با سلام خدمت کاربران در صورتی که با خطای سیستم پرداخت بانکی مواجه شدید از طریق کارت به کارت (6037997535328901 بانک ملی ناصر خنجری ) مقاله خود را دریافت کنید (تا مشکل رفع گردد).
ردیف | عنوان | نوع |
---|---|---|
1 |
A deep learning approach to the screening of malaria infection: Automated and rapid cell counting, object detection and instance segmentation using Mask R-CNN
یک روش یادگیری عمیق برای غربالگری عفونت مالاریا: شمارش خودکار و سریع سلول ها ، تشخیص اشیاء و تقسیم بندی نمونه با استفاده از Mask R-CNN-2021 Accurate and early diagnosis is critical to proper malaria treatment and hence death prevention. Several com- puter vision technologies have emerged in recent years as alternatives to traditional microscopy and rapid diagnostic tests. In this work, we used a deep learning model called Mask R-CNN that is trained on uninfected and Plasmodium falciparum-infected red blood cells. Our predictive model produced reports at a rate 15 times faster than manual counting without compromising on accuracy. Another unique feature of our model is its ability to generate segmentation masks on top of bounding box classifications for immediate visualization, making it superior to existing models. Furthermore, with greater standardization, it holds much potential to reduce errors arising from manual counting and save a significant amount of human resources, time, and cost. Keywords: Malaria diagnosis | Mask R-CNN | Computer vision | Image analysis |
مقاله انگلیسی |
2 |
Aggregate accounting research and development expenditures and the prediction of real gross domestic product
مجموع هزینه های تحقیق و توسعه حسابداری جمع آوری شده و پیش بینی تولید ناخالص داخلی واقعی-2021 The role of accounting information for public policy making has received increased
attention in recent years. Konchitchki and Patatoukas (2014a,b) demonstrate that growth
in aggregate accounting earnings can predict future growth in nominal and real Gross
Domestic Product (GDP). We extend the micro to macro literature by decomposing earnings
into the R&D and pre-R&D components. Using the Almon (1965) finite distributed lag
model, we find that both components can predict future real GDP growth with different
lead-lag structures. Importantly, this decomposition significantly increases the explanatory
power of the predictive model using accounting information. Aggregate accounting R&D can
predict real GDP through the personal consumption, business investment, and net export
channels of GDP. Our study extends prior research on the forecasting usefulness of accounting information at the aggregate level and has practical implications for macro forecasting
and for public policy making regarding innovative activities of publicly listed firms.
keywords: مجموع اعداد حسابداری | هزینه های تحقیق و توسعه | تولید ناخالص داخلی | پیش بینی کلان اقتصادی | ساختارهای تاخیری توزیع شده | Aggregate accounting numbers | Research and development expenditures | Gross domestic product | Macroeconomic forecasting | Distributed lag structures |
مقاله انگلیسی |
3 |
A robust co-state predictive model for energy management of plug-in hybrid electric bus
یک مدل پیش بینی شده مشترک قدرتمند برای مدیریت انرژی اتوبوس برقی هیبریدی پلاگین-2020 This paper proposes a robust co-state predictive model for Pontryagin’s Minimum Principle (PMP)-based
energy management of plug-in hybrid electric bus (PHEB). The main innovation is that the robust costate
predictive model is only expressed by a simplified formula. Moreover, it is exclusively designed
by the Design For Six Sigma (DFSS) method in consideration of noises of driving cycles and stochastic
vehicle mass. Because the DFSS strives to minimize the weighted sum of mean and standard deviation of
fuel consumption, the proposed strategy can simultaneously improve the fuel economy of the PHEB and
its robustness. The DFSS results show that the coefficients of the robust co-state predictive model can be
found; the simulation results demonstrate that the proposed strategy has similar fuel economy to dynamic
programming (DP); the hardware-in-loop (HIL) results demonstrate that the proposed strategy
has good real-time control performance, and can averagely improve the fuel economy by 35.19%
compared to a rule-based control strategy. Keywords: Plug-in hybrid electric bus | Energy management | PMP | Co-state predictive model | Design for six sigma |
مقاله انگلیسی |
4 |
Discrimination of automotive window tint using ATR-FTIR spectroscopy and chemometrics
تبعیض رنگ پنجره اتومبیل با استفاده از طیف سنجی ATR-FTIR و شیمی سنجی-2020 Automotive window tints are commonly applied to motor vehicles to reduce transmittance of light and
heat into the interior. They may hence be encountered as physical evidence in criminal investigations, or
in civil matters where a tint is suspected to originate from a different source than advertised. Establishing
a tint’s provenance would be highly relevant in such cases. However, there are currently a lack of
established guidelines for forensic tint analysis. This study used attenuated total reflectance Fourier
transform infrared (ATR-FTIR) spectroscopy with chemometrics to characterize automotive tints based
on their adhesive composition. Minimal variability was observed within a single roll of tint, however
substantial variability was observed between tints of different brands. Certain individual tint products
were also found to possess highly distinctive spectra. Subsequent predictive models were able to
associate unknown tint samples to their brand, and found to be robust to both adhesive curing and shortterm
environmental exposure over a five-month period. The use of ATR-FTIR spectroscopy and
chemometrics thus offers a rapid and objective approach to discriminating automotive tints for forensic
purposes. Keywords: Polymer | Automotive | Forensic | Chemometrics | Infrared spectroscopy |
مقاله انگلیسی |
5 |
Predicting academic performance with Artificial Intelligence (AI), a new tool for teachers and students
پیش بینی عملکرد تحصیلی با هوش مصنوعی ، ابزاری جدید برای معلمان و دانش آموزان-2020 Abstract—Learning Analytics (LA) is data science
applied to the educational field. It enables the
measurement, collection, and analysis of learners’ data
and their context. In this research we utilized two
algorithms from the field of artificial intelligence (AI): KNearest
Neighbor and Random Forest. These algorithms
trained a predictive model for the academic performance
of students pursuing an engineering degree. This research
found that a general picture of the performance of the
group is enough to improve, despite the forecast for each
student not being accurate. This allowed the instructor to
adapt their teaching technique to get better results.
Finally, most students agree to take advantage of LA and
they think that knowing their predictive results at the
beginning of the course will help them do better in class. Keywords: Artificial intelligence | Educational innovation | Learning analytics | Higher education |
مقاله انگلیسی |
6 |
Big Data Everywhere
داده های بزرگ در همه جا-2020 Big Data and machine-learning approaches to analytics are an important new frontier in
laboratory medicine.
Direct-to-consumer (DTC) testing raises specific challenges in applying these new tools of
data analytics.
Because DTC data are not centralized by default, there is a need for data repositories to
aggregate these values to develop appropriate predictive models.
The lack of a default linkage between DTC results and medical outcomes data also limits
the ability to mine these data for predictive modeling of disease risk.
Issues of standardization and harmonization, which are a significant issue across all laboratory
medicine, may be particularly difficult to correct in aggregated sets of DTC data KEYWORDS : Big Data | Laboratory medicine | Machine learning | Direct-to-consumer testing | DTC | Harmonization |
مقاله انگلیسی |
7 |
Efficient hyperparameter optimization through model-based reinforcement learning
بهینه سازی ابرپارامتر کارآمد از طریق یادگیری تقویتی مبتنی بر مدل-2020 Hyperparameter tuning is critical for the performance of machine learning algorithms. However, a noticeable
limitation is the high computational cost of algorithm evaluation for complex models or for large
datasets, which makes the tuning process highly inefficient. In this paper, we propose a novel modelbased
method for efficient hyperparameter optimization. Firstly, we frame this optimization process as
a reinforcement learning problem and then employ an agent to tune hyperparameters sequentially. In
addition, a model that learns how to evaluate an algorithm is used to speed up the training. However,
model inaccuracy is further exacerbated by long-term use, resulting in collapse performance. We propose
a novel method for controlling the model use by measuring the impact of the model on the policy and
limiting it to a proper range. Thus, the horizon of the model use can be dynamically adjusted. We apply
the proposed method to tune the hyperparameters of the extreme gradient boosting and convolutional
neural networks on 101 tasks. The experimental results verify that the proposed method achieves the
highest accuracy on 86.1% of the tasks, compared with other state-of-the-art methods and the average
ranking of runtime is significant lower than all methods by using the predictive model. Keywords: Hyperparameter optimization | Machine learning | Reinforcement learning |
مقاله انگلیسی |
8 |
Optimization & validation of Intelligent Energy Management System for pseudo dynamic predictive regulation of plug-in hybrid electric vehicle as donor clients
بهینه سازی و اعتبار سنجی سیستم مدیریت انرژی هوشمند برای تنظیم پیش بینی شبه دینامیکی پلاگین در خودروهای برقی هیبریدی به عنوان مشتری دهنده-2020 In developing countries, policies for discarding the existing Internal Combustion (IC) Engine vehicles for
faster adoption of Electric Vehicles’ not only creates burden on the existing power grid but also is
impractical. The conversion of Conventional IC Engine based Online Taxis or public transport vehicles
into Plug-in Hybrid Electric Vehicles donor clients, to participate in Vehicle to Grid & Vehicle to Vehicle
power transfer model, is the solution. These vehicles would not only have emissions within compliance
standards but would also reduces the load on the power grid meanwhile making an income through
power transfer. The Intelligent Energy Management System (IEMS) developed makes use of a Non
Dominated Sorting Genetic Algorithm (NSGA-II) based Pseudo dynamic predictive regulation approach
on the powertrain to optimize the emissions, fuel cost and traction battery SoC. If the vehicle intends to
participate in power transfer, the IEMS would predetermine the amount of SoC that would be used for an
upcoming journey using Global Positioning System(GPS) data interconnected with a server unit which
enables the IEMS to optimize the operating conditions of the vehicle. The modelled IEMS performance is
tested for a given driving cycle with varying traffic levels on a virtual simulation environment using the
IPG CarMaker software. A prototype with a 150 cc, 7.5 kW IC engine integrated to a 3 kW BLDC traction
motor is developed and the response to the predictive model is evaluated and found to provide 27.66%,
13.73% and 7.72% equivalent energy to micro grid for low, medium and high criticality conditions for the
user. Keywords: Vehicle to grid | NSGA-II optimization | State of charge | Emission | GPS | Real-time validation |
مقاله انگلیسی |
9 |
Predicting and explaining corruption across countries: A machine learning approach
پیش بینی و توضیح فساد در سراسر کشور: رویکرد یادگیری ماشینی-2020 In the era of Big Data, Analytics, and Data Science, corruption is still ubiquitous and is perceived as one of the
major challenges of modern societies. A large body of academic studies has attempted to identify and explain the
potential causes and consequences of corruption, at varying levels of granularity, mostly through theoretical
lenses by using correlations and regression-based statistical analyses. The present study approaches the phenomenon
from the predictive analytics perspective by employing contemporary machine learning techniques to
discover the most important corruption perception predictors based on enriched/enhanced nonlinear models
with a high level of predictive accuracy. Specifically, within the multiclass classification modeling setting that is
employed herein, the Random Forest (an ensemble-type machine learning algorithm) is found to be the most
accurate prediction/classification model, followed by Support Vector Machines and Artificial Neural Networks.
From the practical standpoint, the enhanced predictive power of machine learning algorithms coupled with a
multi-source database revealed the most relevant corruption-related information, contributing to the related
body of knowledge, generating actionable insights for administrator, scholars, citizens, and politicians. The
variable importance results indicated that government integrity, property rights, judicial effectiveness, and
education index are the most influential factors in defining the corruption level of significance Keywords: Corruption perception | Machine learning | Predictive modeling | Random forest | Society policies and regulations |Government integrity | Social development |
مقاله انگلیسی |
10 |
The impact of entrepreneurship orientation on project performance: A machine learning approach
تأثیر گرایش کارآفرینی بر عملکرد پروژه: یک رویکرد یادگیری ماشینی-2020 Recent studies in project management have shown the important role of entrepreneurship orientation of the
individuals in project performance. Although identifying the role of entrepreneurship orientation as a critical
success factor in project performance has been considered as an important issue, it is also important to develop a
measurement system for predicting performance based on the degree of an individual’s entrepreneurial orientation.
In this study, we use predictive analytics by proposing a machine learning approach to predict individuals’
project performance based on measures of several aspects of entrepreneurial orientation and
entrepreneurial attitude of the individuals. We investigated this relationship using a sample of 185 observations
and a range of machine learning algorithms including lasso, ridge, support vector machines, neural networks,
and random forest. Our results showed that the best method for predicting project performance is lasso. After
identifying the best predictive model, we then used the Bayesian Information Criterion and the Akaike Information
Criterion to identify the most significant factors. Our results identify all three aspects of entrepreneurial
attitude (social self-efficacy, appearance self-efficacy, and comparativeness) and one aspect of entrepreneurial
orientation (proactiveness) as the most important factors. This study contributes to the relationship between
entrepreneurship skills and project performance and provides insights into the application of emerging tools in
data science and machine learning in operations management and project management research. Keywords: Project performance | Entrepreneurship orientation | Machine learning | Supervised learning | Predictive analytics |
مقاله انگلیسی |