Search

To improve the accuracy of user implicit rating prediction, we combine the traditional latent factor model (LFM) and bidirectional gated recurrent unit neural network (BiGRU) model to propose a hybrid model that deeply mines the latent semantics in the unstructured content of the text and generates a more accurate rating matrix. First, we utilize the user’s historical behavior (favorites records) to build a user rating matrix and decompose the matrix to obtain the latent factor vectors of users and literature. We also apply the BERT model for word embedding of the research papers to obtain the sequence of word vectors. Then, we apply the BiGRU with the user attention mechanism to mine the research paper textual content and to generate the new literature latent feature vectors that are used to replace the original literature latent factor vectors decomposed from the rating matrix. Finally, a new rating matrix is generated to obtain users’ ratings of noninteractive research papers and to generate the recommendation list according to the user latent factor vector. We design experiments on the real datasets and verify that the research paper recommendation model is superior to traditional recommendation models in terms of precision, recall, F1-value, coverage, popularity and diversity.

Machine learning (ML) components are increasingly adopted in many automated systems. Their ability to learn and work with novel input/incomplete knowledge and their generalization capabilities make them highly desirable solutions for complex problems. This has motivated the inclusion of ML techniques/components in products for many industrial domains including automotive systems. Such systems are safety-critical systems since their failure may cause death or injury to humans. Therefore, their safety must be ensured before they are used in their operational environment. However, existing safety standards and Verification and Validation (V&V) techniques do not properly address the special characteristics of ML-based components such as non-determinism, non-transparency, instability. This position paper presents the authors' view on the safety of automotive systems incorporating ML-based components, and it is intended to motivate and sketch a research agenda for extending a safety standard, namely ISO 26262, to address challenges posed by incorporating ML-based components in automotive systems.

In modern academic social network, it is very difficult for scholars to find academic papers consistent with their research direction. Time is a critical factor in paper recommendation. As time goes on, the impact of an academic paper would gradually fade. Likewise, the research interests of users may also change. Therefore, we propose a temporal perceptual neural collaborative filtering model that integrates the multi-dimensional features of papers. We conducted our experiments on the dataset from CiteULike, comparing the recommended results by using four time-decay functions and evaluating our model with multiple evaluation indicators. The satisfactory results show that our model is effective in filtering out the expired papers by considering the characteristics of papers and the changes of scholars' interests.

This paper is review on image fusion for different technique i.e. Non-subsampled Contourlet Transform (NSCT) and Neural Network (NN). NSCT is the pyramid structure and gives the multiscale feature of the image. NSCT used to directional filtering structure and used to filter bank. Neural network is a progression of calculations that tries to perceive fundamental connections in a bunch of information through an interaction that impersonates the manner in which the human mind works. The implemented algorithm will applied to MRI image and with the help of NN to calculated accuracy. The previous algorithm was work of different technique i.e. principal component analysis, discrete wavelet transorm and curvelet transform. The implemented algorithm will implement MATLAB software and calculate mean square error and peak signal to noise ratio.

Most existing studies and public datasets for handwritten Chinese text recognition are based on the regular documents with clean and blank background, lacking research reports for handwritten text recognition on challenging areas such as educational documents and financial bills. In this paper, we focus on examination paper text recognition and construct a challenging dataset named examination paper text (SCUT-EPT) dataset, which contains 50 000 text line images (40 000 for training and 10 000 for testing) selected from the examination papers of 2 986 volunteers. The proposed SCUT-EPT dataset presents numerous novel challenges, including character erasure, text line supplement, character/phrase switching, noised background, nonuniform word size, and unbalanced text length. In our experiments, the current advanced text recognition methods, such as convolutional recurrent neural network (CRNN) exhibits poor performance on the proposed SCUT-EPT dataset, proving the challenge and significance of the dataset. Nevertheless, through visualizing and error analysis, we observe that humans can avoid vast majority of the error predictions, which reveal the limitations and drawbacks of the current methods for handwritten Chinese text recognition (HCTR). Finally, three popular sequence transcription methods, connectionist temporal classification (CTC), attention mechanism, and cascaded attention-CTC are investigated for HCTR problem. It is interesting to observe that although the attention mechanism has been proved to be very effective in English scene text recognition, its performance is far inferior to the CTC method in the case of HCTR with large-scale character set.

Nowadays, there is a rapidly increasing number of conferences and journals in computer science that make a lot of challenges for researchers to find an appropriate venue to submit their scientific work. There is a need for a recommendation system that can support researchers for a better process of paper submission. In this paper, we present an efficient approach for building such a recommendation model by using embedding methods, Global Vector (GloVe) 1 created by Pennington et al. [1] and FastText 2 proposed by Facebook [2], Convolutional Neural Network (CNN) [3], and LSTM. The experimental results show that the combination of CNNs and FastText, CNN + FastText, can achieve the best performance in terms of the Top 1 Accuracy compared with other techniques, including the S2RSCS model, as presented in [4]. Moreover, the performance by using GloVe or FastText is much better, faster, and more stable than S2RSCS in most cases.

1 The aim of the present work is to assess similarities between papers used in a collection of 215 lithographs by Henri Matisse. It combines the use of an anisotropic multiscale analysis built upon the hyperbolic wavelet transform to compute distances between textures, with the fact that four textures of the same print were digitized, corresponding to four different (recto/verso and top/side) views. A permutation algorithm is devised showing the importance of reordering the four views to achieve a meaningful comparison across the papers, thus suggesting possibly different artistic intentions in some prints.

Which are the segmentation algorithms proposed during 2018-2019 in CVPR that have CNN architecture?’ Answering this question involves identifying and analyzing the deep learning architecture diagrams from several research papers. Retrieving such information poses significant challenge as most of the existing academic search engines are primarily based on only the text content. In this paper, we introduce Diag2Graph, an end-to-end framework for parsing deep learning diagram-figures, that enables powerful search and retrieval of architectural details in research papers. Our proposed approach automatically localizes figures from research papers, classifies them, and analyses the content of the diagram-figures. The key steps in analyzing the Figure content is the extraction of the different components data and finding their structural relation. Finally, the extracted components and their relations are represented in the form of a deep knowledge graph. A thorough evaluation on a real-word annotated dataset has been done to demonstrate the efficacy of our approach.

In order to improve the security of medical data, this paper uses a robust and secure dual watermarking technique in RDWT-SVD domain. Dual watermarking is achieved by embedding text watermark in medical watermark image using DWT, generating the final watermark. Before embedding, turbo code is used to encode the text watermark. Further, trade-off between imperceptibility and robustness is controlled in our scheme through firefly optimization. Experimental results indicate the proposed method offers high imperceptibility and robustness, and have superior overall performance.

In many fields, a mass of algorithms with completely different hyperparameters have been developed to address the same type of problems. Choosing the algorithm and hyperparameter setting correctly can promote the overall performance greatly, but users often fail to do so due to the absence of knowledge. How to help users to effectively and quickly select the suitable algorithm and hyperparameter settings for the given task instance is an important research topic nowadays, which is known as the CASH problem. In this paper, we design the Auto-Model approach, which makes full use of known information in the related research paper and introduces hyperparameter optimization techniques, to solve the CASH problem effectively. Auto-Model tremendously reduces the cost of algorithm implementations and hyperparameter configuration space, and thus capable of dealing with the CASH problem efficiently and easily. To demonstrate the benefit of Auto-Model, we compare it with classical Auto-Weka approach. The experimental results show that our proposed approach can provide superior results and achieves better performance in a short time., 12 pages, 3 figures

Sentiment Analysis is done by gathering the data, analyzing the text and aggregating the words in the text. The main focus of this survey is on the evaluation methods used and their respective core solutions. In this paper we are going to study the aspect based sentiment classification, which is classified into three. Those are Implicit, Explicit and Coreferential aspect based sentiment classification. In this paper we are also going to elaborate the polarity of the reviews into positive and negative sentiments. Aspect Based Classification not only promises to suppress the noisy data but also provides high efficiency in identifying the aspects with an addition of performing classification task with higher accuracy. This accuracy can be achieved by using Neural Network algorithms.

The main interest of this paper is concept of Compressive Sensing and its application in 1D biomedical signals. The accent is on analyzing several types of biomedical signals and finding domain in which they are sparse. Therefore, we use gradient-based algorithm to reconstruct these biomedical signals with different number of samples. By working with gradient-based algorithm, we presented reconstruction of Electrocardiogram, Electroencephalogram and Electromyogram biomedical signals and drew a conclusion about differences between them. Several sparsifying basis are tested in order to find the most suitable one, for specific signal type, and we verified the whole theory by experimental results.

The study aimed to realize the automatic production scheduling of waste paper in paper mills. Based on the quantities of the field data (mixing ratios of waste paper and pulp properties) stored in a paper mill, the methods of back propagation neural networks (BP-NN) and support vector machine (SVM), the all data set and the average data set, were firstly applied to develop the prediction models of the pulp properties; and then the genetic algorithm was used to search for the suitable scheduling solutions. The simulation results revealed that: demonstrating its acceptable prediction accuracy and fast training time, the prediction model with SVM method and the average data set was good enough to be used in the genetic algorithm to search for the optimal scheduling ratios of waste paper, the obtained optimal scheduling solution not only had the lowest purchase cost of waste paper, but also met the requirements of the mill: specified types of waste paper, specified ratio of some kind of waste paper and the required pulp brightness.

A secure medical image watermarking framework using SVD, DE, step space filling curve and IntDCT is proposed. The proposed algorithm starts with partition of the host medical image into 8×8 blocks and then IntDCT is applied in each block. By collecting integer DC coefficients, a feeble resolution medical image is obtained. SVD is applied on this feeble resolution image. For maintaining a strong relationship between imperceptibility and robustness, the popular differential evolution (DE) technique is chosen to examine the most suitable scaling factors. The singular values of the host image are modified with the singular values of the watermark to insert the watermark in the host image. Here, DE is playing a very important role to determine the most suitable scaling factors to the embedded process for maintaining the dilemma between robustness and imperceptibility without degrading the nobility of medical images. To increase the security of the watermark image, first it is scuffled with the help of step space filling curve before the embedding process. The experimental observations and results proved the efficiency of the proposed framework. It maintains the quality of the watermarked images and the watermark can be extracted from the seriously malformed images.

Steganography is the art of hiding a secret message inside a publicly visible carrier message. Ideally, it is done without modifying the carrier, and with minimal loss of information in the secret message. Recently, various deep learning based approaches to steganography have been applied to different message types. We propose a deep learning based technique to hide a source RGB image message inside finite length speech segments without perceptual loss. To achieve this, we train three neural networks; an encoding network to hide the message in the carrier, a decoding network to reconstruct the message from the carrier and an additional image enhancer network to further improve the reconstructed message. We also discuss future improvements to the algorithm proposed.

In recent times, Generative Adversarial Networks (GANs) have created a lot of buzz in the research community. GANs are formulated on the zero-sum game theory, where two neural nets compete against each other. The resultant deep model is capable of generating data similar to any data distribution provided. It utilizes the adversarial learning approach and is far more capable in learning features than the traditional machine learning models. This review focusses on the origin and evolution of GANs. Firstly, the traditional GAN is explored in terms of its structure and loss functions. Then come the common challenges of training GANs. Thirdly, the review dives into numerous GAN variants and explains their improvements. The review then lists the wide variety of applications and ends with the conclusion.

We present a mathematical technique for using Hermite functions to reconstruct various types of images. Even though the functions are not orthogonal, they can be used in image processing. The functions permit us to get good localization of the signals and it is shown that we get better reconstructed image quality for a larger number of Hermite functions used. Finally, we use a Hermit function frame to decompose and reconstruct a few types of images and compare the results for various number of Hermit functions.

Paraphrase extraction is the method of collecting words, phrases or sentences that have same meaning from free text or structured documents. Candidate monolingual pararel corpus was build from scientific papers by clustering techniques. The corpus contain couples of citation sentences that mention same index. The corpus evaluted using weighting techniques and proximity measurements based on the density of each sentence cluster. The results of this reserach are expected to simplify paraphrase labeling by language and computer science experts.

Fully convolutional neural networks have shown remarkable success in performing semantic segmentation. The use of convolutional layers for the entire architecture and skip connections to combine different resolution features or predictions have been adopted in successful networks, such as U-Net and DenseNet. However, these models employ several max-pooling layers that cause the network to lose spatial information and require them to mimic an autoencoder architecture to perform semantic segmentation at the original input resolution. In this paper, we propose a network that extracts features automatically with convolutional layers, like the fully convolutional neural network, but retains the spatial information of each of the extracted features. It then utilises the extracted features to make predictions with an efficient upsampling method. We evaluate the network performance on a liver segmentation task where it performs with comparable accuracy to other state-of-the-art networks while being much smaller in terms of the number of parameters as well as faster in computation time.

The problem of classifying a research article into one or more fields of science is of particular importance for the academic search engines and digital libraries. A robust classification algorithm offers the users a wide variety of useful tools, such as the refinement of their search results, the browsing of articles by category, the recommendation of other similar articles, etc. In the current literature we encounter approaches which attempt to address this problem without taking into consideration important parameters such as the previous history of the authors and the categorization of the scientific journals which publish the articles. In addition, the existing works overlook the huge volume of the involved academic data. In this paper, we expand an existing effective algorithm for research articles classification, and we parallelize it on Apache Spark –a parallelization framework which is capable of sharing large amounts of data into the main memory of the nodes of a cluster– to enable the processing of large academic datasets. Furthermore, we present data manipulation methodologies which are useful not only for this particular problem, but also for most parallel machine learning approaches. In our experimental evaluation, we demonstrate that our proposed algorithm is considerably more accurate than the supervised learning approaches implemented within the machine learning library of Spark, whereas it outperforms them in terms of execution speed by a significant margin.

With today’s eye-tracking technologies, it is possible to analyze cognitive processes such as reading. There are many studies on that topic. We have studied some of these studies that analyze eye movements recording during reading by various machine learning algorithms.

Chemical fiber paper tubes are the essential spinning equipment on filament high-speed spinning and winding machine of the chemical fiber industry. The precision of its application directly impacts on the formation of the silk, determines the cost of the spinning industry. Due to the accuracy of its application requirements, the paper tubes with defects must be detected and removed. Traditional industrial defect detection methods are usually carried out using the target operator's characteristics, only to obtain surface information, not only the detection efficiency and accuracy is difficult to improve, due to human judgment, it's difficult to give effective algorithm for some targets. And the existing learning algorithms are also difficult to use the deep features, so they can not get good results. Based on the Faster-RCNN method in depth learning, this paper extracts the deep features of the defective target by Convolutional Neural Network (CNN), which effectively solves the internal joint defects that the traditional algorithm can not effectively detect. As to the external joints and damaged flaws that the traditional algorithm can detect, this algorithm has better results, the experimental accuracy rate can be raised up to 98.00%. At the same time, it can be applied to a variety of lighting conditions, reducing the pretreatment steps and improving efficiency. The experimental results show that the method is effective and worthy of further research.

The process of modifying images to achieve improved resultant images is called image enhancement. Though there are some researches done on image enhancement of Bangla document images but research on enhancement of newsprint Bangla document images is almost an untouched field of research. So for enhancing Bangla newsprint document images we have divided our whole working procedure into three phases. At the first phase, we have tried to explore the main reason behind the failure of well known image enhancement techniques on newsprint papers by running out analytic experimentation. In second phase we proposed a tentative framework by applying commonly used mathematical morphology method in a modified way. But due to huge diversity of Bangla letters this approach also did not provide the expected output. Finally, in the third phase we have proposed a framework addressing the issues of the second phase, where we have used bitmap color substitution method along with the well known Sauvola's algorithm. To enhance the result more we have then modified Sauvola's algorithm to fill missing edges along the horizontal and vertical lines.

A large amount of information is continuously generated in intensive health care. An analysis of these data streams can supply valuable insights to improve the monitoring of the patients. The volume, frequency and complexity of data, which come unlabeled, make their analysis a challenging task. Machine learning (ML) techniques have been successfully employed for mining data streams to extract useful knowledge for health care monitoring. It includes the detection of changes in the behavior of sensors, failures on machines or systems, and data anomalies. Anomaly (or outlier) detection is a ML task that aims to find exceptions or abnormalities in a dataset. These exceptions, in a medical context, can represent a new disease pattern, an event to be further investigated, behavior changes or potential health complications. Despite of its analysis in data streams is a challenging task, temporal abstractions techniques should help due to they deal with the management and abstraction of time based data, offering high level of visualization of each data object in its context. The aim of this paper is to review recent research in anomaly detection and temporal abstractions and discuss the application of their combination to intensive care data streams.

Intelligent computing and maintenance-free powering are two desirable characteristics of wearable IoT devices. Energy harvesting nonvolatile intelligent processor (NIP) with neural network computation capability has the potential to advance these goals. Individual ultraviolet (UV) exposure monitoring progressively becomes one conspicuous application of wearable devices. In resource constrained wearable sensor nodes, we can alleviate the data transmission burden via convolutional neural networks (CNNs) based pattern recognition. Nevertheless, in spite of the substantially improved computing capability of NIP, typically computational and memory intensive CNNs are still too bulky for on-node implementation. We develop an CNN-based pattern recognition system for nonvolatile IoT platform for smart UV monitoring, and propose a optimization method to achieve extremely tiny and efficient CNNs. Experimental results show that the offline-trained CNN can recognize individual UV exposure patterns with accuracy of 85%, and the simplified on-node CNN can achieve 93.2% parameters reduction with only 5% accuracy loss.

In this article, the observability of the Boolean multiplex(multiplevel) networks (BMNs) is studied. First, the dynamical model and structure of BMNs are constructed. By using of Semi-Tensor Product (STP) approach, the logical dynamics of BMNs is converted into an equivalent algebraic representation. Then, the observability of the BMNs is investigated by giving necessary and sufficient conditions. Finally, an example is given to illustrate the efficiency of the obtained theoretical results.

These days digital image forgery has turned out to be unsophisticated because of capable PCs, propelled image editing softwares and high resolution capturing gadgets. Checking the respectability of pictures and identifying hints of altering without requiring additional pre-embedded information of the picture or pre-installed watermarks are essential examine field. An endeavor is prepared to review the current improvements in the research area of advanced picture fraud detection and comprehensive reference index has been exhibited on passive methods for forgery identification. Passive techniques donot require pre-embedded information in the image. Several image forgery detection techniques are arranged first and after that their summed up organization is produced. Author will review the various image forgery detection techniques along with their results and also compare the various different techniques based on their accuracy.

International audience; In this paper, we investigate the retrieval of paper watermark by visual similarity. We propose to perform the visual similarity by encoding small regions of the watermark using a non-negative dictionary optimized on a large collection of watermarks. The local codes are then aggregated into a single vector representing the whole watermark. Experiments are carried out on a test of tracings (manual binarization of watermarks).

This paper discusses the identification of extend relation in scientific papers based on supervised machine learning. Identification of extend relations is conducted by classifying each sentence in scientific papers into extend category. Extend relation is one type of papers' relations that obtained by using the citation context based approach. Citation context is a set of words or phrases in a sentence collection that citing or discussing other papers. Citation context based approach use the semantic of citation sentence to identify the relationship between scientific papers and can identify more varied relationship than two other approaches i.e. content-based and citation analysis. The recently research in papers' relations used a rule-based approach to identify extend relations. In this paper, supervised learning approached with proposed features set was used to identify extend relations. The learning of classifier model is explored by using Naive Bayes, Decision Tree, Ibk and Logistic Regression. Experimental results show that the performance of extend sentence classification based on supervised machine learning with proposed features is superior compared to the baseline. Feature selection based on correlation value is also effective to improve the performance of the extend sentence classification.

In this paper, we focus on a human hand recognition system for human-robot interaction. The motion of the human hand is so dynamic that the hand needs to be tracked at high speed, and non-blurred images are required to achieve stable and correct recognition. Moreover, a wide recognition area is necessary for natural interaction. To achieve these aims, we constructed a new active sensing system that can actively track and recognize the human hand by using a high-speed vision system. An application that clearly requires high-speed recognition and actuation is the rock-paper-scissors game between a human and a robot, and we focused on this task to demonstrate the proposed system. In this situation, we track a human hand at high speed and detect its sign (rock, paper, or scissors) correctly in every frame. Our experiments showed that the robot hand formed its sign to beat the human opponent in natural interaction thanks to its high speed and wide recognition area.

Drivers with a fatigue condition have always become basic problem for many accidents. Driver supposes required to identify the object in front of the vehicle and need to understand the causes of accidents and trouble in vehicle control. Now a day with the increasing traffic conditions, possibility for accidents is increased in a large space because driver is unable to detect themselves the type of fatigue condition they are suffering while driving. So the reason we are going to develop the system which always monitors driver's to avoid accidents in real time. This paper describes dissimilar methods existing for detecting the crisis of driver vigilance and then with comparison of all also chooses best system out of those accessible. As of comparison with all existing methods we have implemented a device which always monitors driver vigilance in real-time using computer vision system which consist of main components as, Raspberry Pi which is specially designed for hardware platform, video camera, and the alarm system for alerting.

In this Plenary paper the most important scientific challenge of knowledge learning is reviewed thought two different approaches: Deep Learning (DL) and Wise Learning (WL).

The effective bandwidth management in multi-service computer networks such as university networks has become a challenge in recent years. The growth of internet traffic and limitation of bandwidth resources persuade the information technology (IT) managers to focus on effective bandwidth allocation policies. One of the important issues discussed in this domain is how to assign the bandwidth fairly based on the priority levels. In this paper, focusing on the "priority-based bandwidth allocation", a hybrid data mining method is developed to manage the limited bandwidth in a university network more effectively. This method is composed of two main steps and uses the clustering and classification techniques. The main purpose is to detect, analyze and predict students' behavioral patterns in a university network and identify the main factors that affect their tendency in using internet. The proposed method is applied on a real data of a network university. The results indicate that "degree level" and "age" are the most important factors that influence students' tendency to use internet. The results would be also useful for prediction purposes. It helps the IT managers to predict a new student's tendency to use internet given his/her characteristics. By analyzing the results, the IT managers can make better decisions to optimize the allocation of bandwidth resources.

We design and study ExamParser, an innovative intelligent semantic automatic indexing method, for orchestrating today's programming classes. ExamParser automatically processes paper-based exams by associating sets of concepts to the exam questions, which provide graders semantic grading guidelines and leave personalized semantic feedback. Results showed that the ExamPraser significantly extract more and diverse concepts from exams. It also achieves high coherence within exam, indicating the automatic concept extraction from exams is promising and could be a potential technological solution to provide personalized feedback for large-size programming classes.

In this paper, a framework for human hand movement recognition is implemented for humans to play rock-paper-scissors game interactively against computer using a webcam. The game is the classic rock-paper-scissors game; however, first player is human and second player is computer. Human hand movements are recognized using a webcam, and the game is fully controlled using hand gestures. A dataset is constructed in this work by collecting hand images of different people. The dataset contains rock, paper, and scissors movements of different humanbeings. There are 8 photos for each movement and person using their left and right hands. The dataset is constructed using a controlled environment with black background and close up hand images. To recognize player movements two different feature extraction methods and two different classification methods are used and compared. Histogram of oriented gradients feature extraction method combined with support vector machines are found to be effective. It presents fast and effective computer vision framework for seamless playing experience with a sensitivity of %92.9.

Falsification of medicines is a big problem in many developing countries, where technological infrastructure is inadequate to detect these harmful products. We have developed a set of inexpensive paper cards, called Paper Analytical Devices (PADs), which can efficiently classify drugs based on their chemical composition, as a potential solution to the problem. These cards have different reagents embedded in them which produce a set of distinctive color descriptors upon reacting with the chemical compounds that constitute pharmaceutical dosage forms. If a falsified version of the medicine lacks the active ingredient or includes substitute fillers, the difference in color is perceivable by humans. However, reading the cards with accuracy takes training and practice, which may hamper their scaling and implementation in low resource settings. To deal with this, we have developed an automatic visual recognition system to read the results from the PAD images. At first, the optimal set of reagents was found by running singular value decomposition on the intensity values of the color tones in the card images. A dataset of cards embedded with these reagents is produced to generate the most distinctive results for a set of 26 different active pharmaceutical ingredients (APIs) and excipients. Then, we train two popular convolutional neural network (CNN) models, with the card images. We also extract some "hand-crafted" features from the images and train a nearest neighbor classifier and a non-linear support vector machine with them. On testing, higher-level features performed much better in accurately classifying the PAD images, with the CNN models reaching the highest average accuracy of over 94\%., Comment: in Proc. IEEE Winter Conference on Applications of Computer Vision (WACV), 2016

Monetary transactions are integral part of our day to day activities, so currency recognition has become one of the active research area at present and it has vast potential applications. In this paper we introduced a system to recognize and classify four different currencies using computer vision. As there are 200+different currencies used in different countries around the world. The technology of currency recognition aims to search and extract the visible as well as hidden marks on paper currency for efficient classification. The features are extracted based on color, texture and shape for four different currencies and they are classified using Artificial Neural network.

There is currently a big demand for automating big data analysis. In the data analysis field, data abstraction or summarization playes an important role in the extraction of generalized information from large scale data. We developped an artificial intelligence computer system with the aim of automating big data analysis and came up with a method that can abstract numerical type data (age, height, time, etc.). However, it could not abstract or summarize label type data (customer ID, product code, name, etc.). In the present work, we have developed a label abstraction method based on information entropy. Experiments using open real data showed that the proposed method achieved an extraction accuracy of 80% evaluated by f measure. We intended to apply the proposed method to our artificial intelligence and perform further evaluations.

Every second, Fraud Detection Systems (FDS) check streams of thousands of credit or debit card transactions. Most of the models in production and in the literature rely on batch learning, wasting part of the most recent data. Incremental learning may be beneficial in terms of compu-tational/architectural cost and confidentiality (e.g. avoiding the storage of sensitive data). The full paper focuses on two questions related to the use of incremental learning for FDS: (1) Can incremental learning be as competitive as batch and retraining approaches? (2) Can combining incremental, ensembles, diversity, and transfer learning lead to efficient models under concept drift? We investigate those elements and provide an experimental evaluation on a real-life case study including more than 150 days of e-commerce transactions (or 50 million transactions).

Europe has been affected by an alarming number of wildfires every year, ravaging nearly half a million hectares of forestry areas and resulting in an unacceptable amount of human losses. The year 2017 was one of the worst ever recorded, with devastating wildfires raging across the globe. Bearing these shocking facts in mind, this position paper aims to lay the foundations of crucial new upcoming research in the field of forestry robotics by presenting an overview of the SEMFIRE project. SEMFIRE proposes the development of a robotic system designed to reduce the accumulation of combustible material in forests, thus assisting in landscaping maintenance tasks for wildfire prevention. This paper describes the multi-robot system envisaged by SEMFIRE, combined with pervasive local positioning, multi-robot navigation and coordination in forests, and an innovative artificial perception architecture.

Digital and paper based documents co-exist in our daily lives. Seamless integration of information from both sources is crucial for efficient knowledge management. This paper address the algorithm that can handle the detection of document so that it can be captured easily to convert it into a digital form for automatic integration of relevant information in electronic work flows. It uses the deep learning technique to provide a solution which is more generalized and flexible than other available solutions.

In imbalanced learning problems, classifiers bias the majority class which would lead to poor predictive accuracy over the minority class. Synthetic minority oversampling technique (SMOTE) is one of the oversampling methods for imbalanced learning problems. However SMOTE may cause over-generalizing. Existing methods for improving SMOTE usually only select the borderline minority samples to over-sample. In this way, some non-borderline informative minority samples can not be fully taken into consideration. To overcome this issue, a new oversampling method named LAD-SMOTE is proposed in this paper based on locally adaptive distance. Both borderline minority samples and non-borderline informative minority samples are over-sampled by our method. In LAD-SMOTE, non-borderline informative minority samples can be fully learned and thus over-generalizing caused by SMOTE can be alleviated. Experimental results show that the algorithm can efficiently improve the classification performance on imbalanced data sets than some existing ones.

We address domain adaptation in the context of clustering where we are given a set of unlabeled data, coming from several domains, and the goal is to group data into different categories regardless of the domain they come from. This is a challenging problem since we do not have any supervision unlike most adaptation scenarios studied earlier, and is very relevant in practical industry applications where labeled data often comes at a premium especially while deploying services that do not have a comparable predecessor. Our philosophy in addressing this problem draws motivation from the concept of dirty paper coding, a communications technique where the signal being transmitted through a noisy channel is encoded with priors on the possible noise patterns to assist reliable decoding of the signal at the receiver. We focus on image applications in this paper, where we encode priors on possible image domain shift factors such as viewpoint, lighting, blur and appearance variations and utilize geometric adaptation mechanisms to perform clustering. We illustrate the utility of our approach on standard datasets involving objects and faces, by obtaining around 18% improvement on average over existing approaches.

In this paper, we propose a novel marker design and its tracking algorithm for room-sized MR/AR environments. The markers and the algorithm are designed to solve the following practical problems: i) the difficulties in creating and arranging markers and ii) the trade-off between inconspicuousness and robustness of markers. The proposed markers are small chips that are cut off a large paper sheet, and are arranged at random positions in an environment or on objects. This paper shows the design concept and feasibility of the proposed markers.

Item test bank system plays very important role in auto generating test or exam paper in assessments in schools and universities. A quite number of researchers have proposed some algorithms in generating test paper based on some well-defined attributes such as time, question type, knowledge point, difficulty level and others. It has always been the aim of these researchers to generate high quality test paper with appropriate level of difficulty in test questions. As a result of this, Bloom taxonomy has been adopted to ensure difficulty level of test questions is appropriate. However, there is no evidence that current test items or questions in the item test bank system are classified in accordance to BT using machine learning approach. Manual classifying is tedious and laborious work and inconsistency in classifying items can take place due to different judgement from instructors. A better approach is to use machine learning namely question classifier such as Support Vector Machine to automate the classification of the test items. Despite some research work has been done on using classifiers to classify questions, there is no evidence that this type of work has been integrated into item bank test system. In view of this, this study proposes a change in existing framework of item test bank system by integrating the facility to automate classifying items in accordance to Bloom taxonomy. With all this in place, the automation of classifying questions or test items in accordance to BT with a reasonable accuracy can be achieved.

In this position paper, we initiate a systematic treatment of reaching consensus in a permissionless network. We prove several simple but hopefully insightful lower bounds that demonstrate exactly why reaching consensus in a permission-less setting is fundamentally more difficult than the classical, permissioned setting. We then present a simplified proof of Nakamoto's blockchain which we recommend for pedagogical purposes. Finally, we survey recent results including how to avoid well-known painpoints in permissionless consensus, and how to apply core ideas behind blockchains to solve consensus in the classical, permissioned setting and meanwhile achieve new properties that are not attained by classical approaches.

Designing efficient self-management algorithms for autonomic managers has been an ongoing and evolving research area. In literature, many machine learning paradigms have been proposed and exploited to make effective decisions in autonomic managers. This position paper proposes to optimize the decision making algorithm closer to the nature inspired decision making process. Core of the proposed framework is case-based reasoning which is an incremental learning mechanism for solving new system problems using the past experience. Fuzzy clustering has been proposed to maintain the knowledge-base of past problems in autonomic managers. New monitored problem in autonomic manager seeks fuzzy memberships amongst different clusters in the knowledge-base and problem solver in autonomic manager exploits existing solutions in different clusters with respect to their fuzzy membership value. It has been tested on a simulated case-study of Autonomic Forest Fire Application and up to 88% accurate predictions have been observed.

In this position paper, we introduce an instance of interlanguage clones in a multilingual web application system and then discuss research challenges on the detection and analysis.

Classification is one of the most important supervised machine learning tasks. During the training of a classification model, the training instances are fed to the model multiple times (during multiple epochs) in order to iteratively increase the classification performance. The increasing complexity of models has led to a growing demand for model interpretability through visualizations. Existing approaches mostly focus on the visual analysis of the final model performance after training and are often limited to aggregate performance measures. In this paper we introduce InstanceFlow, a novel dual-view visualization tool that allows users to analyze the learning behavior of classifiers over time on the instance-level. A Sankey diagram visualizes the flow of instances throughout epochs, with on-demand detailed glyphs and traces for individual instances. A tabular view allows users to locate interesting instances by ranking and filtering. In this way, InstanceFlow bridges the gap between class-level and instance-level performance evaluation while enabling users to perform a full temporal analysis of the training process., conditionally accepted for IEEE VIS 2020 short paper track; minor revisions compared to original version

The development of deep learning has rapidly updated image-based localization techniques. This paper presents a review and comparison of the current state-of-the-art methods for image-based localization using deep learning in the indoor environment. Traditional Global Structure from Motion (SfM) pipeline and learning-based pipeline from the recent techniques have been analyzed. Based on the pipeline, the methods are categorized into three groups: learned features and matching, learned relative pose estimation, and learned absolute pose estimation. Since multiple sensors are used in many applications, sensor fusion techniques including image information, have been briefly reviewed in this paper as well. Furthermore, the paper discusses challenges in these methods and concludes learned features and matching is the more competitive method for indoor localization.