排序方式： 共有11条查询结果，搜索用时 383 毫秒
1.
2.
Olivier Chapelle Shihao Ji Ciya Liao Emre Velipasaoglu Larry Lai Su-Lin Wu 《Information Retrieval》2011,14(6):572-592
We study the problem of web search result diversification in the case where intent based relevance scores are available. A
diversified search result will hopefully satisfy the information need of user-L.s who may have different intents. In this
context, we first analyze the properties of an intent-based metric, ERR-IA, to measure relevance and diversity altogether.
We argue that this is a better metric than some previously proposed intent aware metrics and show that it has a better correlation
with abandonment rate. We then propose an algorithm to rerank web search results based on optimizing an objective function
corresponding to this metric and evaluate it on shopping related queries. 相似文献
3.
Popularity bias is an undesirable phenomenon associated with recommendation algorithms where popular items tend to be suggested over long-tail ones, even if the latter would be of reasonable interest for individuals. Such intrinsic tendencies of the recommenders may lead to producing ranked lists, in which items are not equally covered along the popularity tail. Although some recent studies aim to detect such biases of traditional algorithms and treat their effects on recommendations, the concept of popularity bias remains elusive for group recommender systems. Therefore, in this study, we focus on investigating popularity bias from the view of group recommender systems, which aggregate individual preferences to achieve recommendations for groups of users. We analyze various state-of-the-art aggregation techniques utilized in group recommender systems regarding their bias towards popular items. To counteract possible popularity issues in group recommendations, we adapt a traditional re-ranking approach that weighs items inversely proportional to their popularity within a group. Also, we propose a novel popularity bias mitigation procedure that re-ranks items by incorporating their popularity level and estimated group ratings in two distinct strategies. The first one aims to penalize popular items during the aggregation process highly and avoids bias better, while the second one puts more emphasis on group ratings than popularity and achieves a more balanced performance regarding conflicting goals of mitigating bias and boosting accuracy. Experiments performed on four real-world benchmark datasets demonstrate that both strategies are more efficient than the adapted approach, and empowering aggregation techniques with one of these strategies significantly decreases their bias towards popular items while maintaining reasonable ranking accuracy. 相似文献
4.
Homework is an important component of most physics courses. One of the functions it serves is to provide meaningful formative assessment in preparation for examinations. However, correlations between homework and examination scores tend to be low, likely due to unproductive student behavior such as copying and random guessing of answers. In this study, we attempt to model these two counterproductive learner behaviors within the framework of Item Response Theory in order to provide an ability measurement that strongly correlates with examination scores. We find that introducing additional item parameters leads to worse predictions of examination grades, while introducing additional learner traits is a more promising approach. 相似文献
5.
Emre Özçelik 《Research Policy》2004,33(3):409-424
We examine the determinants of export performance of firms in the Turkish manufacturing industry. Prominent differences show up between innovator and non-innovator firms in terms of the impacts of such variables as firm size, advertisement intensities, ownership structures, and composition of employees. Importance of innovations and R&D activities, conduciveness of capital intensity, and insignificance of the real wage are meaningful as far as a rational international competition policy is concerned. Results are suggestive of a technology-oriented and capital-formative development path, if Turkey is to come up with the international competitive standards. 相似文献
6.
The UK's Research Excellence Framework (REF) is a system that is intended to evaluate the quality of the research produced by higher education institutes (HEIs) in the UK in three areas: quality of research outputs; impact of this research beyond academia; and research environment. For the next REF, the funding bodies have reviewed the importance of the three assessment elements and decided to increase the weight of ‘impact’ to 25% (from 20% in REF2014) and decrease the weight of ‘outputs’ to 60% (from 65% in REF2014). This article first examines the relevance of some factors for the quality of impact submissions in REF2014 and finds that larger submissions and institutes with higher external research income received better impact scores in the REF. The article then examines the units of assessment (UoAs) and HEIs that benefitted from the inclusion of the impact agenda as part of REF2014 by examining the distribution of the quality-related research (QR) funding in the 2017–2018 period and finds that the QR funding gap among different UoAs tends to decrease but the gap among HEIs in most of the UoAs increased. With the increased importance of the impact agenda as a criterion for funding bodies, it is expected that research income will be concentrated in fewer universities in the future, with the increased importance of non-academic impact. This article also discusses some of the gaming strategies and long-term investment priorities that HEIs may engage in based on the new submission rules of the next REF. 相似文献
7.
Since Proportional?+?Integral?+?Derivative (PID) controller is still the workhorse in taking over the workload of process control systems, this article introduces a new design methodology toward improving the performance of such controller. After a PI control law with windup protection is given, it is combined with a derivative path employing a first-order low pass filter in an innovative way to develop a performant controller called PI?+?DF controller. In attempting to attain a high level of control performance, gains of this controller including proportional, integral, derivative and filter gains are tuned choosing the recently introduced Stochastic Fractal Search (SFS) algorithm owing to its superiority to many state-of-the-art algorithms considering convergence, accuracy and robustness. To evaluate the efficacy of SFS, Particle Swarm Optimization (PSO) is also applied to the case study. Furthermore, the presented SFS optimized PI?+?DF controller is compared to a recently reported control scheme through simulation and experimental tests on an identical DC servo system. After providing the stability proof, SFS tuned PI?+?DF controller is found to be the pioneer in exhibiting the most accurate speed response profile under complicated scenarios, which is followed by PSO tuned PI?+?DF controller and the existing control approach, respectively. 相似文献
8.
Berrin Yanikoglu Aytac Gogus Emre Inal 《Educational technology research and development : ETR & D》2017,65(5):1369-1388
Learning through modules on a tablet helps students participate effectively in learning activities in classrooms and provides flexibility in the learning process. This study presents the design and evaluation of an application that is based on handwriting recognition technologies and e-content for the developed learning modules. The application aims to support the handwriting learning process in first grade and provides an efficient interface to the learning modules through the natural form of handwriting input. The learning modules consist of handwriting and arithmetic modules, which were deemed to benefit most from the use of handwriting recognition technologies. The developed e-content was designed to be used in the evaluations of the developed application, rather than being a complete set of exercises. Among the main findings were that the majority of the children in the two evaluations were visibly motivated to use the application for doing exercises and preferred using the tablet over pen–paper, while teachers were concerned about the extra work associated with preparing questions in the new medium. Reflections from the design of the application and students’ and teachers’ perceptions and teachers’ observations in two separate evaluations are presented with related recommendations. 相似文献
9.
Emre Özçelik 《Research Policy》2008,37(2):258-275
In this study, the determinants of private R&D investment are examined at the level of firms in the Turkish manufacturing industry. We focus our attention on the effect of public R&D support programs. Our findings indicate that public R&D support significantly and positively affects private R&D investment. There seems to be even an “acceleration effect” on firm-financed R&D expenditures. Smaller R&D performers benefit more from R&D support and perform more R&D. In addition, technology transfer from abroad and domestic R&D activity show up as complementary processes. Given the scarcity of studies on R&D support in technologically weaker economies, our hope is that the less-developed countries can exploit these findings in constructing socially beneficial technology policies. 相似文献
10.
Humeyra Ozturk Emre Fatma Hande Karpuzoglu Cihan Coskun Ebru Demirel Sezer Ozlem Goruroglu Ozturk Fatma Ucar Hikmet Can Cubukcu Fatma Demet Arslan Levent Deniz Mehmet Senes Mustafa Serteser Cevat Yazici Dogan Yucel Abdurrahman Coskun 《Biochemia medica : ?asopis Hrvatskoga dru?tva medicinskih biokemi?ara / HDMB》2021,31(1)
IntroductionTo interpret test results correctly, understanding of the variations that affect test results is essential. The aim of this study is: 1) to evaluate the clinicians’ knowledge and opinion concerning biological variation (BV), and 2) to investigate if clinicians use BV in the interpretation of test results.Materials and methodsThis study uses a questionnaire comprising open-ended and close-ended questions. Questions were selected from the real-life numerical examples of interpretation of test results, the knowledge about main sources of variations in laboratories and the opinion of clinicians on BV. A total of 399 clinicians were interviewed, and the answers were evaluated using a scoring system ranked from A (clinician has the highest level of knowledge and the ability of using BV data) to D (clinician has no knowledge about variations in laboratory). The results were presented as number (N) and percentage (%).ResultsAltogether, 60.4% of clinicians have knowledge of pre-analytical and analytical variations; but only 3.5% of them have knowledge related to BV. The number of clinicians using BV data or reference change value (RCV) to interpret measurements results was zero, while 79.4% of clinicians accepted that the difference between two measurements results located within the reference interval may be significant.ConclusionsClinicians do not use BV data or tools derived from BV such as RCV to interpret test results. It is recommended that BV should be included in the medical school curriculum, and clinicians should be encouraged to use BV data for safe and valid interpretation of test results. 相似文献