 7.9). Children in Group 1 maintained the scores six months after the program ended. Conclusions: The self-protection program increased and maintained CSA knowledge six months after the intervention finished.
Keywords: Child sexual abuse | Prevention | Primary schools | Self-protection | Low and middle income countries | Latin America
MalDy: Portable, data-driven malware detection using natural language processing and machine learning techniques on behavioral analysis reports
MalDy: تشخیص بدافزارهای قابل حمل ، داده محور با استفاده از تکنیک های پردازش زبان طبیعی و یادگیری ماشین در گزارش های تحلیل رفتاری-2019
In response to the volume and sophistication of malicious software or malware, security investigators rely on dynamic analysis for malware detection to thwart obfuscation and packing issues. Dynamic analysis is the process of executing binary samples to produce reports that summarise their runtime behaviors. The investigator uses these reports to detect malware and attribute threat types leveraging manually chosen features. However, the diversity of malware and the execution environments make manual approaches not scalable because the investigator needs to manually engineer fingerprinting features for new environments. In this paper, we propose, MalDy (mal die), a portable (plug and play) malware detection and family threat attribution framework using supervised machine learning techniques. The key idea of MalDy portability is the modeling of the behavioral reports into a sequence of words, along with advanced natural language processing (NLP) and machine learning (ML) techniques for automatic engineering of relevant security features to detect and attribute malware without the investigator intervention. More precisely, we propose to use bag-of-words (BoW) NLP model to formulate the behavioral reports. Afterward, we build ML ensembles on top of BoW features. We extensively evaluate MalDy on various datasets from different platforms (Android and Win32) and execution environments. The evaluation shows the effectiveness and the portability of MalDy across the spectrum of the analyses and settings.
Keywords: Malware | Android | Win32 | Behavioral analysis | Machine learning | NLP
Forensic psychiatric evaluations of defendants: Italy and the Netherlands compared
ارزیابی روانپزشکی پزشکی قانونی از متهمان: ایتالیا و هلند در مقایسه-2019
Background: Forensic psychiatric practices and provisions vary considerably across jurisdictions. The diversity provides the possibility to compare forensic psychiatric practices, as we will do in this paper regarding Italy and the Netherlands. Aim: We aim to perform a theoretical analysis of legislations dealing with the forensic psychiatric evaluation of defendants, including legal insanity and the management of mentally ill offenders deemed insane. This research is carried out not only to identify similarities and differences regarding the assessment of mentally ill offenders in Italy and the Netherlands, but, in addition, to identify strengths and weaknesses of the legislation and procedures used for the evaluation of the mentally ill offenders in the two countries. Results: Italy and the Netherlands share some basic characteristics of their criminal law systems. Yet, forensic psychiatric practices differ significantly, even if we consider only evaluations of defendants. A strong point of Italy concerns its test for legal insanity which defines the legal norm and enables a straightforward communication between the experts and the judges on this crucial matter. A strong point of the Netherlands concerns more standardized practices including guidelines and the use of risk assessment tools, which enable better comparisons and scientific research in this area. Conclusions: We argue that there appears to be room for improvement on both sides with regards to the evaluation of mentally ill offenders. More generally, a transnational approach to these issues, as applied in this paper, could help to advance forensic psychiatric services in different legal systems.
Keywords: Forensic psychiatry | Legal insanity | Italy | Netherlands | Risk assessment
Psychiatric patients requesting euthanasia: Guidelines for sound clinical and ethical decision making
بیماران روانی درخواست کشتن از سر ترحم: دستورالعمل هایی برای تصمیم گیری بالینی و اخلاقی سالم-2019
Background: Since Belgium legalised euthanasia, the number of performed euthanasia cases for psychological suffering in psychiatric patients has significantly increased, as well as the number of media reports on controversial cases. This has prompted several healthcare organisations and committees to develop policies on the management of these requests. Method: Five recent initiatives that offer guidance on euthanasia requests by psychiatric patients in Flanders were analysed: the protocol of Ghent University Hospital and advisory texts of the Flemish Federation of Psychiatry, the Brothers of Charity, the Belgian Advisory Committee on Bioethics, and Zorgnet-Icuro. These were examined via critical point-by-point reflection, focusing on all legal due care criteria in order to identify: 1) proposed measures to operationalise the evaluation of the legal criteria; 2) suggestions of additional safeguards going beyond these criteria; and 3) remaining fields of tension. Results: The initiatives are well in keeping with the legal requirements but are often more stringent. Additional safeguards that are formulated include the need for at least two positive advices from at least two psychiatrists; an a priori evaluation system; and a two-track approach, focusing simultaneously on the assessment of the patients euthanasia request and on that persons continuing treatment. Although the initiatives are similar in intent, some differences in approach were found, reflecting different ethical stances towards euthanasia and an emphasis on practical clinical assessment versus broad ethical reflection. Conclusions: All initiatives offer useful guidance for the management of euthanasia requests by psychiatric patients. By providing information on, and proper operationalisations of, the legal due care criteria, these initiatives are important instruments to prevent potential abuses. Apart from the additional safeguards suggested, the importance of a decision-making policy that includes many actors (e.g. the patients relatives and other care providers) and of good aftercare for the bereaved are rightly stressed. Shortcomings of the initiatives relate to the aftercare of patients whose euthanasia request is rejected, and to uncertainty regarding the way in which attending physicians should manage negative or conflicting advices, or patients suicide threats in case of refusal. Given the scarcity of data on how thoroughly and uniformly requests are handled in practice, it is unclear to what extent the recommendations made in these guidelines are currently being implemented.
Keywords: Medical assistance in dying | Psychiatry | Mental health | Belgium | Euthanasia | Guidelines
An Evaluation of Machine Learning Approaches for the Prediction of Essential Genes in Eukaryotes Using Protein Sequence-Derived Features
ارزیابی رویکردهای یادگیری ماشینی برای پیش بینی ژنهای ضروری در یوکاریوتها با استفاده از ویژگیهای حاصل از توالی پروتئین-2019
The availability of whole-genome sequences and associated multi-omics data sets, combined with advances in gene knockout and knockdown methods, has enabled large-scale annotation and exploration of gene and protein functions in eukaryotes. Knowing which genes are essential for the survival of eukaryotic organisms is paramount for an understanding of the basic mechanisms of life, and could assist in identifying intervention targets in eukaryotic pathogens and cancer. Here, we studied essential gene orthologs among selected species of eukaryotes, and then employed a systematic machine-learning approach, using protein sequence-derived features and selection procedures, to investigate essential gene predictions within and among species. We showed that the numbers of essential gene orthologs comprise small fractions when comparedwith the total number of orthologs among the eukaryotic species studied. In addition, we demonstrated that machine-learning models trainedwith subsets of essentiality-related data performed better than random guessing of gene essentiality for a particular species. Consistent with our gene ortholog analysis, the predictions of essential genes among multiple (including distantly-related) species is possible, yet challenging, suggesting that most essential genes are unique to a species. The presentwork provides a foundation for the expansion of genome-wide essentiality investigations in eukaryotes using machine learning approaches.
Keywords: Machine-learning | Essential genes | Essentiality prediction | Eukaryotes
Staffs perception of Patients’ affiliation and control in a Highly Secure Psychiatric Setting
درک کارکنان از وابستگی و کنترل بیماران در یک محیط روانی کاملاً ایمن-2019
Effective interactions between patients and staff have been associated with positive ward climate and therapeutic effects, but also pose a challenge in high secure forensic psychiatric settings. The goal of this study was to gain more insight into i) the characteristics that play a role in how staff members perceive the interpersonal style of patients, and ii) whether these perceptions are related to patients’ evaluation of ward climate and satisfaction with daily staff. Staff members (n=69), rated the interpersonal style of 102 male patients. Satisfaction with daily staff and ward climate were rated by 45 patients. Results show that patient characteristics (primary diagnosis, patient age, disruptive behavior, recent problems with symptoms of major mental disorder and recent problems with treatment or supervision response) were related to how staff perceived the interpersonal style (i.e., affiliation and control) of patients. Furthermore, the level of affiliation was positively related to patients’ satisfaction with daily staff. Patients that were seen as more controlling by staff were less satisfied with the safety on their ward (as a factor of ward climate). The results indicate that perception of patients’ interpersonal style entails patient related information and can be relevant for staff to use in their work.
Keywords: Interpersonal style | Forensic psychiatric patients | Staff members