Labeling Problem Research Articles

Towards end-to-end car license plate location and recognition in unconstrained scenarios

Benefiting from the rapid development of convolutional neural networks, the performance of car license plate detection and recognition has been largely improved. Nonetheless, most existing methods solve detection and recognition problems separately, and focus on specific scenarios, which hinders the deployment for real-world applications. To overcome these challenges, we present an efficient and accurate framework to solve the license plate detection and recognition tasks simultaneously. It is a lightweight and unified deep neural network, that can be optimized end-to-end and work in real-time. Specifically, for unconstrained scenarios, an anchor-free method is adopted to efficiently detect the bounding box and four corners of a license plate, which are used to extract and rectify the target region features. Then, a novel convolutional neural network branch is designed to further extract features of characters without segmentation. Finally, the recognition task is treated as sequence labeling problems, which are solved by Connectionist Temporal Classification (CTC) directly. Several public datasets including images collected from different scenarios under various conditions are chosen for evaluation. Experimental results indicate that the proposed method significantly outperforms the previous state-of-the-art methods in both speed and precision.

Friend or Foe: A Review and Synthesis of Computational Models of the Identity Labeling Problem

ABSTRACT We introduce the identity labeling problem – given an individual in a social situation, can we predict what identity(ies) they will be labeled with by someone else? This problem remains a theoretical gap and methodological challenge, evidenced by the fact that models of social-cognition often sidestep the issue by treating identities as already known. We build on insights from existing models to develop a new framework, entitled Latent Cognitive Social Spaces, that can incorporate multiple social cues including sentiment information, socio-demographic characteristics, and institutional associations to estimate the most culturally expected identity. We apply our model to data collected in two vignette experiments, finding that it predicts identity labeling choices of participants with a mean absolute error of 10.9%, a 100% improvement over previous models based on parallel constraint satisfaction and affect control theory.

Chinese grammatical error diagnosis based on sequence tagging methods

In view of the problems such as exploding gradient or vanishing gradient or inefficiency caused by parallel problems when traditional neural networks deal with long text grammar error correction. In this paper, Chinese grammatical error detection is proposed as a sequence tagging problem for named entity recognition, and a Chinese grammatical error detection model based on BERT BILSTM CRF is designed and implemented by using Bert model to construct word vectors. Meanwhile, three groups of comparative tests were designed: CRF, BILSTM CRF and BERT CRF. In the end, BERT BILSTM CRF model achieves the error detection accuracy of 66.67% and 52.94% in M (missing error) and W (ordering error), respectively. BERT CRF and CRF models have also achieved good results in S (selection error) and R (redundant words). The research shows that it is feasible to solve grammatical error detection as a sequence labeling problem. On the other hand, a Chinese grammatical error detection model based on BERT BILSTM CRF is proposed for the first time.

End-to-End Detection-Segmentation System for Face Labeling

In this paper, we propose an end-to-end detection-segmentation system to implement detailed face labeling. Fully convolutional networks (FCN) has become the mainstream algorithm in the field of semantic segmentation due to the state-of-the-art performance. However, a general FCN usually produces smooth and homogeneous results. Moreover, when semantic category is extremely unbalanced in samples such as face labeling problem, features for some categories cannot be well explored by FCN. To alleviate these problems, a face image is firstly encoded to multi-level feature maps by a pyramid FCN, then features of different facial components are extracted separately according to the bounding box provided by a one-stage detection head. Three class-specific sub-networks are employed to process the extracted features to obtain the respective segmentation results. The skin-hair region can be decoded directly from the back end of the pyramid FCN. Finally, the overall segmentation result is obtained by combining different branches. Moreover, the proposed method trained on a single-face labeled dataset, can be directly used to implement detailed multi-face labeling tasks without any network modification and additional module or data. The overall structure can be trained in an end-to-end manner while maintaining a small network size (12 MB). Experiments show that the proposed method can generate more accurate (single or multi) face labeling results comparing with previous works and gets the state-of-the-art results in HELEN face dataset.

Non-compliance of sanitary standards in the offer of animal products in e-commerce and the risks to public health

Introdução: A comercialização de produtos de origem animal (POA) em marketplace sem a prévia inspeção e fiscalização expõe os consumidores a produtos que podem apresentar problemas de autenticidade, rotulagens inadequadas, fraudes ou não conformidades de ordem higiênicosanitária ou tecnológica, desrespeitando os direitos do consumidor, além da possibilidade de causar danos à saúde pública. Objetivo: Identificar e traçar o perfil dos produtos ofertados via e-commerce e verificar o cumprimento às normas de regulação sanitária de alimentos e de proteção e defesa do consumidor no Brasil. Método: Anúncios de venda de POA foram rastreados, enumerados e tabulados de forma qualitativa nas categorias: tipo de produto; carimbo oficial da Inspeção Sanitária e origem – divisão regional do país. Em seguida, os dados tabulados foram analisados e submetidos à estatística descritiva e teste de qui-quadrado (χ2). Resultados: Foram identificados 498 POA sendo ofertados, e houve a predominância daqueles que não apresentavam o carimbo oficial de inspeção e fiscalização sanitária. Os produtos cárneos representaram a maioria dos produtos comercializados e, na categoria origem, a região Sudeste apresentou a maior frequência de anúncios. Os resultados obtidos demonstraram o descumprimento à legislação sanitária. Conclusões: Foram identificadas diversas categorias de POA sendo ofertados de forma deliberada em descumprimento à legislação higiênico-sanitária para a produção e comercialização de produtos alimentícios.

Cost-sensitive learning for semi-supervised hit-and-run analysis

Hit-and-run crashes not only degrade the morality, but also result in delays of medical services provided to victims. However, class imbalance problem exists as the number of hit-and-run crashes is much smaller than that of non-hit-and-run crashes. The missing label problem also exists in the crash analysis due to reasons like data barrier such that the information hidden in the unlabelled samples has not been effectively utilised. In this paper, a cost-sensitive semi-supervised logistic regression (CS3LR) model is proposed for hit-and-run analysis, in order to tackle class-imbalanced data distribution and missing label problem, based on the crash dataset of Victorian, Australia (2013–2019). By performing label estimation with logistic regression jointly utilising both labelled and unlabelled data with pseudo labels in a well-designed cost-sensitive semi-supervised maximum likelihood framework, the proposed model can obtain an unbiased likelihood parameter for hit-and-run prediction and analysis. Comparing the experimental results of CS3LR model with two logistic regression models and seven machine learning methods, better performance of CS3LR model is demonstrated. The most significant contributing factors to hit-and-run crashes extracted by CS3LR with only 10% labelled data show a high degree of consistency with the true contributing factors obtained by the supervised cost-sensitive logistic regression with complete hit-and-run labels. The effects of class-weighted ratio and hyper-parameter λ on the performance of hit-and-run crash prediction model have also been analysed. The results can further provide recommendations and implications on the policies and counter-measures for preventing hit-and-run collisions and crimes. The methodology proposed in this paper can also be employed to analyse crash data with other types of missing labels, such as crash severity.

Spatial‐tree filter for cost aggregation in stereo matching

The desired solution to many labelling problems in computer vision is a spatially smooth result where label changes are aligned with the edges of the guidance image. It can be obtained traditionally by smoothing the label costs using edge-aware filters. However, local filters incorporate the information in a local support region to obtain locally-optimal and non-local tree-based filters, which often overuse piece-wise constant assumptions. In this paper, we propose a spatial-tree filter for cost aggregation. The tree model incorporates the spatial affinity into the tree structure. The tree distance between two pixels on our spatial tree is an approximated geodesic distance, which acts as a pixel similarity metric. The filtering process was implemented by recursively techniques in four directions: Top-to-bottom, left-to-right, and vice-versa. Thus, the complexity of our approach is linear to the number of image pixels. Extensive experiments demonstrate the effectiveness and efficiency of our spatial-tree filter in image smoothing and stereo matching. Our method performs better than the existing tree-based non-local method in cost aggregation.

Biomedical named entity recognition using BERT in the machine reading comprehension framework

Recognition of biomedical entities from literature is a challenging research focus, which is the foundation for extracting a large amount of biomedical knowledge existing in unstructured texts into structured formats. Using the sequence labeling framework to implement biomedical named entity recognition (BioNER) is currently a conventional method. This method, however, often cannot take full advantage of the semantic information in the dataset, and the performance is not always satisfactory. In this work, instead of treating the BioNER task as a sequence labeling problem, we formulate it as a machine reading comprehension (MRC) problem. This formulation can introduce more prior knowledge utilizing well-designed queries, and no longer need decoding processes such as conditional random fields (CRF). We conduct experiments on six BioNER datasets, and the experimental results demonstrate the effectiveness of our method. Our method achieves state-of-the-art (SOTA) performance on the BC4CHEMD, BC5CDR-Chem, BC5CDR-Disease, NCBI-Disease, BC2GM and JNLPBA datasets, achieving F1-scores of 92.92%, 94.19%, 87.83%, 90.04%, 85.48% and 78.93%, respectively.

Learning with noisy labels method for unsupervised domain adaptive person re-identification

Unsupervised domain adaptive (UDA) person re-identification (re-ID) aims to adapt the model trained on a labeled source domain to an unlabeled target domain. For pseudo-label-based UDA methods, pseudo labels noise is the main problem for model degradation and the factors that cause noise are complex. In this paper, a novel learning with noisy labels (LNL) method for UDA person re-ID is proposed to address this problem by analyzing the noise data itself. LNL learns with noise data from two aspects, including noise correction and noise resistance. According to the idea of neighbor consistency, pseudo labels correction (PLC) based on sample similarity is designed to correct the noisy pseudo labels before training. In order to solve the problem of noise labels in deep learning, noise recognition based on similarity and confidence relationship (SACR) is designed. Then, an easy-to-hard model collaborative training (MCT) strategy is developed, which can resist noise during the training process and obtain a more robust training model. To further avoid overfitting of noisy samples, the re-weighting (RW) method is employed in MCT. The proposed LNL model achieves considerable results of 75.2%/88.9% and 62.5%/77.4% mAP/Rank-1 on DukeMTMC-reID-to-Market-1501 and Market-1501-to-DukeMTMC-reID UDA tasks.

L(2,1,1)-Labeling of Circular-Arc Graphs

Graph labeling problem has been broadly studied in recent past for its wide applications, in mobile communication system for frequency assignment, radar, circuit design, X-ray crystallography, coding theory, etc. An L211-labeling (L211L) of a graph G = (V, E) is a function γ : V → Z∗ such that |γ(u) − γ(v)| ≥ 2, if d(u, v) = 1 and |γ(u) − γ(v)| ≥ 1, if d(u, v) = 1 or 2, where Z∗ be the set of non-negative integers and d(u, v) represents the distance between the nodes u and v. The L211L numbers of a graph G, are denoted by λ2,1,1(G) which is the difference between largest and smallest labels used in L211L. In this article, for circular-arc graph (CAG) G we have proved that λ2,1,1(G) ≤ 6∆ − 4, where ∆ represents the degree of the graph. Beside this we have designed a polynomial time algorithm to label a CAG satisfying the conditions of L211L. The time complexity of the algorithm is O(n∆2), where n is the number of nodes of the graph G.

ОСНОВЫ ФОРМАЛЬНОЙ ТЕОРИИ КЛАССИФИКАЦИИ ТЕХНИЧЕСКИХ СИСТЕМ НА ПРИМЕРЕ ОБЪЕКТОВ ЭНЕРГЕТИКИ

The article describes the power generating and other energy facilities as complex technical systems interacting with the material, financial, informational and legal environment. To ensure functioning and development of a technical system it is important to determine its place among other technical systems: technical systems and their elements must be classified. There have been presented the principles of formal theory of technical system classification. The classification objects have been given a definition. The concepts of classification, classification attribute and meaning of the classification attribute are formalized. The goal and tasks of the classification are formulated. The stages of forming and applying classifiers include developing the composition and structure of classifier kernel, marking interpretation elements and selecting an element corresponding to the user's request. The alphabet, inference rules, axioms and main theorems of the classification formal theory are presented. It is proved that a complete consistent independent classification theory is decidable. The taxonomy of the classifier types is considered: elementary (incomplete and complete), basic (discrete hierarchical, discrete matrix, continuous band), complex (combined and limiting). Examples of using classifiers of various types in the description of energy objects are given. The algorithms, labeling and selection problems are considered. The use of a grounded and structured classification theory allows better understanding and description of the relationship between technical and related systems and l increases the efficiency of functioning and development of technical systems.

A Hybrid Model for Family History Information Identification and Relation Extraction: Development and Evaluation of an End-to-End Information Extraction System.

BackgroundFamily history information is important to assess the risk of inherited medical conditions. Natural language processing has the potential to extract this information from unstructured free-text notes to improve patient care and decision making. We describe the end-to-end information extraction system the Medical University of South Carolina team developed when participating in the 2019 National Natural Language Processing Clinical Challenge (n2c2)/Open Health Natural Language Processing (OHNLP) shared task.ObjectiveThis task involves identifying mentions of family members and observations in electronic health record text notes and recognizing the 2 types of relations (family member-living status relations and family member-observation relations). Our system aims to achieve a high level of performance by integrating heuristics and advanced information extraction methods. Our efforts also include improving the performance of 2 subtasks by exploiting additional labeled data and clinical text-based embedding models.MethodsWe present a hybrid method that combines machine learning and rule-based approaches. We implemented an end-to-end system with multiple information extraction and attribute classification components. For entity identification, we trained bidirectional long short-term memory deep learning models. These models incorporated static word embeddings and context-dependent embeddings. We created a voting ensemble that combined the predictions of all individual models. For relation extraction, we trained 2 relation extraction models. The first model determined the living status of each family member. The second model identified observations associated with each family member. We implemented online gradient descent models to extract related entity pairs. As part of postchallenge efforts, we used the BioCreative/OHNLP 2018 corpus and trained new models with the union of these 2 datasets. We also pretrained language models using clinical notes from the Medical Information Mart for Intensive Care (MIMIC-III) clinical database.ResultsThe voting ensemble achieved better performance than individual classifiers. In the entity identification task, our top-performing system reached a precision of 78.90% and a recall of 83.84%. Our natural language processing system for entity identification took 3rd place out of 17 teams in the challenge. We ranked 4th out of 9 teams in the relation extraction task. Our system substantially benefited from the combination of the 2 datasets. Compared to our official submission with F1 scores of 81.30% and 64.94% for entity identification and relation extraction, respectively, the revised system yielded significantly better performance (P<.05) with F1 scores of 86.02% and 72.48%, respectively.ConclusionsWe demonstrated that a hybrid model could be used to successfully extract family history information recorded in unstructured free-text notes. In this study, our approach to entity identification as a sequence labeling problem produced satisfactory results. Our postchallenge efforts significantly improved performance by leveraging additional labeled data and using word vector representations learned from large collections of clinical notes.

Semi-supervised anomaly detection in dynamic communication networks

To ensure the security and stabilization of the communication networks, anomaly detection is the first line of defense. However, their learning process suffers two major issues: (1) inadequate labels: there are many different kinds of attacks but rare abnormal nodes in mt of these atstacks; and (2) inaccurate labels: considering the heavy network flows and new emerging attacks, providing accurate labels for all nodes is very expensive. The inadequate and inaccurate label problem challenges many existing methods because the majority normal nodes result in a biased classifier while the noisy labels will further degrade the performance of the classifier. To tackle these issues, we propose SemiADC, a Semi-supervised Anomaly Detection framework for dynamic Communication networks. SemiADC first approximately learns the feature distribution of normal nodes with regularization from abnormal ones. It then cleans the datasets and extracts the nodes sasainaccurate labels by the learned feature distribution and structure-based temporal correlations. These self-learning processes run iteratively with mutual promotion, and finally help increase the accuracy of anomaly detection. Experimental evaluations on real-world datasets demonstrate the effectiveness of our SemiADC, which performs substantially better than the state-of-art anomaly detection approaches without the demand of adequate and accurate supervision.

HAL: Hybrid active learning for efficient labeling in medical domain

The success of the deep convolutional neural networks in computer vision tasks mainly relies on massive labeled training data. However, in the field of medical images, it is difficult to construct large labeled datasets since the labeling of medical images is time-consuming, labor-intensive, and medical expertise demanded. To meet the challenge, we propose a hybrid active learning framework HAL for efficient labeling in the medical domain, which integrates active learning into deep learning to reduce the cost of manual labeling and take the advantages of deep neural networks. The proposed HAL utilizes a hybrid sampling strategy considering both sample diversity and prediction loss simultaneously. The effectiveness and efficiency of proposed HAL are validated on three medical image datasets. The experimental results show that the proposed HAL outperforms several state-of-the-art active learning methods. On the Hyper-Kvasir Dataset, with only 10% of the labels, the HAL achieves 95% performance of the deep learning method trained on the entire dataset. The quantitative and qualitative analysis proves that HAL can greatly reduce the number of labels needed for training a deep neural network, which is robust to address efficient labeling problems even with imbalanced data distribution.

Radio Mean Labeling Of Paths And Its Total Graph

A graph labeling problem is an assignment of labels to the vertices or edges (or both) of a graph G satisfying some mathematical condition. Radio Mean Labeling, a vertex-labeling of graphs with non-negative integers has a significant application in the study of problems related to radio channel assignment. The maximum label used in a radio mean labeling is called its span, and the lowest possible span of a radio mean labeling is called the radio mean number of a graph. In this paper, we obtain the radio mean number of paths and total graph of paths.

Biomedical event trigger extraction based on multi-layer residual BiLSTM and contextualized word representations

Biomedical event extraction is an important branch of biomedical information extraction. Trigger extraction is the most essential sub-task in event extraction, which has been widely concerned. Existing trigger extraction studies are mostly based on conventional machine learning or neural networks. But they neglect the ambiguity of word representations and the insufficient feature extraction by shallow hidden layers. In this paper, trigger extraction is treated as a sequence labeling problem. We introduce the language model to dynamically compute contextualized word representations and propose a multi-layer residual bidirectional long short-term memory (BiLSTM) architecture. First, we concatenate contextualized word embedding, pretrained word embedding and character-level embedding as the feature representations, which effectively solves the tokens’ ambiguity in biomedical corpora. Then, the designed BiLSTM block with residual connection and gated multi-layer perceptron is adopted to extract features iteratively. This architecture improves the ability of our model to capture information and avoids gradient exploding or vanishing. Finally, we combine the multi-layer residual BiLSTM with CRF layer to obtain more reasonable label sequences. Comparing with other state-of-the-art methods, the proposed model achieves the competitive performance (F1-score: 80.74%) on the biomedical multi-level event extraction (MLEE) corpus without any manual participation and feature engineering.

Multi-granularity Deep Local Representations for Irregular Scene Text Recognition

Recognizing irregular text from natural scene images is challenging due to the unconstrained appearance of text, such as curvature, orientation, and distortion. Recent recognition networks regard this task as a text sequence labeling problem and most networks capture the sequence only from a single-granularity visual representation, which to some extent limits the performance of recognition. In this article, we propose a hierarchical attention network to capture multi-granularity deep local representations for recognizing irregular scene text. It consists of several hierarchical attention blocks, and each block contains a Local Visual Representation Module (LVRM) and a Decoder Module (DM). Based on the hierarchical attention network, we propose a scene text recognition network. The extensive experiments show that our proposed network achieves the state-of-the-art performance on several benchmark datasets including IIIT-5K, SVT, CUTE, SVT-Perspective, and ICDAR datasets under shorter training time.

Pendampingan Dan Pelatihan Desain Grafis Untuk Kewirausahaan Bagi Siswa SMK Muhammadiyah 9 Wagir

The problem faced by students is at the production stage, namely the problem of product labeling. Until now labeling was done simply or without labeling. If students can design their own product labels, of course this will add to the selling value of the product. The existence of labels and brochures also add value to the promotion of student products. The aim of this mentoring activity is to provide training and graphic design assistance to students of SMK. The steps of this community service activities are observations, socialization, graphic design training, assistance in making labels and brochures for student entrepreneurial products, as well as evaluating and reflecting on service activities. The result of this service activity was in the form of training and graphic design assistance for students. All of the activities was well done . This can be seen from the number of students who participated in training and mentoring for three days. In addition, there are positive responses from teachers and students that have an impact on the success of this graphic design training and assistance. This can be seen from the results of student designs that are attractive and can be used as a commercial tool.

Nonlocal graph theory based transductive learning for hyperspectral image classification

Hyperspectral Image classification plays an important role in the maintenance of remote image analysis, which has been attracting a lot of research interest. Although various approaches, including unsupervised and supervised methods, have been proposed, obtaining a satisfactory classification result is still a challenge. In this paper, an efficient transductive learning method using variational nonlocal graph theory for hyperspectral image classification is proposed. First, the nonlocal vector neighborhood similarity is employed to build sparse graph representation. Then the variational segmentation framework is extended to label space, and the vectorization nonlocal energy function is constructed. Next, a fast comprehensive alternating minimization iteration algorithm is designed to implement labels transductive learning. At the same time, the labeled sample constraints are doubled ensured with simplex projection. Finally, experiments on six widely used hyperspectral image datasets are implemented, compared with other state-of-the-art classification methods, the classification results demonstrate that the proposed method has higher classification performance. Benefiting from graph theory and transductive idea, the proposed classification method can propagate labels and overcome the very high dimensionality and limited labeling problem to some extent.

A noisy label and negative sample robust loss function for DNN-based distant supervised relation extraction

As a major method for relation extraction, distantly supervised relation extraction (DSRE) suffered from the noisy label problem and class imbalance problem (these two problems are also common for many other NLP tasks, e.g., text classification). However, there seems no existing research in DSRE or other NLP tasks that can simultaneously solve both problems, which is a significant insufficiency in related researches. In this paper, we propose a loss function which is robust to noisy label and efficient for the imbalanced class dataset. More specific, first we quantify the negative impacts of the noisy label and class imbalance problems. And then we construct a loss function that can minimize these negative impacts through a linear programming method. As far as we know, this seems to be the first attempt to address the noisy label problem and class imbalance problem simultaneously. We evaluated the constructed loss function on the distantly labeled dataset, our artificially noised dataset, human-annotated dataset of Docred, as well as the artificially noised dataset of CoNLL 2003. Experimental results indicate that a DNN model adopting the constructed loss function can outperform other models that adopt the state-of-the-art noisy label robust or negative sample robust loss functions.