با سلام خدمت کاربران عزیز، به اطلاع می رساند ترجمه مقالاتی که سال انتشار آن ها زیر 2008 می باشد رایگان بوده و میتوانید با وارد شدن در صفحه جزییات مقاله به رایگان ترجمه را دانلود نمایید.
Probabilistic active learning: An online framework for structural health monitoring
یادگیری فعال احتمالی: یک چارچوب آنلاین برای نظارت بر سلامت ساختاری-2019
A novel, probabilistic framework for the classification, investigation and labelling of data is suggested as an online strategy for Structural Health Monitoring (SHM). A critical issue for data-based SHM is a lack of descriptive labels (for measured data), which correspond to the condition of the monitored system. For many applications, these labels are costly and/or impractical to obtain, and as a result, conventional supervised learning is not feasible. This fact forces a dependence on outlier analysis, or one-class classifiers, in practical applications, as a means of damage detection. The model suggested in this work, however, allows for the definition of a multi-class classifier, to aid both damage detection and identification, while using a limited number of the most informative labelled data. The algorithm is applied to three datasets in the online setting; the Z24 bridge data, a machining (acoustic emission) dataset, and measurements from ground vibration aircraft tests. In the experiments, active learning is shown to improve the online classification performance for damage detection and classification.
Keywords: Damage detection | Pattern recognition | Semi-supervised learning |Structural health monitoring
مقایسه برداشت ها و نگرش های دانشجویان درباره تجربه دستِ اول تنفس سنجی دربرابر یادگیری فعال مبتنی بر کاغذ
سال انتشار: 2019 - تعداد صفحات فایل pdf انگلیسی: 6 - تعداد صفحات فایل doc فارسی: 17
زمینه و هدف: تنفس سنجی اغلب ازنظر فنی برای بیماران چالش برانگیز است. مطالعات قبلی سودهای بالقوه تنفس سنجی را در زمینه داروخانه های جامعه ای نشان داده اند. این مطالعه برداشت ها و نگرش های دانشجویان رشته داروسازی نسبت به انجام دادن تنفس سنجی و نیز اجرای تنفس سنجی در کلینیک ها و داروخانه های جامعه ای ازطریق تجربه کردن تمرینات یادگیری تنفس سنجی دست اول را با تمرینات یادگیری فعالانه مبتنی بر کاغذ مقایسه می کند. فعالیت آموزشی و شرایط: دانشجویان سال اول (102 نفر) و سال دوم (70 نفر) رشته داروسازی با مواد درسی پیش از کلاس یکسان جهت یادگیری فرآیند تنفس سنجی فراهم آمدند. درطی کلاس، دانشجویان سال اول آزمایشات تنفس سنجی را انجام دادند درحالیکه دانشجویان سال دوم تمرینات یادگیری فعالانه مبتنی بر کاغذ را درباره تنفس سنجی بدون انجام دادن آزمایش انجام دادند. یک ارزیابی برای هر گروه در انتهای کلاس جهت (1) مقایسه برداشت دانشجویان درباره دشواری انجام تنفس سنجی و (2) شناسایی موانع بیمار، کلینیکی و دارویی اجرای آزمایش تنفس سنجی انجام شد. یافته ها: دانشجویان سال اول این چنین برداشت کردند که انجام تنفس سنجی به صورت قابل توجهی درمقایسه با دانشجویان سال دوم دشوارتر است. هم دانشجویان سال اول و هم سال دوم این چنین برداشت کردند که نحوه قرارگیری صحیح و روش نفس زنی و ناراحتی بیمار دشوارترین بخشهای انجام تنفس سنجی هستند. دانشجویان سال اول (1/91%) درمقایسه با دانشجویان سال دوم (1/54%) به صورت قابل توجهی تنفس سنجی را یک ابزار کمک کننده تر و غیرتعارضی تر برای رصد کردن بیماری های ریوی دانستند. خلاصه:دانشجویانی که تنفس سنجی را تجربه کردند نسبت به دانشجویانی که تمرینات یادگیری فعال مبتنی بر کاغذ را انجام دادند این چنین براشت کردند که تنفس سنجی دشوارتر است. وارد کردن تنفس سنجی به برنامه درسی رشته داروسازی می تواند فرصتی برای افزایش بینش دانشجویان درباره دشواری انجام دادن تنفس سنجی و افزایش قدردانی آنها از خدمات کلینیکی دارویی باشد.
|مقاله ترجمه شده|
Active deep learning for the identification of concepts and relations in electroencephalography reports
یادگیری عمیق فعال برای شناسایی مفاهیم و روابط در گزارشات الکتروانسفالوگرافی-2019
The identification of medical concepts, their attributes and the relations between concepts in a large corpus of Electroencephalography (EEG) reports is a crucial step in the development of an EEG-specific patient cohort retrieval system. However, the recognition of multiple types of medical concepts, along with the many attributes characterizing them is challenging, and so is the recognition of the possible relations between them, especially when desiring to make use of active learning. To address these challenges, in this paper we present the Self- Attention Concept, Attribute and Relation (SACAR) identifier, which relies on a powerful encoding mechanism based on the recently introduced Transformer neural architecture (Dehghani et al., 2018). The SACAR identifier enabled us to consider a recently introduced framework for active learning which uses deep imitation learning for its selection policy. Our experimental results show that SACAR was able to identify medical concepts more precisely and exhibited enhanced recall, compared with previous methods. Moreover, SACAR achieves superior performance in attribute classification for attribute categories of interest, while identifying the relations between concepts with performance competitive with our previous techniques. As a multi-task network, SACAR achieves this performance on the three prediction tasks simultaneously, with a single, complex neural network. The learning curves obtained in the active learning process when using the novel Active Learning Policy Neural Network (ALPNN) show a significant increase in performance as the active learning progresses. These promising results enable the extraction of clinical knowledge available in a large collection of EEG reports.
Keywords: Deep learning | Electroencephalography | Active learning | Long-distance relation identification | Concept detection | Attribute classification
Breast mass detection from the digitized X-ray mammograms based on the combination of deep active learning and self-paced learning
تشخیص توده سینه از ماموگرافی های دیجیتالی شده اشعه ایکس بر اساس ترکیبی از یادگیری عمیق فعال و یادگیری خود گام-2019
Breast mass detection is a challenging task in mammogram, since mass is usually embedded and surrounded by various normal tissues with similar density. Recently, deep learning has achieved impressive performance on this task. However, most deep learning methods require large amounts of well-annotated datasets. Generally, the training datasets is generated through manual annotation by experienced radiologists. However, manual annotation is very time-consuming, tedious and subjective. In this paper, for the purpose of minimizing the annotation efforts, we propose a novel learning framework for mass detection that incorporates deep active learning (DAL) and self-paced learning (SPL) paradigm. The DAL can significantly reduce the annotation efforts by radiologists, while improves the efficiency of model training by obtaining better performance with fewer overall annotated samples. The SPL is able to alleviate the data ambiguity and yield a robust model with generalization capability in various scenarios. In detail, we first employ a few of annotated easy samples to initialize the deep learning model using Focal Loss. In order to find out the most informative samples, we propose an informativeness query algorithm to rank the large amounts of unannotated samples. Next, we propose a self-paced sampling algorithm to select a number of the most informative samples. Finally, the selected most informative samples are manually annotated by experienced radiologists, which are added into the annotated samples for the model updating. This process is looped until there are not enough most informative samples in the unannotated samples. We evaluate the proposed learning framework on 2223 digitized mammograms, which are accompanied with diagnostic reports containing weakly supervised information. The experimental results suggest that our proposed learning framework achieves superior performance over the counterparts. Moreover, our proposed learning framework dramatically reduces the requirement of the annotated samples, i.e., about 20% of all training data.
Keywords: Breast cancer | Mammography | Mass detection | Deep active learning | Self-paced learning
I explain, therefore I learn: Improving students’ assessment literacy and deep learning by teaching
من شرح می دهم، بنابراین من یاد می گیرم: بهبود سواد ارزیابی دانش آموزان و یادگیری عمیق با تدریس-2019
Partnership in higher education emphasises an active role for students in both teaching and learning. This pedagogical culture is likely to make students assessment literate and engage them in deep learning. In this study, Iranian students experiencing learning-by-teaching (LbT) in private language institutes were interviewed to compare their perceptions toward assessment and learning with their counterparts without this experience. Findings show that LbT fosters students’ assessment literacy and deep learning. Results also reveal that by teaching other students, quasi-teachers promote a broader understanding of assessment and grade practices in comparison to other students. Unlike their counterparts, quasi-teachers de-emphasised grades and showed a greater focus on learning. Moreover, explaining the materials to other students provided them with a deeper cognitive process resulting in deeper learning. These results underscore the perceived importance of partnership in higher education.
Keywords: Learning-by-teaching (LbT) | Assessment literacy | Deep learning | Active learning | Learning approach
Joint temporal context exploitation and active learning for video segmentation
استخراج زمینه زمانی مشترک و یادگیری فعال برای تقسیم بندی ویدیو-2019
The segmentation of video, or separating out objects in the foreground, is an important application of pattern recognition and computer vision. Segmentation errors in pattern recognition approaches mainly come from difficulties in selecting maximally informative frames for learning. In this paper, we develop an approach to video segmentation that relies on temporal features by modeling the uncertainty of the distribution of different feature mask forms. We use those uncertainty values for unsupervised active learning. We evaluate our approach on the DAVIS16 annotated video data set and Shining3D dental video data set, and the results show our approach to be more accurate than other video segmentation ap- proaches
Keywords: Video segmentation | Deep learning | Computer vision
Impact of using classroom response systems on students entrepreneurship learning experience
تأثیر استفاده از سیستم های پاسخ کلاس در تجربه یادگیری کارآفرینی دانش آموزان-2019
Technology-based teaching devices that promote interaction and communication between instructors and learners benefit active learning. Emerging technologies for Classroom Response Systems (CRS) and mobile devices can potentially help instructors create a student-centered interactive classroom. In this study, the authors aim to evaluate students experiences of using mobile-based CRS technology in the context of an entrepreneurship course. This study involves 22 graduate students enrolled in an 18-week course in Entrepreneurship Management. This study examines how their learning could be supported and enhanced by CRS technology. Results indicate that mobile-based CRS technology is a useful and effective tool for facilitating interaction among learners and content, enhancing students engagement with entrepreneurial knowledge acquisition, and improving students motivation toward increased entrepreneurial capability. In particular, students experience innovative, active, and deep learning in a mobile-based and CRS-supported classroom regardless of time and location.
Keywords: Classroom response systems | Entrepreneurship education | Mobile learning | Curriculum assessment | Learning | Participation | Higher education
The lean and resilient management of the supply chain and its impact on performance
مدیرت زنجیره تامین متکی و برجهنده و تاثیر آن روی عملکرد-2018
The relationship between lean management and resilience in the supply chain, whether negative or positive, is still not clear from the existing literature. This paper aims to investigate the relationship and links between lean and resilient supply chain (SC) practices and their impact on SC performance. To achieve this objective, the aerospace manufacturing sector (AMS) is chosen as the study sector because of the importance of both paradigms. Interpretive Structural Modeling (ISM) approach is used in order to identify linkages among various lean and resilience practices and SC performance metrics through a single systemic framework. ISM is an interactive learning process based on graph theory where experts knowledge is extracted and converted into a powerful well-structured model. For that purpose, a heterogeneous panel of experts in the AMS was formed, providing a complete view of all SC levels in the sector. The final ISM model revealed that lean SC practices act as drivers for resilient SC practices, since implementing the former in isolation could lead to a more vulnerable SC. The findings also show that lean SC practices lead to a higher performance improvement than resilient SC practices. This is due to the fact that resilient SC practices do not exert influence over all SC performance metrics as it occurs with lean SC practices. In addition, several managerial implications regarding the most convenient practices in terms of the companys objectives are drawn from this study.
keywords: Lean supply chain management |Resilient supply chain management |Interpretive structural modeling |Aerospace manugacturing sector
Learning symbols from permanent and transient visual presentations: Dont overplay the hand
نمادهای یادگیری از ارائه های دائم و موقت بصری: دردست اغراق نکن-2018
Instructional dynamic pictures (animations and videos) contain transient visual information. Consequently, when learning from dynamic pictures, students must process in working memory the current images while trying to remember the images that left the screen. This additional activity in working memory may lead dynamic pictures to be less suitable instructional materials than comparable static pictures, which are more permanent. In order to directly show the influence of transient visual information on dynamic learning environments, we designed a well-matched comparison between a permanent and a transient presentation of an abstract-symbol memory task on the computer. In the task, 104 university students (50% females) had to memorize the type, color, and position of the symbols in a rectangular configuration. In addition, an embodied cognition factor was included where the symbols in the task were either shown with a precision grasping static hand or not. We also assessed how individual characteristics (spatial ability, spatial memory span, and gender) influenced performance. Results showed that (a) permanent outperformed transient presentations, (b) observing hands hindered learning, and (c) high spatial ability and high spatial memory span were beneficial, but gender did not affect performance.
keywords: Applications in subject areas| Gender studies| Interactive learning environments| Media in education| Multimedia/hypermedia systems
The effect of online argumentation of socio-scientific issues on students scientific competencies and sustainability attitudes
تاثیر بحث های آنلاین درباره مسائل اجتماعی - علمی روی شایستگی ها علمی و ویژگی های ماندگاری دانشجویان-2018
One focal point of science learning is to develop students ability to actively participate in discussions of socio-scientific issues (SSIs) in their daily lives. This study proposed the SSIs-Online-Argumentation Pattern (SOAP) to develop a pedagogical strategy enabling students to participate in online argumentation of SSIs. Two quasi-experiments were conducted to investigate the variations in scientific competencies and sustainability attitudes of students following the SOAP strategy. The participants were 127 senior high school students and 68 undergraduates respectively. Students scientific competencies and sustainability attitudes were assessed using quantitative methods. The results showed that the SOAP strategy led to differences in high school students scientific competencies. The mean scientific competency of the experimental group was higher than that of the comparison group in the post-test and in the delayed test. Specifically, for the constructs ‘identifying scientific issues’ and ‘using scientific evidence’, the difference between the two groups did not reach significance in the post-test and in the delayed test. The results showed that the SOAP strategy resulted in differences in undergraduates sustainability attitudes. In the post-test, the mean sustainability attitude of the experimental group was higher than that of the comparison group. Specifically, for the constructs of ‘economic’ aspect, the post-test difference between the two groups did not reach significance. Finally, this research proposed suggestions and implications for future studies related to SSIs and science education.
keywords: Computer-mediated communication| Interactive learning environments| Lifelong learning| Pedagogical issues| Teaching/learning strategies