Core Code Research Articles

A Multi-Module Based Method for Generating Natural Language Descriptions of Code Fragments

Code fragment natural language description generation, also known as code summarization, refers to obtaining a natural language sequence describing a given code fragment's functionality. It is broadly agreed that applying code summarization into production can significantly improve the efficiency of software development and maintenance. In recent years, syntactic analysis (SA) technology and Latent Dirichlet Allocation (LDA) has been widely used in code summarization and has achieved good results. However, most of the existing techniques focus on core code statements, and thus their generated code summarization lacks a logical description of the code fragment's holistic information. To this end, we propose a code summarization method based on multiple modules to generate natural language for each code statement by constructing a new type of natural language template. Meanwhile, to utilize the code fragment's holistic information, we adopt the code statement partition rules and cosine similarity measure to rank and optimize the weight of the overall information of the code fragment, and finally generate the holistic natural language description of the code fragment. The experimental results demonstrate that our method can generate more concise and logical natural language descriptions than existing models.

AN IRSN CONTRIBUTION TO THE UAM PROJECT: THERMAL-HYDRAULIC AND NEUTRONIC UNCERTAINTIES PROPAGATION IN A ROD EJECTION, FIRST RESULTS

The paper presents our first results of the exercise III-I-2c from the OECD-NEA UAM-LWR benchmark intended to an elaboration of the methodology of uncertainty propagation. The considered case studied a full PWR core behavior in fast (~0.1 sec) rod ejection transient. According to the benchmark, the core represented a Hot Zero Power state. Authors used brute-force sampling propagating nuclear data and thermo-fluid uncertainties using 3D computational IRSN chain HEMERA. It couples the reactor physics code CRONOS and thermal-hydraulic core code FLICA4. The nuclear data uncertainties were represented in a form of cross sections standard deviations (in percentage of the mean cross sections values) supplied by the UAM team. In addition to the original benchmark, the study includes a case with an increased power peak by supplementary rod ejection, i.e. with higher reactivity. Both the results are similar to what we obtained in the mini-core rod ejection: the power standard deviation follows, in percentage of the mean power, the mean power curve. We split the variance with a direct calculation: once the cross sections are modified and the thermal-hydraulics inputs are kept constant, another time the contrary. The results show that uncertainties dues to nuclear data dominate over ones due to the thermal-flow area. Furthermore, the major contributors in peak-of-power variance lie in a fast group of cross sections.

Using Deep Learning to Detect Defects in Manufacturing: A Comprehensive Survey and Current Challenges.

The detection of product defects is essential in quality control in manufacturing. This study surveys stateoftheart deep-learning methods in defect detection. First, we classify the defects of products, such as electronic components, pipes, welded parts, and textile materials, into categories. Second, recent mainstream techniques and deep-learning methods for defects are reviewed with their characteristics, strengths, and shortcomings described. Third, we summarize and analyze the application of ultrasonic testing, filtering, deep learning, machine vision, and other technologies used for defect detection, by focusing on three aspects, namely method and experimental results. To further understand the difficulties in the field of defect detection, we investigate the functions and characteristics of existing equipment used for defect detection. The core ideas and codes of studies related to high precision, high positioning, rapid detection, small object, complex background, occluded object detection and object association, are summarized. Lastly, we outline the current achievements and limitations of the existing methods, along with the current research challenges, to assist the research community on defect detection in setting a further agenda for future studies.

DDOS Botnets Attacks Detection in Anomaly Traffic: A Comparative Study.

One of the major challenges that faces the acceptance and growth rate of business and governmental sites is a Botnet-based DDoS attack. A flooding DDoS strikes a victim machine by means of sending a vast amount of malicious traffic, causing a significant drop in the service quality (QoS) in IoT devices. Nonetheless, it is not that easy to detect and tackle flooding DDoS attacks, owing to the significant number of attacking machines, the usage of source-address spoofing, and the common areas shared between legitimate and malicious traffic. New kinds of attacks are identified daily, and some remain undiscovered, accordingly, this paper aims to improve the traffic classification algorithm of network traffic, that hackers use to try to be ambiguous or misleading. A recorded simulated traffic was used for both samples; normal and DDoS attack traffic, approximately 104.000 cases of each, where both datasets -which were created for this study- represent the input data in order to create a classification model, to be used as a tool to mitigate the risk of being attacked. The next step is putting datasets in a format suitable for classification. This process is done through preprocessing techniques, to convert categorical data into numerical data. A classification process is applied to capture datasets, to create a classification model, by using five classification algorithms which are; Decision Tree, Support Vector Machine, Naive Bayes, K-Neighbours and Random Forest. The core code used for classification is the python code, which is controlled by a user interface. The highest prediction, precision and accuracy are obtained using the Decision Tree and Random Forest classification algorithms, which also have the lowest processing time.

A Novel Cis-Acting RNA Structural Element Embedded in the Core Coding Region of the Hepatitis C Virus Genome Directs Internal Translation Initiation of the Overlapping Core+1 ORF.

Hepatitis C virus (HCV) genome translation is initiated via an internal ribosome entry site (IRES) embedded in the 5′-untranslated region (5′UTR). We have earlier shown that the conserved RNA stem-loops (SL) SL47 and SL87 of the HCV core-encoding region are important for viral genome translation in cell culture and in vivo. Moreover, we have reported that an open reading frame overlapping the core gene in the +1 frame (core+1 ORF) encodes alternative translation products, including a protein initiated at the internal AUG codons 85/87 of this frame (nt 597–599 and 603–605), downstream of SL87, which is designated core+1/Short (core+1/S). Here, we provide evidence for SL47 and SL87 possessing a novel cis-acting element that directs the internal translation initiation of core+1/S. Firstly, using a bicistronic dual luciferase reporter system and RNA-transfection experiments, we found that nucleotides 344–596 of the HCV genotype-1a and -2a genomes support translation initiation at the core+1 frame AUG codons 85/87, when present in the sense but not the opposite orientation. Secondly, site-directed mutagenesis combined with an analysis of ribosome–HCV RNA association elucidated that SL47 and SL87 are essential for this alternative translation mechanism. Finally, experiments using cells transfected with JFH1 replicons or infected with virus-like particles showed that core+1/S expression is independent from the 5′UTR IRES and does not utilize the polyprotein initiation codon, but it requires intact SL47 and SL87 structures. Thus, SL47 and SL87, apart from their role in viral polyprotein translation, are necessary elements for mediating the internal translation initiation of the alternative core+1/S ORF.

Building open source audiovisual collections management systems on Samvera

Building and using an open source community-supported system to manage audiovisual materials for a digital archive/library has many advantages and challenges. Being able to dictate the features, have a system speak to specific needs, and have staff on hand that can change or fix problems is certainly appealing. Embarking on this effort with an established open community, such as Samvera, has the advantages of a robust community of developers and service vendors to turn to for help. Managing needed customisations to core code base, keeping track of updates and contributing to the community, however, is challenging. For WGBH, a public television station with a robust 60-year archive, most customisations are due to the use PBCore to structure the metadata of the audio-visual items. This paper focuses on WGBH’s efforts to build a system for its Media Library and Archives based on the Samvera digital repository framework and its Hyrax and Avalon Media System ‘products’.

4차 산업혁명 시대에서 홈 트레이닝 경험에 대한 이론적 패러다임 적용 : 기술수용모델(TAM)을 중심으로

Purpose: The purpose of this study is to understand in depth about participants’ home training experience and technology acceptance in the 4th Industrial Revolution era based on the technology acceptance model (TAM).<BR> Method: This research was conducted through grounded theory. Conclusions were drawn through open coding, axis coding, and selective coding of the grounded theory. In the open coding process, conceptual words are derived, and categorized based on commonalities and differences. In axis coding, a paradigm model is created based on the categorization results. The core code was derived from the codes identified through selective coding and the story outline was constructed. A paradigm model was presented thorough grounded theory and the data was analyzed in depth based on the TAM.<BR> Results: As a result of coding through grounded theory, 124 conceptual words, 23 sub-categories, and 12 upper-categories were derived in the first open coding stage. In the second axis coding stage, a paradigm model was derived. The paradigm model consisted of causal conditions, contextual context, central phenomena, interventional conditions, action-interaction strategies, and outcomes. The key category was the emergence of new technology expectations. In-depth analysis based on TAM showed that many of the techniques combined with home training can be explained by the perceived ease of use of TAM in that they have spent a lot of time planning their own exercise routines and reducing the effort to study or search. In addition, the study participants believed that the usefulness of the innovative technologies combined with home training would improve their performance and quality of life, and these results can be explained through the perceived usefulness of TAM. It was found that this perceived ease of use and perceived usefulness influenced the positive attitude and intention to use new technologies.<BR> Conclusion: Home training participants have positive attitude about their home training experience and acceptance of innovative technologies. As a result, it has been shown that they are actively utilizing innovative home training technologies. In addition, the convergence of home training and technology can be a corner-stone for many people to enjoy sports equally and conveniently.

Genotypic Regulation of Type I Interferon Induction Pathways by Frameshift (F) Proteins of Hepatitis C Virus.

Hepatitis C virus (HCV) has evolved mechanisms to evade innate immunity that are leading to chronic infections. The immunological function of the HCV frameshift (F) protein, which is a frameshift product of core coding sequences, has not been well characterized. The HCV F protein is produced during natural HCV infections and is found most commonly in genotype 1 HCV. In this study, we investigated whether the F protein plays a role in type I interferon (IFN) induction pathways. We engineered F expression constructs from core coding sequences of 4 genotypes (1a, 2a, 3a, and 4a) of HCV as well as the sequences which would only be able to produce core proteins. The peptide lengths and amino acids sequences of F proteins are highly variable. We hypothesized that F proteins from different genotypes might control the type I IFN production and response differently. We found that both IFN-beta (IFN-β) promoter activities are significantly higher in genotype 1a F protein (F1a)-expressing cells. Conversely, the IFN-β promoter activities are lower in genotype 2a F (F2a) protein-expressing cells. We also used real-time PCR to confirm IFN-β mRNA expression levels. By generating chimera F proteins, we discovered that the effects of F proteins were determined by the amino acid sequence 40 to 57 of genotype 1a. The regulation of type I IFN induction pathway is related but not limited to the activity of F1a to interact with proteasome subunits and to disturb the proteasome activity. Further molecular mechanisms of how F proteins from different genotypes of HCV control these pathways differently remain to be investigated.IMPORTANCE Although naturally present in HCV infection patient serum, the virological or immunological functions of the HCV F protein, which is a frameshift product of core coding sequences, remain unclear. Here, we report the effects of the HCV F protein between genotypes and discuss a potential explanation for the differential responses to type I IFN-based therapy among patients infected with different genotypes of HCV. Our study provides one step forward to understanding the host response during HCV infection and new insights for the prediction of the outcome of IFN-based therapy in HCV patients.

A tutorial on deep learning-based data analytics in manufacturing through a welding case study

Abstract Over the past decade, machine learning and deep learning have been increasingly reshaping manufacturing towards smart manufacturing. This paper aims to provide a tutorial for researchers to understand the basic principles of deep learning and its applications to manufacturing, using welding as an example. In this tutorial, we first present an overview of welding processes and the advantages of deep learning in solving welding problems, such as process monitoring and product quality prediction. Then, deep learning characteristics are summarized and two representative deep learning techniques, conventional neural networks (CNNs) and recurrent neural networks (RNNs) that are suitable for image processing and sequential modeling, are discussed. A case study on welding quality prediction that predicts the back-side bead width from top-side images through a CNN is demonstrated, with detailed procedures and core codes from building a CNN to testing the network performance. Prospects for deep learning in a manufacturing context are examined from the authors’ perspective.

Uniform Heated Scaled-Down Standard Fuel Block Test to Validate Core Thermofluid Analysis Code for Prismatic Gas-Cooled Reactor

The Korea Atomic Energy Research Institute (KAERI) has developed the Core Reliable Optimization and thermofluid Network Analysis (CORONA) code for core thermofluid analysis of a prismatic high-temperature gas-cooled reactor (HTGR). KAERI performed scaled-down standard fuel block (SFB) heated tests at a helium experimental loop to validate the CORONA code. The scaled-down SFB was designed based on the core thermofluid design for a 350-MW(thermal) HTGR. The reference test condition was selected to maintain the Reynolds number of the coolant channels and the bypass gaps. The test section had seven coolant holes and 12 fuel holes considering KAERI’s helium loop circulator design. The material of the fuel block was Al2O3, selected to simulate the low thermal conductivity of the irradiated graphite at the high-temperature condition. The bypass gap structure was made of stainless steel 304 to minimize gap size deformation at the heated condition. This paper presents a comparison between the test results and the CORONA analysis results. The test parameter was the nitrogen flow velocity (3.6 to 6.0 kg/min) and constant heated condition.

Wannier90 as a community code: new features and applications

Wannier90 is an open-source computer program for calculating maximally-localised Wannier functions (MLWFs) from a set of Bloch states. It is interfaced to many widely used electronic-structure codes thanks to its independence from the basis sets representing these Bloch states. In the past few years the development of Wannier90 has transitioned to a community-driven model; this has resulted in a number of new developments that have been recently released in Wannier90 v3.0. In this article we describe these new functionalities, that include the implementation of new features for wannierisation and disentanglement (symmetry-adapted Wannier functions, selectively-localised Wannier functions, selected columns of the density matrix) and the ability to calculate new properties (shift currents and Berry-curvature dipole, and a new interface to many-body perturbation theory); performance improvements, including parallelisation of the core code; enhancements in functionality (support for spinor-valued Wannier functions, more accurate methods to interpolate quantities in the Brillouin zone); improved usability (improved plotting routines, integration with high-throughput automation frameworks), as well as the implementation of modern software engineering practices (unit testing, continuous integration, and automatic source-code documentation). These new features, capabilities, and code development model aim to further sustain and expand the community uptake and range of applicability, that nowadays spans complex and accurate dielectric, electronic, magnetic, optical, topological and transport properties of materials.

LearnBlock: A Robot-Agnostic Educational Programming Tool

Education is evolving to prepare students for the current sociotechnical changes. An increasing effort to introduce programming and other STEM-related subjects into the core curriculum of primary and secondary education is taking place around the world. The use of robots stands out among STEM initiatives, since robots are proving to be an engaging tool for learning programming and other STEM-related contents. Block-based programming is the option chosen for most educational robotic platforms. However, many robotics kits include their own software tools, as well as their own set of programming blocks. LearnBlock, a new educational programming tool, is proposed here. Its major novelty is its loosely coupled software architecture which makes it, to the best of our knowledge, the first robot-agnostic educational tool. Robot-agnosticism is provided not only in block code, but also in generated code, unifying the translation from blocks to the final programming language. The set of blocks can be easily extended implementing additional Python functions, without modifying the core code of the tool. Moreover, LearnBlock provides an integrated educational programming environment that facilitates a progressive transition from a visual to a general-purpose programming language. To evaluate LearnBlock and demonstrate that it is platform-agnostic, several tests were conducted. Each of them consists of a program implementing a robot behaviour. The block code of each test can run on several educational robots without changes.

Hybrid Henry Gas Solubility Optimization Algorithm Based on the Harris Hawk Optimization

Henry Gas Solubility Optimization Algorithm (HGSO) is a physical-based algorithm based on Henry's Law and simulates the process that the solubility of gas in liquid changes with temperature. The search strategy of the HGSO is very simple, which results in the algorithm having poor exploitation ability and unable to find a more accurate optimal solution. The Harris Hawk optimization algorithm (HHO) has strong exploitation because of its diverse exploitation strategies. In order to make up for the shortcomings of HGSO algorithm, this article proposes an improved Henry gas solubility optimization algorithm (HHO-HGSO) based on the Harris Hawk optimization. During the iteration, taking the escaping energy function of the HHO algorithm as an indicator and determining a threshold, when the escaping energy function value is greater than the threshold, the algorithm conducts the search strategy of the HGSO algorithm, when less than the threshold, executes four exploitation strategies of Harris Hawk. In order to verify the performance of the proposed algorithm, HHO-HGSO is used to optimize CEC2005 and CEC2017 benchmark test functions and solve 4 real engineering design problems. Marine predators algorithm (MPA), Whale optimization algorithm (WOA), Lightning search algorithm (LSA), Water cycle algorithm (WCA), HGSO, and HHO were used to conduct comparative experiments. The simulation results show that the proposed HHO-HGSO algorithm has strong optimization ability. The core code of the paper has been uploaded in https://github.com/xwaiyy123/HHO-HGSO.

Neutronic study of the 2-MW TRIGA MARK-II research reactor by the deterministic codes DRAGON5 & DONJON5

The main objective of this work is to perform a neutronic study of the 2 MW TRIGA MARK-II research reactor of the National Centre of Sciences, Energy and Nuclear Techniques (CNESTEN), Rabat, Morocco and then validate the results by comparing the experimental values and those published for an ordinary 2 MW TRIGA MARK II research reactor. The core diffusion code DONJON5 and the lattice code DRAGON5 were coupled to perform a full model of the TRIGA core and their consistency and accuracy were established by benchmarking the TRIGA experiments. In this study, the nuclear data libraries ENDF/B-VII.1 and JEFF3.1 based on 172 energy groups were used. The group constants of all the reactor components were generated using DRAGON5 code and the collision probability method. These group constants were used then in the DONJON5 core code to calculate the multiplication factor, core excess reactivity, total and integral control rods worth as well as power peaking factors. Good agreement found between the calculated and measured results.

A new fuel depletion code based on group-wised energy Bateman ODEs; SUDEPLET code for point-like PWR core

A new efficient fuel depletion code for the PWRs (containing less than 5% average enrichment) is developed based on the Bateman five-energy groups coupled ODEs, and is called SUDEPLET (Shiraz University DEPLETion code). In the current article, SUDEPLET physical concepts and its approximation procedures are given, and also the SUDEPLET.exe file are attached and released as user friendly software without any limiting-access. The SUDEPLET execution schemes and inputs are explained, and the SUDEPLET is executed for three different suggested PWRs to find its verification. The SUDEPLET results (which include more than 200 tracked isotopes) are benchmarked with those results using other familiar fuel depletion analyzers such as ORIGEN 2.1 and/or MCNPX 2.6 (based on the BURNS card). Lack of a complicated input is the main improvement of the SUDEPLET code. Actually, thermal power, the BOC uranium isotopes masses, and cycle-time (in terms of days) are the main necessary input parameters for SUDEPLET executions. Another capability of the SUDEPLET is its ability to calculate the whole core average EXCESS reactivity due to uranium depletion of the core, temperature defects, and mixed burnable poisons in the fuel (or BARs). Among hundreds tracked isotopes; Cs-131, Cs-135, Cs-137, Sr-90, Tc-99, I-129 (which are problematic isotopes in the High Level Waste analysis); also Am-241 and Np-237 (which are most important problematic actinides in the HLW analysis) are highlighted and tabulated in the “Principle Window” of the SUDEPLET. Obviously, time-dependent uranium and plutonium isotopes masses are the main elements which are tracked during the cycles.

Fast Service Migration Method Based on Virtual Machine Technology for MEC

In the era of the Internet of Things (IoT), mobile edge computing (MEC) has become an effective solution to meet the energy efficiency and delay requirements of IoT applications. In MEC systems, tasks can be offloaded from lightweight mobile devices to edge nodes that are nearer to the users. To improve user experience, we combine remote loading and redirection to accelerate the service migration. By tracing historic access patterns, the proposed method first generates a loading request list that locates the core codes in the image file of service applications for booting. The core codes are then be prefetched and cached automatically. Furthermore, to avoid the potential UI lagging caused by incomplete service migration, edge nodes can continuously load the remaining codes in the image file. Once the image file is completely migrated, the file will be reconstructed. The running virtual machine (VM) then switches data access to the merged image file. Experiments show that this method can observably reduce the loading time of a VM-based application.

Embracing Multiple Stakeholder Perspectives in Defining Trainee Competence.

To explore how multiple stakeholder groups contribute to the understanding of trainee competence. The authors conducted a constructivist qualitative study in 2015 using focus group discussions to explore the perceptions of different stakeholder groups (patients, nurses/nurse practitioners, supervisors/senior physicians, leaders/administrators, trainees) regarding trainee competence in the emergency department. The authors used a conventional content analysis, a comparative analysis of supervisors'/senior physicians' versus other stakeholders' perspectives, and a directed analysis informed by stakeholder theory to analyze the focus group transcripts. Forty-six individuals participated in nine focus groups. Four categories of competence were identified: Core Clinical Activities, Patient Centeredness, Aligning Resources, and Code of Conduct. Stakeholders generally agreed in their overall expectations regarding trainee competence. Within individual categories, each stakeholder group identified new considerations, details, and conflicts, which were a replication, elaboration, or complication of a previously identified theme. All stakeholders stressed those aspects of trainee competence that were relevant to their work or values. Trainees were less aware of the patient perspective than that of the other stakeholder groups. Considering multiple stakeholder perspectives enriched the description and conceptualization of trainee competence. It also can inform the development of curricula and assessment tools and guide learning about inter- and intradisciplinary conflicts. Further research should explore how trainees' perceptions of value are influenced by their organizational context and, in particular, how trainees adapt their learning goals in response to the divergent demands of key stakeholders.

Development and validation of reactor nuclear design code CORCA-3D

Abstract The advanced node core code CORCA-3D is one of the independent developed codes of NPIC for the nuclear reactor core design. CORCA-3D code can calculate the few-group cross section, solve the 3D diffusion equations, consider the thermal-hydraulic feedback, reconstruct the pin-by-pin power. It has lots of functions such as changing core status calculation, critical searching, control rod value calculation, coefficient calculation and so on. The main theory and functions of CORCA-3D code are introduced and validated with a lot of reactor measured data and the SCIENCE system. Now, CORCA-3D code has been applied in ACP type reactor nuclear cores design.

An Efficiently Decoupled Implicit Method for Complex Natural Gas Pipeline Network Simulation

The simulation of a natural gas pipeline network allows us to predict the behavior of a gas network system under different conditions. Such predictions can be effectively used to guide decisions regarding the design and operation of the real system. The simulation is generally associated with a high computational cost since the pipeline network is becoming more and more complex, as well as large-scale. In our previous study, the Decoupled Implicit Method for Efficient Network Simulation (DIMENS) method was proposed based on the ‘Divide-and-Conquer Approach’ ideal, and its computational speed was obviously high. However, only continuity/momentum Equations of the simple pipeline network composed of pipelines were studied in our previous work. In this paper, the DIMENS method is extended to the continuity/momentum and energy Equations coupled with the complex pipeline network, which includes pipelines and non-pipeline components. The extended DIMENS method can be used to solve more complex engineering problems than before. To extend the DIMENS method, two key issues are addressed in this paper. One is that the non-pipeline components are appropriately solved as the multi-component interconnection nodes; the other is that the procedures of solving the energy Equation are designed based on the gas flow direction in the pipeline. To validate the accuracy and efficiency of the present method, an example of a complex pipeline network is provided. From the result, it can be concluded that the accuracy of the proposed method is equivalent to that of the Stoner Pipeline Simulator (SPS), which includes commercially available simulation core codes, while the efficiency of the present method is over two times higher than that of the SPS.

Microarchitecture-Aware Code Generation for Deep Learning on Single-ISA Heterogeneous Multi-Core Mobile Processors

While single-ISA heterogeneous multi-core processors are widely used in mobile computing, typical code generations optimize the code for a single target core, leaving it less suitable for the other cores in the processor. We present a microarchitecture-aware code generation methodology to mitigate this issue. We first suggest adopting Function-Multi-Versioning (FMV) to execute application codes utilizing a core at full capacity regardless of its microarchitecture. We also propose to add a simple but powerful backend optimization pass in the compiler to further boost the performance of applicable cores. Based on these schemes, we developed an automated flow that analyzes the program and generates multiple versions of hot functions tailored to different microarchitectures. At runtime, the running core chooses an optimal version to maximize computation performance. The measurements confirm that the methodology improves the performance of Cortex-A55 and Cortex-A75 cores in Samsung's next-generation Exynos 9820 processor by 11.2% and 17.9%, respectively, while running TensorFlow Lite.