Search

Researchers face difficulties in finding relevant papers to their research interest as the number of scientific publication is rapidly increasing on the web. Scientific paper recommenders have emerged as a leading solution to help researchers by automatically suggesting relevant and useful publications. Several approaches have been proposed on improving recommender systems. However, most existing approaches depend on priori user profiles, and thus they cannot recommend papers to new user. Furthermore, the existing approaches utilize non-public contextual information, and thus it cannot adequately find similarities between papers due to copyright restrictions. Also, the existing approaches consider only single level paper-citation relation to identify similarities between papers. Considering the above challenges, this paper presents a collaborative filtering based recommendation approach for scientific papers that does not depend on priori user profiles and which utilizes only public contextual information. Using citation context, we utilized 2-level paper-citation relations to find hidden associations between papers. The rational underlying this approach is that, two papers are co-occurred with same cited paper(s) and two papers are co-occurring with same citing paper(s) are significantly similar to some extent. To evaluate the performance of the proposed approach, publicly available datasets are used to conduct extensive experiments. The experimental results demonstrate that the proposed approach has significantly outperforms the baseline approaches in terms of precision, recall, F1, mean average precision, and mean reciprocal rank, which are commonly used information retrieval metrics. The novelty of this study is that, with the proposed approach, researchers are able to find relevant and useful publications over the internet regardless of their previous research experiences and research area.

In paper production, a jumbo reel is cut into multiple intermediate rolls, and each intermediate roll is then sheeted as finished goods. This problem is called a cutting stock problem and is proven to be NP-hard. The objective is to minimize material waste or trim loss from all the cuttings. In the case that any intermediate roll is not entirely used for its associated order, the intermediate roll itself could turn to be a dead stock. We use the concept of universal sizes of intermediate rolls to eliminate the dead stock. A pre-defined number of universal sizes of intermediate rolls is to be used to serve all the orders. The problem is solved using Reinforcement Artificial Bee Colony algorithm with Integer Linear Programming subroutine. This proposed approach is then tested with a set of 1,055 orders and 127 different sizes of sheet papers from a paper manufacturer. The results reveal that our method outperforms other algorithms. Our method offers the total trim loss of 3.51%, compared to the trim loss reported by the industry of at least 5%. This approach not only reduces the number of partially cut rolls, but also decreases the number of the jumbo reels needed to serve all the orders. Therefore, both the inventory cost and material cost can be saved.

The current study revolves around development of paper from dhaincha fiber (Sesbania aculeata). Dhaincha is a basically cultivated for improvisation of soil quality. Due to deforestation worldwide, various sources of raw materials are used in paper industries. Indian pulp and paper industry is the world's major paper industry. In this study extraction of dhaincha fiber was done followed by retting, combing, chemical analysis and by making of pulp with dhaincha fiber. The dhaincha pulp was then used for making a paper from dhaincha fiber. The thickness of the dhaincha paper was then tested with paper thickness tester. By doing further processing dhaincha paper can be used commercially by paper industry.

The NIST selection process for standardizing Post-Quantum Cryptography Mechanisms is currently running. Many papers already studied their theoretical security, but the resistance in deployed device has not been much investigated so far. In particular, fault attack is a serious threat for algorithms implemented in embedded devices. One particularly powerful technique is to use safe-error attacks. Such attacks exploit the fact that a specific fault may or may not lead to a faulty output depending on a secret value. In this paper, we investigate the resistance of various Post-Quantum candidates algorithms against such attacks.

The industry of disposables has undergone a huge revolution in this century owing to the busy lifestyles and technological advancements in its production and manufacturing process. Paper cups are one of the widely used products in disposable industry. It has multi utility options that is it can be used to drink water as well as beverages, soups, liquids etc. This paper entirely focuses on increasing the quality factor of production and manufacturing process of these paper cups by adding additional circuitry to the existing machinery to reduce waste of raw materials as well as the per unit cost involved in wastage of electricity. This research work is based on using proximity sensors to sense the materials and molded paper cups infront of machinery and reduce the electric wastage through a relay Mechanism assembly. This is a PLC based live project example of effectively increasing the overall efficiency of the manufacturing plant

For the detection of product color uniformity and seam texture defects in the production process of test paper, detection methods are proposed respectively. For the detection of the color uniformity, a uniformity detection method based on dynamic light correction is proposed. The illumination correction image is used for color dynamic compensation of the test paper image to solve the influence of uneven illumination on the test paper. For the detection of texture defects such as seams, a texture defect detection method based on gray gradient aggregation intensity is proposed. This method uses one-way Gaussian convolution to strengthen features, and then calculates the gradient, divides the region and calculates the gradient extremum aggregation strength to judge texture defects, which can quickly and efficiently find seam defects. The experimental results show that the above methods have obvious effects on the color uniformity and seam detection of the test paper. The color uniformity accuracy rate is 96%, the seam detection accuracy rate is 97%. The total detection time of a single test paper is about 90ms. The algorithm is real-time and accurate, and can be used in actual production detection.

Electromagnetic Fault Injection (EMFI) is a well known method of introducing faults for security analysis of digital devices. Such faults can be seen as analogous to the faults which are known to naturally occur in digital devices, a known problem with designing safety-critical systems.Numerous standards have been developed for safety-critical systems, including the development of standards for increasing the rate of naturally occurring faults using particle sources. In this work, we demonstrate that desktop EMFI tooling can be used to accomplish similar testing, but with more control, effectively speeding up the evaluation process.We demonstrate using EMFI tooling for safety evaluation to recreate a highly publicized safety issue present in an automotive ECU – one that could not easily be recreated with other techniques.

Oil-paper insulation is the key insulation structure of the transformers, whose aging condition is closely related to the operations of the equipment. The degree of polymerization (DP) is the direct parameter characterizing the aging condition of oil-paper insulation. Recently, the near infrared spectroscopy (NIRS) measurement powered by quantitative analysis is used to evaluate DP of the insulating papers mainly. While the applications of NIRS for DP evaluation is constrained because the present spectral quantitative analysis method is not accurate and stable enough, especially for onsite tests. In this paper, we propose a Gaussian process regression (GPR) method to predict DP of the oil- paper insulation in laboratory as well as in field. Firstly, the basic principles of GPR algorithm are illustrated. A GPR model for DP prediction is established based on the spectra of differently aged insulating paper samples which are prepared in laboratory. The GPR model show a high prediction accuracy both for training set and testing set. The established GPR model is finally applied on the DP prediction of insulating papers originating from a de-tanked transformer. The accurate predicted DP and the reliable aging assessment results indicate that the established model can be implemented on site.

Because the traditional insulation paper state detection method has some shortcomings, this paper proposes a non-destructive detection method, which uses the texture feature value of insulation paper image to establish a relationship with the degree of polymerization (DP) of insulation paper state, and then indirectly estimates the aging state of insulation paper. Firstly, the insulation paper images in different aging states are collected, and the texture feature values of the images are extracted by gray histogram, gray level co-occurrence matrix and gray level run-length matrix; then, the texture feature values are selected by correlation; then, the validity of the selected texture feature values to represent the insulation paper state is verified by support vector machine (SVM); finally, the multiple linear regression method is used. The relationship between the degree of polymerization and the texture feature value is established and verified by the actual results. The results show that the texture feature value has a good fitting relationship with the degree of polymerization, which verifies the effectiveness of this method in evaluating the state of insulating paper.

The ageing status of the oil-paper insulation is closely related to the operation of the transformers. Recently, near infrared spectroscopy (NIRS) is used to identify the ageing condition of oil-paper insulation for it is a more rapid and nondestructive detection than the traditional viscometry method. However, the identification accuracy of existing quantitative analysis method of NIRS are still insufficient. This paper proposed an improved Linear discriminant analysis (LDA) method to classify the spectra and to identify the ageing condition of the insulating paper. Based on DP measurement, the insulating paper specimens prepared in laboratory are divided into varying ageing groups. The principle of LDA method is introduced. The classic LDA method has been applied to identify the ageing status of specimens respectively from 4 ageing groups and 3 ageing groups. The results show that LDA has a better performance in classifying four groups than three groups. To achieve an accurate ageing condition identification with balanced accuracy of the model and enough categories of ageing status, the improved LDA method is proposed. The identification results show that the improved method has satisfactory performance and indicates great application potentials in the ageing assessment of the oil-paper impregnated insulation.

Prediction and detection of skin diseases have always been a challenging and critical task for health care professionals. There are many clinics and skin practitioners who are offering their services at exorbitant costs. At the same time, skin condition prevails in our country in every part. In the current scenario majority of the skincare practitioners are using traditional methods to diagnose the disease which may take a considerable amount of time. Skin Diseases are serious problems in recent times as it is a matter of environmental factors, socioeconomic factors, lack of full diet, etc. This paper is a comparative study about understanding various skin diseases related to normal skin issues as well as cosmetology. A comparison between oily, dry, and normal skin is also projected in this study. A survey of different papers is done on basis of technologies used, results with accuracy, ethical conduct, number of diseases diagnosed, datasets, etc are done.

This paper describes approaches to the use of wearables and related monitoring and diagnostic tools to glean and present data to patients and their caregivers in the form of Personal Health Dashboard. Raw data from devices such as fitness watches, blood glucose and blood pressure monitors can be analyzed by machine learning algorithms ensembles over the cloud or at the edge, and the resulting patterns can be presented to allow for actionable responses. The advent of low-cost computational, sensing, monitoring and diagnostic tools allow for cost-effective solutions that can have wide adaptability. The paper draws from prior and ongoing research conducted by the team to present these positions in the advancement of healthcare.

The economy of the paper industry largely depends on the quality of the paper manufactured and the quality depends on the precise operation of Headbox, which is the key element of the paper machine and is highly responsible for the quality of the paper manufactured. Being a nonlinear multivariable system, a headbox needs to be controlled efficiently to have the desired response which in turn reflects in the quality of the paper. The headbox is subjected to various disturbances and thus it is required not only to suppress the effect of disturbances but also to ensure the smooth start of the headbox and fast response to the set-point changes. A control technique is proposed in this paper for a paper machine headbox based on the determination of ultimate gain and frequency using the fractional-order integrator (FOI). The PID tuning technique proposed has been applied on paper machine headbox and the responses are compared with the existing techniques.

Using a new opinion dynamics using the polarity analysis of measured data of Twitter about "toilet paper" and "COVID-19", we analyzed the chain of fear that people feel for the rumor of lack of toilet paper and COVID-19 itself. We use opinion dynamics for the analysis of the fears, and we obtain the opinion distribution of the fear calculated using the opinion dynamics and the observed distribution. By comparing the two distribution, we found that the opinion distribution for the fear is consisted with the mixture of trust and distrust.

Researchers from academia and the corporate-sector rely on scholarly digital libraries to access articles. Attackers take advantage of innocent users who consider the articles' files safe and thus open PDF-files with little concern. In addition, researchers consider scholarly libraries a reliable, trusted, and untainted corpus of papers. For these reasons, scholarly digital libraries are an attractive-target and inadvertently support the proliferation of cyber-attacks launched via malicious PDF-files. In this study, we present related vulnerabilities and malware distribution approaches that exploit the vulnerabilities of scholarly digital libraries. We evaluated over two-million scholarly papers in the CiteSeerX library and found the library to be contaminated with a surprisingly large number (0.3-2%) of malicious PDF documents (over 55% were crawled from the IPs of US-universities). We developed a two layered detection framework aimed at enhancing the detection of malicious PDF documents, Sec-Lib, which offers a security solution for large digital libraries. Sec-Lib includes a deterministic layer for detecting known malware, and a machine learning based layer for detecting unknown malware. Our evaluation showed that scholarly digital libraries can detect 96.9% of malware with Sec-Lib, while minimizing the number of PDF-files requiring labeling, and thus reducing the manual inspection efforts of security-experts by 98%.

The accelerating rate of scientific publications makes it extremely difficult for researchers to find out the relevant papers and related works. Recommender systems that aim at solving the information overload problem have attracted lots of attention. However, existing paper recommendation works generally rely on the simple citation-ships between papers, which ignore the heterogeneity of the academic graphs. In this paper, we solve the personalized paper recommendation problem in the setting of heterogeneous information networks. A heterogeneous graph representation based recommendation method named HGRec is proposed. First, the author and paper profiles are constructed based on the extracted contents information. Second, we initialize the node vectors by employing the word-embedding technique. Third, we jointly update the node embeddings in the heterogeneous graph by proposing two meta-path based proximity measures. Finally, the paper recommendation is completed by calculating the similarity of the generated author and paper feature vectors. We present experiments on a real academic network, the DBLP network. The comparative results demonstrate the effectiveness of the proposed personalized recommendation approach compared to state-of-the-art methods.

Cryptography is the methodology in which the information is cast into a form that forestalls the abuse of it. In this advanced time of the wireless, it is imperative to guarantee the safe conveyance of the pertinent information which are otherwise exposed to disgusting goals. The procedure of transformation of plaintext to encode text is known as encryption and its converse is known as decryption. The three kinds of cryptography methods, viz., Asymmetric key, Symmetric-key, and Hash Functions. A huge segment of this contemporary method is the key; which is a specific component of extreme significance, as this is what is utilized while scrambling or descrambling the desired information. In this paper, we have compared the two techniques. Incipiently, Particle Swarm Optimization, a strategy of delivering keys to design a stream figure for text encryption. The tale methodology named Particle Swarm Optimization Key Generation Algorithm (PKGA) uses a character code table for encoding the keys. The essential ideal situation of this procedure is that it diminishes the quantity of keys to be fused and coursed. On the other hand is Ant Colony Optimization Techniques to produce keys for the encryption of the information. This entire philosophy depends on the conduct of the ants in their characteristic inquiry of food. The drawback of using Ant Colony Optimization is to downscale the number of keys to be utilized during encipherment when contrasted with different strategies of cryptography in presence. ACO would warrant the generation of such keys with utmost proficiency.

In this work, we describe the application of the ‘steerable pyramid’ (Portilla and Simoncelli, 2000), a high-dimensional texture descriptor, to the discrimination and classification of a large collection of photographic paper textures. This work is part of a broader effort to bring precise physical measurements to the characterization of photographic papers, and our intended audience includes photograph conservators, curators, and collectors. For our purposes, it is important both to enable fine-grained local comparisons of very similar textures and to identify salient groupings across the entire landscape of photographic papers. The steerable pyramid was developed with the explicit goal of synthesizing textures; as such, it is an overcomplete representation. Its high native dimensionality is an advantage for highly precise local comparisons of texture, as in nearest neighbor search; it is, however, a disadvantage for grouping tasks, which require a well-structured global topology. For grouping tasks, we describe a number of compression and grouping approaches designed to meet a variety of analytical needs. Finally, we validate the steerable pyramid descriptor numerically using ‘one versus all’ classification on an augmented data set that includes multiple samples per paper type.

The online monitoring is becoming increasingly important for operators of high voltage assets. One of the most important assets are transformers, especially power transformers, which are mainly insulated with paper together with an insulating liquid for impregnation. An important parameter is the aging condition of the insulating paper, which allows an estimation of the operational safety. This variable is determined by the degree of polymerization (DP-value) and provides information about the mechanical stability of the paper. However, paper samples are currently required for the measurement, which is carried out in a laboratory in accordance with IEC 60450, whereby the paper is destroyed. This study is conducted to determine the DP-value of a paper insulation using a novel, patented optical sensor. The investigation shows that oil-free paper samples have very strong linear correlation between the value measured with the sensor and the DP-value. Therefore, a condition assessment is possible within seconds, as the measurement by means of the optical method is significantly faster and more precise than the conventional viscometric method. It also shows that the assessment is independent of the insulating oil, in which the paper is aged. The sensor could be used for an improved condition assessment of transformers to increase the operational safety.

This paper presents a set of modeling procedures which are substantial for developing a data-driven soft sensor for paper strength parameters in production environment. Special focus is put on describing methods for efficient data aggregation, data cleaning and feature selection, which were employed to appropriately model the inherently complex process of papermaking and handle the challenging nature of collected data. The resulting soft sensor for tensile index deployed at a paper machine shows solid performance. Moreover, good quality of calculated real-time predictions is also confirmed by means of experiment-based validation.

This paper presents the main disadvantages of receipts printed on thermal paper, underlining the impact that BPA (component of thermal paper) has on the environment and human health. The conceptual solution proposes to use the near-field communication (NFC) protocol available in smartphones, integrating a virtual receipt printing device (called e-receipt) at the end of the contactless payment chain.

Paper-based digital microfluidic biochip (PB-DMFB) technology provides a promising solution to many biochemical applications. However, PB-DMFB manufacturing process may suffer from potential security threats. For example, both Trojan insertion and man-in-the-middle attack may affect the functionality of PB-DMFBs. To ensure the correct functionality of PB-DMFBs, we propose a watermarking scheme to hides information in the PB-DMFB layout, which allows users to check design integrity and authenticate the source of the PB-DMFB design. As a result, it serves as a first countermeasure against both Trojan insertion and man-in-the-middle attacks for PB-DMFBs.

The last decade has witnessed an impressive paradigm shift in practically all networking fields: from hardware-based appliances to software-based infrastructures. Network functions traditionally hard-coded in the network devices have been redesigned in software and deployed in virtualized networking scenarios, with impressive advantages in terms of management flexibility and upgradable capability. Still, at a time in which we strive for more and more network processing and computing resources, such a softwarization process risks to bring about a severe toll in terms of performance and energy consumption. We believe, and posit, that programmable hardware architectures tailored for supporting high performance network processing tasks may come to the rescue. With this position paper, after motivating our position, we further briefly review the past and present of programmable hardware for networking, and we discuss its possible evolution towards more comprehensive "higher layer" HW-accelerated processing tasks.

We propose the original automated information technology for transformation of paper archives of drawing and design documentation into electronic 3D model of an object for CALS and BIM ideologies of designing and manufacturing of an object. The technology assumes the full automated cycle of processing of projective kinds from initial paper drawings to electronic frame 3D model of an object. Original models of representation and algorithms of solving the given problem are proposed.

Basic repair and maintenance tasks of industrial machines are usually performed by local service personnel at the location of the component to be maintained. Complex maintenance tasks require an expert who travels to the customer to solve a specific problem. This process is time and cost intensive. Augmented Reality (AR) Remote Assistance offers means to support newcomers on site by guiding personnel through the maintenance process with novel visualizations. This paper describes an Augmented Reality based remote assistance tool for industrial purposes. We present the technology and our implementation approach, as well as the challenges and limitations we have encountered. The potential of the tool was evaluated and analyzed in an industry oriented use case for repairing an industrial PC. The results were compared with a maintenance procedure that was carried out using standard paper repair instructions. Our approach enabled the users to perform complex tasks faster and with fewer errors. Overall, the error rate decreased significantly due to the AR-based remote support. We also found that paper-based instructions are still more efficient in solving simple tasks.

Electronic paper (i.e., e-paper) is a display technology that aims to imitate and substitute the conventional paper. Previous studies of e-paper mainly focus on evaluating or making practical use of its readability. However, there is little research to explore the potential of e-paper on input-oriented applications. In this paper, we introduce a document composition system named SwiftPad for e-paper. Specifically, SwiftPad renovates the famous TEX typesetting system, enabling users to compose high typographic quality documents on e-paper in a WYSIWYG (what you see is what you get), offline-first, and collaborative fashion. Building such a system on resource-constrained e-paper with low screen refresh rate creates unique challenges. In this paper, we identity these challenges and provides workable solutions. We also provide a preliminary evaluation of the new system.

A Particle Swarm Optimisation (PSO) based PID controller has been proposed in this paper for a paper machine headbox. The considered headbox model has been perturbed and the proposed controller has been implemented on the perturbed model to assess the robustness of the system. The robustness of the proposed controller has been evaluated on the basis of sensitivity and complementary sensitivity functions. Sensitivity and complementary sensitivity functions provide significant insight into the system's behavior for disturbance, uncertainties, and noise (due to sensors/actuators). Hence, the controller to be designed for a given system should be concerned about these functions. Apart from these functions, worst-case sensitivity analysis has also been carried out for the perturbed headbox. The responses obtained from the proposed controller have been compared with some conventional PID tuning techniques and it is observed that the proposed controller yields better robustness.

During the almost 27 years of its existence, the business ecosystem research has developed a substantial level of ambiguity and multifacetedness. Due to the technological advancements that promote interconnectedness and value co-creation, the field has consequently spun off into more domain specific branches, such as the arenas of digital business ecosystems or business ecosystems driven by IoT. Nonetheless, despite the efforts to mend the theoretical foundations and close the gap between academia and empirical practice, the absolute majority of IoT driven business ecosystem literature follows the trend of conceptual randomness while expanding the volume of publications exponentially. Therefore, in order to tackle this conceptual inflation, the author encourages other researchers to make extended effort and increase the external validity of their researches, as well as the relevance of the research stream as a whole, by bounding the IoT driven business ecosystems on a rigorous basis and deploying the extant theory in an careful and appropriate manner. Overall, this position paper contributes to the research area of IoT driven business ecosystems by unfolding the importance of focusing on developing conceptual consistency within the discussed emerging field of interest.

Prologue: This paper is written after numerous focus-group meetings with several types of people/experts of many genders. Statements given under are not meant to be insulting any gender, but these are focus-group observations. If you do not like any such statements, we are sorry in that you have missed the points we are trying to convey. This paper addresses various perceptions on genders that concerns abstract entities such as various disciplines of studies. This paper defines genders for disciplines through historical biases and perceptions – e.g., birth, religion, language, region, brought up, to say a few; these were obtained thru’ several focus-group discussions. However, technologies, such as computers, the World Wide Web (WWW), Moodle, sensor networks, Internet of Things (IoT), Virtualization and Immersion technologies are constantly changing the nature of education and learning. As a consequence, and through a series of arguments, we establish that the genders of many popular disciplines are becoming feminine in nature. We also define two terms, namely, Gender Capital and Gender Capitalization, through which respectively one could possibly define and analyze ways and means by which the changing nature of genders of disciplines can be profitably exploited for the global goodness.

The security of software systems relies so heavily on the use of secrets, ranging from credentials, API keys, and tokens to secret keys for cryptographic security. This paper examines approaches for managing and handling secrets in software development life cycle, focusing on mobile applications where software must be distributed to devices and operate in an untrusted environment. This paper points out the shortcomings in activities related to secrets management and handling in mobile application development life cycle and outlines future directions in this area of research.

Microfluidics allows analyzing small volumes of samples, which makes it an ideal technology to build a point-of-care device. Microfluidic paper-based analytical devices (μPADs) have been used on multiple applications. Paper is promising since colorimetric assays could be read with the naked eyed or a cell phone camera. Here we systematically tested several possible readers for bio-assays that use as a base a μPADs biosensors. Specifically, we tested digital imaging of an assay that produces a change of colors, imaging of a fluorescence-based assay, and an automatic scanning based on an in-house Multispectral-image Reconstruction System for colorimetric assays. We used commercial tests, both chemical and enzyme-based tests; as well as an in house test that used the emerging technology of DNA aptamers. Similarly, we evaluated the advantages and disadvantages of each setup and demonstrate how adaptable, fast, and versatile μPADs analysis can be. s. Finally, we suggest points of consideration to develop commercial point-of-care applications based on μPADs.

Currently, the game has been used as learning media in educational field. In developing educational media, it is necessary to match the content and user interface with the target user and match the content with the goals of educational media. The development process is more effective and the chances of success are greater if tested as early as possible. Testing at this design stage uses a paper prototype to get user feedback about the content and the user interface. In this paper, we share experiences where a paper prototype is used to test the suitability of the content and UI before entering the stage of making visual novel games. We identify the problems encountered when applying this method to users, analyze the results, and describe how this problem is solved.

In this paper, we study the state-dependent Gaussian wiretap channel with action encoder and noncausal state at the transmitter, which is also called the action-dependent dirty paper wiretap channel. Two lower bounds on the secrecy capacity of this model are provided, and we find that the two bounds have their advantages and disadvantages, i.e., none of the two always performs better than the other. In addition, we find that taking action on the state helps to increase the achievable secrecy rate of the dirty paper wiretap channel.

The paper presents the research begin modelling and control of the paper drying. The researches have been generated by the investigations of the industrial paper drying plant. The paper is structured in four parts. In first part they are analysed the references of the drying process, modelling process and control process. The second part presents a detailed description of the drying process and industrial paper drying plant. In this part there are analysed the major operating problems of the plant. The third part represents a author's contribution of paper drying process modelling. The last part is destined to design of paper quality control system.

The purpose of this work is to explore the applicability and effectiveness of deep learning methods for the task------moves recognition in abstract of research paper. We firstly build a large corpus for moves recognition. Then we choose the traditional machine learning method SVM as a benchmark, and develop four moves recognition methods based on DNN, LSTM, Attention-BiLSTM and BERT. Finally, we design two groups of experiments with sample size 10,000 and 50,000 and then compare experimental results. The results show that most of the deep learning methods outperform the traditional machine learning method SVM especially in large-scale sample experiments, in which the BERT with a re-pre-trained model achieves the best results in both groups of experiments. Deep learning methods are proved applicable and effective for moves recognition in research paper abstracts.

© 2013 IEEE. With the rapid growth of scientific publications, it is hard for researchers to acquire appropriate papers that meet their expectations. Recommendation system for scientific articles is an essential technology to overcome this problem. In this paper, we propose a novel low-rank and sparse matrix factorization-based paper recommendation (LSMFPRec) method for authors. The proposed method seamlessly combines low-rank and sparse matrix factorization method with fine-grained paper and author affinity matrixes that are extracted from heterogeneous scientific network. Thus, it can effectively alleviate the sparsity and cold start problems that exist in traditional matrix factorization based collaborative filtering methods. Moreover, LSMFPRec can significantly reduce the error propagated from intermediate outputs. In addition, the proposed method essentially captures the low-rank and sparse characteristics that exist in scientific rating activities; therefore, it can generate more reasonable predicted ratings for influential and uninfluential papers. The effectiveness of the proposed LSMFPRec is demonstrated by the recommendation evaluation conducted on the AAN and CiteULike data sets.

With the rapid growth of scientific publications, research paper recommendation which suggests relevant research papers to users can bring great benefits to researchers. In this paper, we focus on the problem of recommending conference papers to the conference attendees. While most of the related existing methods depend on the content-based filtering, we propose a unified conference paper recommendation method named $CPRec$ , which exploits both the contents and the authorship information of the papers. In particular, besides the contents, we exploit the relationships between a user and the authors of a paper for recommendation. In our method, we extract several features for a user-paper pair from the citation network, the coauthor network, and the contents, respectively. In addition, we derive a user’s pairwise preference towards the conference papers from the user’s bookmarked papers in each conference. Furthermore, we employ a pairwise learning to rank model which exploits the pairwise user preference to learn a function that predicts a user’s preference towards a paper based on the extracted features. We conduct a recommendation performance evaluation using real-world data and the experimental results demonstrate the effectiveness of our proposed method.

Machine learning (ML) has increasingly been adopted in safety-critical systems such as autonomous vehicles (AVs) and industrial robotics. In these domains, reliability and safety are important considerations, and hence it is critical to ensure the resilience of ML systems to faults and errors. On the other hand, soft errors are becoming more frequent in commodity computer systems due to the effects of technology scaling and reduced supply voltages. Further, traditional solutions for masking hardware faults such as Triple-Modular Redundancy (TMR) are prohibitively expensive in terms of their energy and performance overheads. Therefore, there is a compelling need to ensure the resilience of ML applications to soft errors on commodity hardware platforms.We first experimentally assess the resilience of safety-critical ML applications to soft errors. We demonstrate through fault injection experiments that even a single bit flip due to a soft error can lead to misclassification in Deep Neural Network (DNN) applications deployed in AVs, leading to safety violations. However, not all the errors in an DNN will result in serve consequences such as safety violations, and hence it is sufficient to protect the DNN from the ones that do. Unfortunately, finding all possible errors that result in safety violations is a very compute intensive task. We propose BinFI, a fault injection approach that efficiently injects critical faults that are highly likely to result in safety violations, based on the unique properties of DNNs. Finally, we propose Ranger, an approach to protect DNNs from critical faults with minimal performance overheads and no accuracy loss. We will conclude by presenting some of our ongoing work, and the future challenges in this area.

The exchange of academic marks between HEIs (Higher Education Institutions) is mandatory in every student mobility programs (i.e. the EU Erasmus Program) but that process remains to present date with insufficient technological support and the absence of a comprehensive reference model that allows the integration of potential technological solutions for the exchange of academic data with existing Academic Information Systems seems to limit greatly the possibility of adopting solutions of this type referred to in the existing literature. This work addresses this issue, conducting an initial bibliographic review aimed at the identification of the fundamental requirements of such an architecture as well as explores some of the technologies that are showing potential for usage in the safe exchange of academic results between partner HEIs, with particular interest in blockchain technology applied in an educational context. info:eu-repo/semantics/publishedVersion

Objective: We propose a MATLAB-based tool to convert electrocardiography (ECG) waveforms from paper-based ECG records into digitized ECG signals that is vendor-agnostic. The tool is packaged as an open source standalone graphical user interface (GUI) based application. Methods and procedures: To reach this objective we: (1) preprocess the ECG records, which includes skew correction, background grid removal and linear filtering; (2) segment ECG signals using Connected Components Analysis (CCA); (3) implement Optical Character Recognition (OCR) for removal of overlapping ECG lead characters and for interfacing of patients’ demographic information with their research records or their electronic medical record (EMR). The ECG digitization results are validated through a reader study where clinically salient features, such as intervals of QRST complex, between the paper ECG records and the digitized ECG records are compared. Results: Comparison of clinically important features between the paper-based ECG records and the digitized ECG signals, reveals intra- and inter-observer correlations of 0.86–0.99 and 0.79–0.94, respectively. The kappa statistic was found to average at 0.86 and 0.72 for intra- and inter-observer correlations, respectively. Conclusion: The clinically salient features of the ECG waveforms such as the intervals of QRST complex, are preserved during the digitization procedure. Clinical and Healthcare Impact: This open-source digitization tool can be used as a research resource to digitize paper ECG records thereby enabling development of new prediction algorithms to risk stratify individuals with cardiovascular disease, and/or allow for development of ECG-based cardiovascular diagnoses relying upon automated digital algorithms.

This paper describes a user identification method conducted across online social networks (OSNs) using information regarding friends’ locations, in contrast to a conventional method based on the similarity of two display names. This latter method encounters the problem of decreased identification accuracy if a user registers different display names across OSNs. Our proposed method aims at addressing this issue by utilizing information on friends’ locations. We convert location information extracted from an OSN to that of such geographic administrative units as country, state and city, and calculate the weighted occurrence frequency of a location pair on the basis of distance between the pair. This is based on the hypothesis that a friend list relatively rarely includes an account pair whose locations are extremely distant from one another. We calculate a similarity score between two accounts’ weighted occurrence frequencies for each geographic administrative unit. Finally, we identify a user on the basis of individual similarity scores or weighted average scores for a given geographic administrative unit. Evaluation experiments show that our proposed method improves performance over that of a conventional method (accuracy: 83.1%$\rightarrow$91.2%).

International audience; Internet of Things (IoT) is a network where several and various interconnected, smart objects continuously generate alarge amount of data transmitted over Internet. Several efforts and research works in IoT focused on hardware constraints, lowcost, application design, as well as good communication technologies. However, due to the intrinsic characteristics of diversity,heterogeneity, large-scale, dynamic and large volume of data generated and various IoT applications needs, using traditionaldatabase management mechanisms and analytics architectures is not generally suitable. Then, it is challenging to provideefficient IoT data storage and query processing mechanisms for satisfying IoT application needs. This paper identifies the mainspecifications and mechanisms of data and query management for IoT and classifies them. Moreover, this work presentsadvances on data and query management mechanisms on IoT, shows their advantages and limits and discusses the challengingopen research issues that need to be addressed for providing guidelines for further contributions.

Several applications can benefit from recording information about the places a mobile entity visits and the length of time it spends there (e.g., shoppers, employees, buses, movable equipment, autonomous robots, etc.). In this paper, we present our approach to recording spatio-temporal information in a secure and inviolable way using a Distributed Ledger Technology. So far in this work, we observe that, although sending data to the blockchain is a heavy process, there are ways that this transaction time does not disrupt the flow of new data such as adding these data to a queue. However, this makes the sending process overall longer. We intend to continue this study by testing different blockchain services and comparing their performances.

UmpleOnline is a widely used modeling tool that allows practitioners to create class diagrams, state diagrams and several other model types textually, with the option to edit class diagrams graphically. In this paper we propose how UmpleOnline can be used as a tool to conduct modeling empirical studies. UmpleOnline would be instrumented to allow any of its 180,000 annual visitors to volunteer to participate in one of many studies that would be pre-loaded by any member of the modeling research community. Studies could focus on research questions relating to model quality, modeling strategies, modeling errors or many other things. UmpleOnline would be an ideal vehicle for such studies since it is zero-footprint and has demonstrated high usability.

The problem of joint information transmission and input signalling estimation over a state-dependent channel is considered. The general single-letter upper and lower bounds of the optimal capacity-distortion trade-offs are provided. For the Gaussian dirty paper channel, we calculate the bounds by choosing parameters carefully, and obtain a computable form eventually. In such a way, we obtain a complete characterization for this problem setting.

Our proposed method focuses on offline friends who are acquainted with a target user in the real world. We use Density-Based Spatial Clustering of Applications with Noise (DBSCAN) to distinguish between online friends of a target user who are also offline friends and online friends who are acquainted with the target user only in cyberspace. By utilizing the probability of friendship as a function of the geographical distance between users’ home locations, our proposed method achieves an estimation accuracy 1.7 percentage point better than that of a conventional method. We have also been able to verify the effectiveness of an optimization method that narrows down candidate areas for a target user’s home location.

This work is centered on the paper making process in an industrial paper mill. There exist plenty of sub-processes in a paper mill for transformation of raw material to finished product. A particular process that is taken up in this research is paper basis weight control process, which is very important for good quality paper production. But the controller design for this process is very challenging due to the involvement of severe dead time value, which in addition to delaying the control output, makes the control system unstable. So, two different approaches are used here, in MATLAB, for controller design, one is the dead time approximation using Pade approximation technique and the other is dead time compensation using Smith predictor technique. Hence, the performance of the designed control systems is compared.