Showing papers on "Conditional random field published in 2022"
33 citations
TL;DR: Wang et al. as discussed by the authors proposed a CNN-LSTM model to predict gas field production based on a gas field in southwest China, where the convolutional neural network (CNN) has a feature extraction ability, and the LSTM can learn sequence dependence.
33 citations
TL;DR: In this paper , a multiscale residual attention (MRA-UNet) was proposed for brain tumor regions segmentation using three consecutive slices as input to preserve the sequential information and employs multiscaling learning in a cascade fashion, enabling it to exploit the adaptive region of interest scheme to segment enhanced and core tumor regions accurately.
27 citations
TL;DR: Wang et al. as discussed by the authors proposed a multi-head self-attention based Bi-directional Long Short-Term Memory (MUSA-BiLSTM-CRF) model for Chinese clinical named entity recognition.
26 citations
TL;DR: A method based on graph convolutional neural (GCN) network and conditional random field (CRF) for predicting human lncRNA-miRNA interactions, named GCNCRF, which has higher prediction accuracy than the other six state-of-the-art methods.
Abstract: Long non-coding RNA (lncRNA) and microRNA (miRNA) are two typical types of non-coding RNAs (ncRNAs), their interaction plays an important regulatory role in many biological processes. Exploring the interactions between unknown lncRNA and miRNA can help us better understand the functional expression between lncRNA and miRNA. At present, the interactions between lncRNA and miRNA are mainly obtained through biological experiments, but such experiments are often time-consuming and labor-intensive, it is necessary to design a computational method that can predict the interactions between lncRNA and miRNA. In this paper, we propose a method based on graph convolutional neural (GCN) network and conditional random field (CRF) for predicting human lncRNA-miRNA interactions, named GCNCRF. First, we construct a heterogeneous network using the known interactions of lncRNA and miRNA in the LncRNASNP2 database, the lncRNA/miRNA integration similarity network, and the lncRNA/miRNA feature matrix. Second, the initial embedding of nodes is obtained using a GCN network. A CRF set in the GCN hidden layer can update the obtained preliminary embeddings so that similar nodes have similar embeddings. At the same time, an attention mechanism is added to the CRF layer to reassign weights to nodes to better grasp the feature information of important nodes and ignore some nodes with less influence. Finally, the final embedding is decoded and scored through the decoding layer. Through a 5-fold cross-validation experiment, GCNCRF has an area under the receiver operating characteristic curve value of 0.947 on the main dataset, which has higher prediction accuracy than the other six state-of-the-art methods.
25 citations
TL;DR: Wang et al. as discussed by the authors proposed a continuous CRF graph convolution (CRFConv) for point cloud segmentation, which can capture the structure of features well to improve the representation ability of features rather than simply smoothing.
17 citations
TL;DR: Wang et al. as discussed by the authors proposed a continuous CRF graph convolution (CRFConv) for point cloud segmentation, which can capture the structure of features well to improve the representation ability of features rather than simply smoothing.
17 citations
TL;DR: Wang et al. as discussed by the authors proposed a multi-scale segmentation squeeze-and-excitation UNet with a conditional random field (M-SegSEUNet-CRF) to automatically segment lung tumors from CT images.
14 citations
TL;DR: The proposed approach explores two deep learning methods to achieve both negation and uncertainty detection in clinical texts written in Spanish using bidirectional Long-Short Term Memory with a Conditional Random Field layer and Bidirectional Encoder Representation for Transformers for Transformers.
Abstract: Detecting negation and uncertainty is crucial for medical text mining applications; otherwise, extracted information can be incorrectly identified as real or factual events. Although several approaches have been proposed to detect negation and uncertainty in clinical texts, most efforts have focused on the English language. Most proposals developed for Spanish have focused mainly on negation detection and do not deal with uncertainty. In this paper, we propose a deep learning-based approach for both negation and uncertainty detection in clinical texts written in Spanish. The proposed approach explores two deep learning methods to achieve this goal: (i) Bidirectional Long-Short Term Memory with a Conditional Random Field layer (BiLSTM-CRF) and (ii) Bidirectional Encoder Representation for Transformers (BERT). The approach was evaluated using NUBES and IULA, two public corpora for the Spanish language. The results obtained showed an F-score of 92% and 80% in the scope recognition task for negation and uncertainty, respectively. We also present the results of a validation process conducted using a real-life annotated dataset from clinical notes belonging to cancer patients. The proposed approach shows the feasibility of deep learning-based methods to detect negation and uncertainty in Spanish clinical texts. Experiments also highlighted that this approach improves performance in the scope recognition task compared to other proposals in the biomedical domain.
13 citations
23 May 2022
TL;DR: This work proposes EmotionFlow for ERC with the consideration of the spread of participants' emotions during a conversation, and conducts extensive experiments on a public dataset Multimodal EmotionLines Dataset (MELD), demonstrating the effectiveness of the model.
Abstract: Emotion recognition in conversations (ERC) has attracted increasing interests in recent years, due to its wide range of applications, such as customer service analysis, health-care consultation, etc. One key challenge of ERC is that users' emotions would change due to the impact of others' emotions. That is, the emotions within the conversation can spread among the communication participants. However, the spread impact of emotions in a conversation is rarely addressed in existing researches. To this end, we propose EmotionFlow for ERC with the consideration of the spread of participants' emotions during a conversation. EmotionFlow first encodes users' utterance by concatenating the context with an auxiliary question, which helps to learn user-specific features. Then, conditional random field is applied to capture the sequential information at emotional level. We conduct extensive experiments on a public dataset Multimodal EmotionLines Dataset (MELD), and the results demonstrate the effectiveness of our proposed model.
13 citations
TL;DR: An uncertainty-based active learning strategy called the lowest token probability (LTP), which combines the input and output of conditional random field (CRF) to select informative instances, which is a simple and powerful strategy that does not favor long sequences and does not need to revise the model.
TL;DR: A novel NER model is proposed to learn information about syntactic dependency graphs with graph neural networks, and merge learned information into the classic Bidirectional Long Short-Term Memory - Conditional Random Field NER scheme.
Abstract: Named entity recognition (NER) isa preliminary task in natural language processing (NLP). Recognizing Chinese named entities from unstructured texts is challenging due to the lack of word boundaries. Even if performing Chinese Word Segmentation (CWS) could help to determine word boundaries, it is still difficult to determine which words should be clustered together for entity identification, since entities are often composed of multiple-segmented words. As dependency relationships between segmented words could help to determine entity boundaries, it is crucial to employ information related to syntactic dependency relationships to improve NER performance. In this paper, we propose a novel NER model to learn information about syntactic dependency graphs with graph neural networks, and merge learned information into the classic Bidirectional Long Short-Term Memory (BiLSTM) - Conditional Random Field (CRF) NER scheme. In addition, we extract various kinds of task-specific hidden information from multiple CWS and part-of-speech (POS) tagging tasks, to further improve the NER model. We finally leverage multiple self-attention components to integrate multiple kinds of extracted information for named entity identification. Experimental results on three public benchmark datasets show that our model outperforms the state-of-the-art baselines in most scenarios.
TL;DR: NL-LinkNet as mentioned in this paper proposes an efficient nonlocal LinkNet with nonlocal blocks (NLBs) that can grasp relations between global features and results in more accurate road segmentation.
Abstract: Road extraction from very high resolution (VHR) satellite images is one of the most important topics in the field of remote sensing. In this letter, we propose an efficient nonlocal LinkNet with nonlocal blocks (NLBs) that can grasp relations between global features. This enables each spatial feature point to refer to all other contextual information and results in more accurate road segmentation. In detail, our single model without any postprocessing like conditional random field (CRF) refinement performed better than any other published state-of-the-art ensemble model in the official DeepGlobe Challenge. Moreover, our nonlocal LinkNet (NL-LinkNet) beat the D-LinkNet, the winner of the DeepGlobe challenge (Demir et al. , 2018), with 43% less parameters, less giga floating-point operations per seconds (GFLOPs), and shorter training convergence time. We also present empirical analyses on the proper usages of NLBs for the baseline model.
TL;DR: Zhang et al. as discussed by the authors proposed an attentive multilevel feature fusion (AMFF) model for NER, which captures the multilelevel features in the current context from various perspectives.
Abstract: In the era of information explosion, named entity recognition (NER) has attracted widespread attention in the field of natural language processing, as it is fundamental to information extraction. Recently, methods of NER based on representation learning, e.g., character embedding and word embedding, have demonstrated promising recognition results. However, existing models only consider partial features derived from words or characters while failing to integrate semantic and syntactic information, e.g., capitalization, inter-word relations, keywords, and lexical phrases, from multilevel perspectives. Intuitively, multilevel features can be helpful when recognizing named entities from complex sentences. In this study, we propose a novel attentive multilevel feature fusion (AMFF) model for NER, which captures the multilevel features in the current context from various perspectives. It consists of four components to, respectively, capture the local character-level (CL), global character-level (CG), local word-level (WL), and global word-level (WG) features in the current context. In addition, we further define document-level features crafted from other sentences to enhance the representation learning of the current context. To this end, we introduce a novel context-aware attentive multilevel feature fusion (CAMFF) model based on AMFF, to fully leverage document-level features from all the previous inputs. The obtained multilevel features are then fused and fed into a bidirectional long short-term memory (BiLSTM)-conditional random field (CRF) network for the final sequence labeling. Extensive experiments on four benchmark datasets demonstrate that our proposed AMFF and CAMFF models outperform a set of state-of-the-art baseline methods and the features learned from multiple levels are complementary.
TL;DR: A deep learning-based information extraction algorithm (named BERT-BiLSTM-CRF) for automatically extracting temporal information from social media messages is proposed and outperforms the current state-of-the-art models.
TL;DR: Experimental results on two public multispectral datasets indicate the effectiveness and robustness of the proposed method when compared with other state-of-the-art methods.
Abstract: Multispectral image change detection is an important application in the field of remote sensing. Multispectral images usually contain many complex scenes, such as ground objects with diverse scales and proportions, so the change detection task expects the feature extractor is superior in adaptive multi-scale feature learning. To address the above-mentioned problems, a multispectral image change detection method based on multi-scale adaptive kernel network and multimodal conditional random field (MSAK-Net-MCRF) is proposed. The multi-scale adaptive kernel network (MSAK-Net) extends the encoding path of the U-Net, and designs a weight-sharing bilateral encoding path, which simultaneously extracts independent features of bi-temporal multispectral images without introducing additional parameters. A selective convolution kernel block (SCKB) that can adaptively assign weights is designed and embedded in the encoding path of MSAK-Net to extract multi-scale features in images. MSAK-Net retains the skip connections in the U-Net, and embeds an upsampling module (UM) based on the attention mechanism in the decoding path, which can give the feature map a better expression of change information in both the channel dimension and the spatial dimension. Finally, the multimodal conditional random field (MCRF) is used to smooth the detection results of the MSAK-Net. Experimental results on two public multispectral datasets indicate the effectiveness and robustness of the proposed method when compared with other state-of-the-art methods.
TL;DR: Zhang et al. as discussed by the authors proposed a novel method based on relational graph convolutional network (RGCN) to utilize multi-source knowledge in a unified manner for Chinese Clinical NER.
TL;DR: In this paper , an RGB-D SLAM approach was proposed for accurate camera pose tracking in dynamic environments. But the method is limited to a short time-span of consecutive frames.
Abstract: Accurate camera pose estimation is essential and challenging for real world dynamic 3D reconstruction and augmented reality applications. In this article, we present a novel RGB-D SLAM approach for accurate camera pose tracking in dynamic environments. Previous methods detect dynamic components only across a short time-span of consecutive frames. Instead, we provide a more accurate dynamic 3D landmark detection method, followed by the use of long-term consistency via conditional random fields, which leverages long-term observations from multiple frames. Specifically, we first introduce an efficient initial camera pose estimation method based on distinguishing dynamic from static points using graph-cut RANSAC. These static/dynamic labels are used as priors for the unary potential in the conditional random fields, which further improves the accuracy of dynamic 3D landmark detection. Evaluation using the TUM and Bonn RGB-D dynamic datasets shows that our approach significantly outperforms state-of-the-art methods, providing much more accurate camera trajectory estimation in a variety of highly dynamic environments. We also show that dynamic 3D reconstruction can benefit from the camera poses estimated by our RGB-D SLAM approach.
18 Jan 2022
TL;DR: In this article , transfer learning was used to solve the problem of nested named-entity recognition using the transfer-learning approach and different variants of fine-tuned, pretrained, BERT-based language models were used for the problem using the joint labeling modeling technique.
Abstract: Named-entity recognition (NER) is one of the primary components in various natural language processing tasks such as relation extraction, information retrieval, question answering, etc. The majority of the research work deals with flat entities. However, it was observed that the entities were often embedded within other entities. Most of the current state-of-the-art models deal with the problem of embedded/nested entity recognition with very complex neural network architectures. In this research work, we proposed to solve the problem of nested named-entity recognition using the transfer-learning approach. For this purpose, different variants of fine-tuned, pretrained, BERT-based language models were used for the problem using the joint-labeling modeling technique. Two nested named-entity-recognition datasets, i.e., GENIA and GermEval 2014, were used for the experiment, with four and two levels of annotation, respectively. Also, the experiments were performed on the JNLPBA dataset, which has flat annotation. The performance of the above models was measured using F1-score metrics, commonly used as the standard metrics to evaluate the performance of named-entity-recognition models. In addition, the performance of the proposed approach was compared with the conditional random field and the Bi-LSTM-CRF model. It was found that the fine-tuned, pretrained, BERT-based models outperformed the other models significantly without requiring any external resources or feature extraction. The results of the proposed models were compared with various other existing approaches. The best-performing BERT-based model achieved F1-scores of 74.38, 85.29, and 80.68 for the GENIA, GermEval 2014, and JNLPBA datasets, respectively. It was found that the transfer learning (i.e., pretrained BERT models after fine-tuning) based approach for the nested named-entity-recognition task could perform well and is a more generalized approach in comparison to many of the existing approaches.
TL;DR: Wang et al. as discussed by the authors built a connected network of hazards and accidents, forming a knowledge graph (KG), and applied it to railway hazard identification and risk assessment, and the results showed that this approach realizes the visualization and quantitative description of the potential relationship among hazards, faults, and accidents by exploring the topological relationship of the railway accident network, further assisting the formulation of railway risk preventive measures.
Abstract: To clarify the risk factors and propagation characteristics affecting railway safety, we learn from historical reports to build a connected network of hazards and accidents, forming a knowledge graph (KG), and apply it to railway hazard identification and risk assessment. First, the open source-British railway accident/incident reports are selected as the data source. The text augmentation algorithm in the text mining technology is introduced and optimized to achieve data enhancement. An ensemble model is constructed based on the hidden Markov model, conditional random field (CRF) algorithm, bidirectional long short-term memory (Bi-LSTM), and Bi-LSTM-CRF deep learning network, completing the named entity recognition of the reports. Then, using the random forest algorithm, the standardized classification of entities is accomplished, and the multi-dimensional knowledge graph network is established. Finally, after defining a series of safety-related feature parameters, the obtained KG is applied to the quantitative assessment of the corresponding risk level of the hazards. The results show that this approach realizes the visualization and quantitative description of the potential relationship among hazards, faults, and accidents by exploring the topological relationship of the railway accident network, further assisting the formulation of railway risk preventive measures.
TL;DR: This work proposes a new system capable of real-time ship instance segmentation during maritime surveillance missions by unmanned aerial vehicles using an onboard standard RGB camera and creates a new dataset comprising synthetic video sequences of maritime surveillance scenarios (MarSyn).
Abstract: This work proposes a new system capable of real-time ship instance segmentation during maritime surveillance missions by unmanned aerial vehicles using an onboard standard RGB camera. The implementation requires two stages: an instance segmentation network able to produce fast and reliable preliminary segmentation results and a post-processing 3D fully connected Conditional Random Field, which significantly improves segmentation results by exploring temporal correlations between nearby frames in video sequences. Moreover, due to the absence of maritime datasets consisting of properly labeled video sequences, we create a new dataset comprising synthetic video sequences of maritime surveillance scenarios (MarSyn). The main advantages of this approach are the possibility of generating a vast set of images and videos, being able to represent real-world scenarios without the necessity of deploying the real vehicle, and automatic labels, which eliminate human labeling errors. We train the system with the MarSyn dataset and with aerial footage from publicly available annotated maritime datasets to validate the proposed approach. We present some experimental results and compare them to other approaches, and we also illustrate the temporal stability provided by the second stage in missing frames and wrong segmentation scenarios.
TL;DR: The authors used reference information and applied two typical methods of unsupervised extraction methods (TF*IDF and TextRank), two representative traditional supervised learning algorithms (Naïve Bayes and Conditional Random Field) and a supervised deep learning model (BiLSTM-CRF) to analyze the specific performance of reference information on keyphrase extraction.
Abstract: With the development of Internet technology, the phenomenon of information overload is becoming more and more obvious. It takes a lot of time for users to obtain the information they need. However, keyphrases that summarize document information highly are helpful for users to quickly obtain and understand documents. For academic resources, most existing studies extract keyphrases through the title and abstract of papers. We find that title information in references also contains author-assigned keyphrases. Therefore, this article uses reference information and applies two typical methods of unsupervised extraction methods (TF*IDF and TextRank), two representative traditional supervised learning algorithms (Naïve Bayes and Conditional Random Field) and a supervised deep learning model (BiLSTM-CRF), to analyze the specific performance of reference information on keyphrase extraction. It is expected to improve the quality of keyphrase recognition from the perspective of expanding the source text. The experimental results show that reference information can increase precision, recall, and F1 of automatic keyphrase extraction to a certain extent. This indicates the usefulness of reference information on keyphrase extraction of academic papers and provides a new idea for the research on automatic keyphrase extraction.
TL;DR: In this paper , the authors proposed two models based on Gated Recurrent Units (GRU) neural networks for aspect-based sentiment analysis (ABSA) and interactive attention network based on bidirectional GRU to identify sentiment polarity toward extracted aspects.
TL;DR: Zhang et al. as discussed by the authors proposed an Adaptive Reward Markov Random Field (ARMRF) layer to model three intuitions on user label relations and assign them different learnable rewards.
Abstract: With the growing popularity of social media, spammers unfairly overpower legitimate users with unwanted content to achieve their illegal purposes, which encourages research on spammer detection. The existing spammer detection methods can be characterized into feature-based and propagation-based detection. However, feature-based methods (e.g., GCN) cannot capture the user's following relations, while propagation-based methods cannot utilize the rich text features. To this end, we consider combining these two methods and propose an Adaptive Reward Markov Random Field(ARMRF) layer. ARMRF layer models three intuitions on user label relations and assign them different learnable rewards. Besides, we learn the reward weights by stacking the ARMRF layer on top of GCN for end-to-end training, and we call the stacked model ARMGCN. To further improve ARMGCNs expressive ability, we propose the Markov-Driven Graph Convolutional Network(MDGCN), which integrates conditional random fields(CRF) and ARMGCN. CRF establishes the label joint probability distribution conditioned features for learning user dependencies, and the distribution can be optimized by a variational EM algorithm. We extensively evaluate the proposed method on two real-world Twitter datasets, and the experimental results demonstrate that MDGCN outperforms the state-of-the-art baselines. In addition, the ARMRF layer can be integrated with existing detection methods to improve performance further.
TL;DR: This work proposes S-NER, a span-based NER model that directly obtains the types of spans by conducting entity type classifications on span semantic representations, which eliminates the requirement for label dependency.
Abstract: Named entity recognition (NER) is a task that seeks to recognize entities in raw texts and is a precondition for a series of downstream NLP tasks. Traditionally, prior NER models use the sequence labeling mechanism which requires label dependency captured by the conditional random fields (CRFs). However, these models are prone to cascade label misclassifications since a misclassified label results in incorrect label dependency, and so some following labels may also be misclassified. To address the above issue, we propose S-NER, a span-based NER model. To be specific, S-NER first splits raw texts into text spans and regards them as candidate entities; it then directly obtains the types of spans by conducting entity type classifications on span semantic representations, which eliminates the requirement for label dependency. Moreover, S-NER has a concise neural architecture in which it directly uses BERT as its encoder and a feed-forward network as its decoder. We evaluate S-NER on several benchmark datasets across three domains. Experimental results demonstrate that S-NER consistently outperforms the strongest baselines in terms of F1-score. Extensive analyses further confirm the efficacy of S-NER.
TL;DR: In this paper , a novel adversarial training-based Lattice LSTM model called AT-Lattice is proposed to address the problem of latent information in the massive textual data (text records).
Abstract: Learning and identifying key concepts from past fault records are essential for us to understand the causes of these faults, which lay the foundation for the fault diagnosis and prognosis. At present, faults in many fields (e.g., rail, automobile, and smart grid) are recorded in textual form. Due to the lack of effective mining and analysis tools, latent information in the massive textual data (text records) has not been fully unearthed. In this paper, a novel Adversarial Training-based Lattice LSTM model called AT-Lattice is proposed to address this problem. In this model, the Named Entity Recognition (NER) is achieved by Lattice LSTM and Conditional Random Field (CRF), where the Lattice LSTM is used to provide sequence information between words, and the CRF is used to get the final entity prediction result. In addition, the Chinese Word Segmentation (CWS) task is introduced to conduct the adversarial training with the NER task. The framework of the adversarial training is able to make full use of the boundary information and filter out the noise caused by the introduced CWS task. More importantly, extensive experiments are conducted on five different train fault datasets collected by a rail transit company. The results demonstrate that the proposed model outperforms the state-of-the-art baselines.
TL;DR: It is shown that DistilProtBert preforms very well on singlet, doublet, and even triplet-shuffled versions of the human proteome, with AUC of 0.92, 0.91, and 0.87, and it is suggested that by examining the small number of false-positive classifications the authors may be able to identify de-novo potential natural-like proteins based on random shuffling of amino acid sequences.
Abstract: Summary Recently, Deep Learning models, initially developed in the field of Natural Language Processing (NLP), were applied successfully to analyze protein sequences. A major drawback of these models is their size in terms of the number of parameters needed to be fitted and the amount of computational resources they require. Recently, “distilled” models using the concept of student and teacher networks have been widely used in NLP. Here, we adapted this concept to the problem of protein sequence analysis, by developing DistilProtBert, a distilled version of the successful ProtBert model. Implementing this approach, we reduced the size of the network and the running time by 50%, and the computational resources needed for pretraining by 98% relative to ProtBert model. Using two published tasks, we showed that the performance of the distilled model approaches that of the full model. We next tested the ability of DistilProtBert to distinguish between real and random protein sequences. The task is highly challenging if the composition is maintained on the level of singlet, doublet and triplet amino acids. Indeed, traditional machine learning algorithms have difficulties with this task. Here, we show that DistilProtBert preforms very well on singlet, doublet, and even triplet-shuffled versions of the human proteome, with AUC of 0.92, 0.91, and 0.87 respectively. Finally, we suggest that by examining the small number of false-positive classifications (i.e., shuffled sequences classified as proteins by DistilProtBert) we may be able to identify de-novo potential natural-like proteins based on random shuffling of amino acid sequences. Availability https://github.com/yarongef/DistilProtBert Contact yaron.geffen@biu.ac.il
TL;DR: Results demonstrate that the CRF contextual refinement algorithm can improve the classification performance of five sleep staging pre-classifiers, including overlapping-based and nonoverlapping-based models, and the algorithm works both on healthy subjects and patients with sleep disorder.
Abstract: Automatic sleep stage classification has gained much attention in recent researches. Various classification algorithms have been proposed for automatic sleep staging, including deep neural networks and traditional machine learning models. However, the output of those models has unreasonable sleep stage transitions, as temporal dependence of sleep stage label of adjacent data segment is ignored. In this article, we propose a novel sleep stage contextual refinement algorithm based on conditional random fields (CRFs). The algorithm works as a post-processing step to rectify the hypnogram produced by sleep staging pre-classifiers. Unreasonable sleep stage transitions can be corrected via our algorithm to further improve the classification performance. We use CNN-based, CNN-LSTM-based, random forest, and two existing sleep staging models UTSN and UTSN-L as pre-classifiers. Our algorithm is evaluated on three sleep datasets, Sleep-EDF-20, DRM-SUB, and SVUH-UCD datasets. Results demonstrate that our CRF contextual refinement algorithm can improve the classification performance of five sleep staging pre-classifiers, including overlapping-based and nonoverlapping-based models, and the algorithm works both on healthy subjects and patients with sleep disorder. When using CNN as pre-classifiers, our algorithm improves the overall accuracy and macro F1-score by 2.5% and 4.7% on Sleep-EDF-20, by 3.6% and 6.6% on DRM-SUB, and by 5.5% and 7.8% on the SVUH-UCD dataset.
TL;DR: In this article , a two-step liver and tumour segmentation method is presented, where a cascade framework is used during the segmentation process, and a fully connected conditional random field (CRF) method is used to refine the liver segmentation result.