Domain-specific Classifiers Research Articles

A Target Domain-Specific Classifier Weight Partial Transfer Adversarial Network for Bearing Fault Diagnosis

In actual industry applications, the failure categories of practical equipment are usually a subset of laboratory conditions failure categories. Due to the strict constraints, partial transfer learning can address a more practical diagnostic scenario. In view of this, this paper proposes a target domain-specific classifier weight partial transfer adversarial network. Initially, the 1-D convolutional neural network is employed as the basic architecture. By training the domain discriminator and feature generator with an adversarial strategy, the recognition ability of the domain discriminant network and the feature extraction ability of the feature generation network can be enhanced. After that, a weighted learning strategy is introduced to guide the model to learn the cross-domain invariant feature. Also, a specific target domain classifier is utilized to redivide the target domain decision boundary to accurately classify the unlabeled target domain samples. Finally, five mainstream deep neural network methods are taken for comparison using the data from Western Reserve University and the motor-magnetic brake test designed by us. The results show that the proposed method reaches 90.18% and 96.53% classification accuracy on two datasets, respectively, which demonstrates superior performance compared with the state-of-the-art methods.

Masked Image Modeling Auxiliary Pseudo-Label Propagation with a Clustering Central Rectification Strategy for Cross-Scene Classification

Cross-scene classification focuses on setting up an effective domain adaptation (DA) way to transfer the learnable knowledge from source to target domain, which can be reasonably achieved through the pseudo-label propagation procedure. However, it is hard to bridge the objective existing severe domain discrepancy between source and target domains, and thus, there are several unreliable pseudo-labels generated in target domain and involved into pseudo-label propagation procedure, which would lead to unreliable error accumulation to deteriorate the performance of cross-scene classification. Therefore, in this paper, a novel Masked Image Modeling Auxiliary Pseudo-Label Propagation called MIM-AP2 with clustering central rectification strategy is proposed to improve the quality of pseudo-label propagation for cross-scene classification. First, in order to gracefully bridge the domain discrepancy and improve DA representation ability in-domain, a supervised class-token contrastive learning is designed to find the more consistent contextual clues to achieve knowledge transfer learning from source to target domain. At the same time, it is also incorporated with a self-supervised MIM mechanism according to a low random masking ratio to capture domain-specific information for improving the discriminability in-domain, which can lay a solid foundation for high-quality pseudo-label generation. Second, aiming to alleviate the impact of unreliable error accumulation, a clustering central rectification strategy is designed to adaptively update robustness clustering central representations to assist in rectifying unreliable pseudo-labels and learning a superior target domain specific classifier for cross-scene classification. Finally, extensive experiments are conducted on six cross-scene classification benchmarks, and the results are superior to other DA methods. The average accuracy reached 95.79%, which represents a 21.87% improvement over the baseline. This demonstrates that the proposed MIM-AP2 can provide significantly improved performance.

Multi-source domain adaptation based tempo-spatial convolution network for cross-subject EEG classification in RSVP task

Objective. Many subject-dependent methods were proposed for electroencephalogram (EEG) classification in rapid serial visual presentation (RSVP) task, which required a large amount of data from new subject and were time-consuming to calibrate system. Cross-subject classification can realize calibration reduction or zero calibration. However, cross-subject classification in RSVP task is still a challenge. Approach. This study proposed a multi-source domain adaptation based tempo-spatial convolution (MDA-TSC) network for cross-subject RSVP classification. The proposed network consisted of three modules. First, the common feature extraction with multi-scale tempo-spatial convolution was constructed to extract domain-invariant features across all subjects, which could improve generalization of the network. Second, the multi-branch domain-specific feature extraction and alignment was conducted to extract and align domain-specific feature distributions of source and target domains in pairs, which could consider feature distribution differences among source domains. Third, the domain-specific classifier was exploited to optimize the network through loss functions and obtain prediction for the target domain. Main results. The proposed network was evaluated on the benchmark RSVP dataset, and the cross-subject classification results showed that the proposed MDA-TSC network outperformed the reference methods. Moreover, the effectiveness of the MDA-TSC network was verified through both ablation studies and visualization. Significance. The proposed network could effectively improve cross-subject classification performance in RSVP task, and was helpful to reduce system calibration time.

ResMon: Domain-Adaptive Wireless Respiration State Monitoring via Few-Shot Bayesian Deep Learning

Under the outbreak of the COVID-19 pandemic, respiration state monitoring plays an important role in assisting respiratory disease diagnosis and treatment. Thanks to the non-intrusive nature and low deployment cost, Wi-Fi based wireless respiration state monitoring methods have gained increasing popularity. By analyzing the variation of channel state information (CSI) of Wi-Fi signals, the respiration states of a target person under the wireless coverage, such as cough, sneeze, and yawn, can be accurately detected. A major problem of the current wireless respiration state monitoring methods is being overly domain-dependent. That is, a sensing algorithm fine-tuned to a specific device placement and background setting (i.e., a domain) can result in drastic drop in detection accuracy when applied to a dissimilar new domain. To enhance the robustness of wireless sensing and reduce the sensing cost across different domains, we propose in this paper a domain-adaptive Respiration state Monitoring system (ResMon) that achieves highly accurate cross-domain detection performance while requiring very limited labeled samples in the new domain. In a nutshell, the proposed ResMon consists of a source domain meta-training stage and a target domain meta-testing stage. In the meta-training stage, we leverage the rich source domain labeled dataset to train an embedding model as a feature extractor of high-dimensional CSI data measurements. In particular, we apply the statistical Bayesian deep learning technique to improve the generalization performance of the embedding model in cross-domain applications. In the meta-testing stage, we combine the embedding model with a few-shot learning technique to train a domain-specific classifier using very limited labeled samples in the target domain. Experiment results show that the proposed ResMon can achieve on average 87.26% cross-domain detection accuracy in a 4-class respiration state classification task using only 5 labeled samples per class, which significantly outperforms the considered benchmark methods.

Multimodal-Boost: Multimodal Medical Image Super-Resolution Using Multi-Attention Network With Wavelet Transform.

Multimodal medical images are widely used by clinicians and physicians to analyze and retrieve complementary information from high-resolution images in a non-invasive manner. Loss of corresponding image resolution adversely affects the overall performance of medical image interpretation. Deep learning-based single image super resolution (SISR) algorithms have revolutionized the overall diagnosis framework by continually improving the architectural components and training strategies associated with convolutional neural networks (CNN) on low-resolution images. However, existing work lacks in two ways: i) the SR output produced exhibits poor texture details, and often produce blurred edges, ii) most of the models have been developed for a single modality, hence, require modification to adapt to a new one. This work addresses (i) by proposing generative adversarial network (GAN) with deep multi-attention modules to learn high-frequency information from low-frequency data. Existing approaches based on the GAN have yielded good SR results; however, the texture details of their SR output have been experimentally confirmed to be deficient for medical images particularly. The integration of wavelet transform (WT) and GANs in our proposed SR model addresses the aforementioned limitation concerning textons. While the WT divides the LR image into multiple frequency bands, the transferred GAN uses multi-attention and upsample blocks to predict high-frequency components. Additionally, we present a learning method for training domain-specific classifiers as perceptual loss functions. Using a combination of multi-attention GAN loss and a perceptual loss function results in an efficient and reliable performance. Applying the same model for medical images from diverse modalities is challenging, our work addresses (ii) by training and performing on several modalities via transfer learning. Using two medical datasets, we validate our proposed SR network against existing state-of-the-art approaches and achieve promising results in terms of structural similarity index (SSIM) and peak signal-to-noise ratio (PSNR).

Correspondence-Free Domain Alignment for Unsupervised Cross-Domain Image Retrieval

Cross-domain image retrieval aims at retrieving images across different domains to excavate cross-domain classificatory or correspondence relationships. This paper studies a less-touched problem of cross-domain image retrieval, i.e., unsupervised cross-domain image retrieval, considering the following practical assumptions: (i) no correspondence relationship, and (ii) no category annotations. It is challenging to align and bridge distinct domains without cross-domain correspondence. To tackle the challenge, we present a novel Correspondence-free Domain Alignment (CoDA) method to effectively eliminate the cross-domain gap through In-domain Self-matching Supervision (ISS) and Cross-domain Classifier Alignment (CCA). To be specific, ISS is presented to encapsulate discriminative information into the latent common space by elaborating a novel self-matching supervision mechanism. To alleviate the cross-domain discrepancy, CCA is proposed to align distinct domain-specific classifiers. Thanks to the ISS and CCA, our method could encode the discrimination into the domain-invariant embedding space for unsupervised cross-domain image retrieval. To verify the effectiveness of the proposed method, extensive experiments are conducted on four benchmark datasets compared with six state-of-the-art methods.

MASS: A Multisource Domain Adaptation Network for Cross-Subject Touch Gesture Recognition

Touch gesture recognition (TGR) plays a pivotal role in many applications, such as socially assistive robots and embodied telecommunication. However, one obstacle to practicality of existing TGR methods is the individual disparities across subjects. Moreover, a deep neural network trained with multiple existing subjects can easily lead to overfitting for a new subject. Hence, how to mitigate the discrepancies between the new and existing subjects and establish a generalized network for TGR is a significant task to realize reliable human–robot tactile interaction. In this article, a novel framework for Multisource domain Adaptation via Shared-Specific feature projection (MASS) is proposed, which incorporates intradomain discriminant, multidomain discriminant, and cross-domain consistency into a deep learning network for cross-subject TGR. Specifically, the MASS method first extracts the shared features in the common feature space of training subjects, with which a domain-general classifier is built. Then, the specific features of each pair of training and testing subjects are mapped and aligned in their common feature space, and multiple domain-specific classifiers are trained with the specific features. Finally, the domain-general classifier and domain-specific classifiers are ensembled to predict the label for the touch samples of a new subject. Experimental results performed on two datasets show that our proposed MASS method achieves remarkable results for cross-subject TGR. The code of MASS is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/AI-touch/MASS</uri> .

Multi-source alignment domain adaptation with similarity measurement for cross-domain bearing fault diagnosis

Variations in bearing operating conditions may cause a shift in the feature distribution of fault signals, weakening the generalization ability of the fault diagnosis model. Traditional methods with domain adaptation (DA) extract invariant cross-domain fault features, but they are mostly applied to a single-source domain. In the case of multi-source cross-domain transfer, however, determining which source domain has a better transfer effect to the target domain is difficult. In this paper, we aim to combine the transferable knowledge contained in all available source domains to improve the transfer performance. Low-correlation features are often more prone to bias under variable working conditions. The contribution of high-correlation features to the model must be increased. Accordingly, a multi-source alignment DA network with similarity measurement (MADASM) is proposed for this issue. First, DA is performed for each combination of source and target domains, and similarity measurement is introduced to constrain the similarity between the input and the central features. Second, the learned features are inputted into the corresponding domain-specific classifiers, of which the classifier discrepancy loss is then introduced to minimize the difference between the outputs of each domain-specific classifier, resulting in consistent prediction results for classifiers on the target domain. Finally, the average of output probabilities from all the classifier is calculated as the diagnosis result. Experimental results show that MADASM can fully utilize fault category information in multiple source domains to extract fault features with high correlation even in challenging scenarios with unknown data labels and lacking of prior knowledge in target domain.

Multisource Unsupervised Subdomain Adaptation Network for Rotary Machine Fault Diagnosis

Domain adaptation-based methods have been widely developed for fault diagnosis. However, the existing approaches mainly focus on the global distribution alignment of single-to-single domain without considering multiple scenarios, and overlooking the alignment of subdomains, which causes misclassification near the class boundaries. Thus, a multisource unsupervised subdomain adaptation network is proposed in this paper to solve fault diagnosis of rotary machines under multiple and variable working conditions. Using data images as inputs, a domain feature extractor is constructed to extract the domain-invariant features and map each source-target pair into the advanced feature space. Multi-kernel local maximum mean discrepancy is introduced to align subdomains. Moreover, the domain-specific classifiers are applied to diagnose fault category, and diagnosis result is jointly determined by multiple classifiers through weighted decision-making strategy. The experiment results on two sets of rotary machines achieve average accuracy of 98.01% and 96.04%, respectively, which demonstrates the validity of the proposed method.

Adaptive Domain Generalization Via Online Disagreement Minimization.

Deep neural networks suffer from significant performance deterioration when there exists distribution shift between deployment and training. Domain Generalization (DG) aims to safely transfer a model to unseen target domains by only relying on a set of source domains. Although various DG approaches have been proposed, a recent study named DomainBed (Gulrajani and Lopez-Paz, 2020), reveals that most of them do not beat simple empirical risk minimization (ERM). To this end, we propose a general framework that is orthogonal to existing DG algorithms and could improve their performance consistently. Unlike previous DG works that stake on a static source model to be hopefully a universal one, our proposed AdaODM adaptively modifies the source model at test time for different target domains. Specifically, we create multiple domain-specific classifiers upon a shared domain-generic feature extractor. The feature extractor and classifiers are trained in an adversarial way, where the feature extractor embeds the input samples into a domain-invariant space, and the multiple classifiers capture the distinct decision boundaries that each of them relates to a specific source domain. During testing, distribution differences between target and source domains could be effectively measured by leveraging prediction disagreement among source classifiers. By fine-tuning source models to minimize the disagreement at test time, target-domain features are well aligned to the invariant feature space. We verify AdaODM on two popular DG methods, namely ERM and CORAL, and four DG benchmarks, namely VLCS, PACS, OfficeHome, and TerraIncognita. The results show AdaODM stably improves the generalization capacity on unseen domains and achieves state-of-the-art performance.

Source-free unsupervised domain adaptation with maintaining model balance and diversity

Source-free unsupervised domain adaptation (SFUDA) uses the knowledge learned from the source domain pre-trained model to perform the task on the target domain without directly accessing the source domain. However, the source domain pre-trained model is biased towards the source domain but deviates from the target domain, so its pseudo-label on the target domain tends to have a large noise. Thus, we should avoid using noisy pseudo-labels to train the model. In this article, we propose a novel kind of source-free unsupervised domain adaptation with maintaining model balance and diversity (SFMBD), which designs a target domain-specific classifier whose classification boundary is far from the high-density area of the target domain feature distribution. In addition, we keep the model balance and promote the model diversity while maintaining its ability to discriminate the target domain. Experimental evaluation of multiple benchmark datasets illustrates the effectiveness of our proposed approach on SFUDA.

A Novel Multi-Source Domain Adaptation Method with Dempster–Shafer Evidence Theory for Cross-Domain Classification

In this era of big data, Multi-source Domain Adaptation (MDA) becomes more and more popular and is employed to make full use of available source data collected from several different, but related domains. Although multiple source domains provide much information, the processing of domain shifts becomes more challenging, especially in learning a common domain-invariant representation for all domains. Moreover, it is counter-intuitive to treat multiple source domains equally as most existing MDA algorithms do. Therefore, the domain-specific distribution for each source–target domain pair is aligned, respectively. Nevertheless, it is hard to combine adaptation outputs from different domain-specific classifiers effectively, because of ambiguity on the category boundary. Subjective Logic (SL) is introduced to measure the uncertainty (credibility) of each domain-specific classifier, so that MDA could be bridged with DST. Due to the advantage of information fusion, Dempster–Shafer evidence Theory (DST) is utilized to reduce the category boundary ambiguity and output reasonable decisions by combining adaptation outputs based on uncertainty. Finally, extensive comparative experiments on three popular benchmark datasets for cross-domain image classification are conducted to evaluate the performance of the proposed method via various aspects.

Domain consistency regularization for unsupervised multi-source domain adaptive classification

Deep learning-based multi-source unsupervised domain adaptation (MUDA) has been actively studied in recent years. Compared with single-source unsupervised domain adaptation (SUDA), domain shift in MUDA exists not only between the source and target domains but also among multiple source domains. Most existing MUDA algorithms focus on extracting domain-invariant representations among all domains whereas the task-specific decision boundaries among classes are largely neglected. In this paper, we propose an end-to-end trainable network that exploits domain Consistency Regularization for unsupervised Multi-source domain Adaptive classification (CRMA). CRMA aligns not only the distributions of each pair of source and target domains but also that of all domains. For each pair of source and target domains, we employ an intra-domain consistency to regularize a pair of domain-specific classifiers to achieve intra-domain alignment. In addition, we design an inter-domain consistency that targets joint inter-domain alignment among all domains. To address different similarities between multiple source domains and the target domain, we design an authorization strategy that assigns different authorities to domain-specific classifiers adaptively for optimal pseudo label prediction and self-training. Extensive experiments show that CRMA tackles unsupervised domain adaptation effectively under a multi-source setup and achieves superior adaptation consistently across multiple MUDA datasets.

Multi-label, multi-domain learning identifies compounding effects of HIV and cognitive impairment.

Older individuals infected by Human Immunodeficiency Virus (HIV) are at risk for developing HIV-Associated Neurocognitive Disorder (HAND), i.e., from reduced cognitive functioning similar to HIV-negative individuals with Mild Cognitive Impairment (MCI) or to Alzheimer's Disease (AD) if more severely affected. Incompletely understood is how brain structure can serve to differentiate cognitive impairment (CI) in the HIV-positive (i.e., HAND) from the HIV-negative cohort (i.e., MCI and AD). To that end, we designed a multi-label classifier that labels the structural magnetic resonance images (MRI) of individuals by their HIV and CI status via two binary variables. Proper training of such an approach traditionally requires well-curated datasets containing large number of samples for each of the corresponding four cohorts (healthy controls, CI HIV-negative adults a.k.a. CI-only, HIV-positive patients without CI a.k.a. HIV-only, and HAND). Because of the rarity of such datasets, we proposed to improve training of the multi-label classifier via a multi-domain learning scheme that also incorporates domain-specific classifiers on auxiliary single-label datasets specific to either binary label. Specifically, we complement the training dataset of MRIs of the four cohorts (Control: 156, CI-only: 335, HIV-only: 37, HAND: 145) acquired by the Memory and Aging Center at the University of California - San Francisco with a CI-specific dataset only containing MRIs of HIV-negative subjects (Controls: 229, CI-only: 397) from the Alzheimer's Disease Neuroimaging Initiative and an HIV-specific dataset (Controls: 75, HIV-only: 75) provided by SRI International. Based on cross-validation on the UCSF dataset, the multi-domain and multi-label learning strategy leads to superior classification accuracy compared with one-domain or multi-class learning approaches, specifically for the undersampled HIV-only cohort. The 'prediction logits' of CI computed by the multi-label formulation also successfully stratify motor performance among the HIV-positive subjects (including HAND). Finally, brain patterns driving the subject-level predictions across all four cohorts characterize the independent and compounding effects of HIV and CI in the HAND cohort.

Deep Adversarial Domain Adaptation With Few-Shot Learning for Motor-Imagery Brain-Computer Interface

Electroencephalography (EEG) is the most prevalent signal acquisition technique for brain-computer interface (BCI). However, the statistical distribution of EEG data varies across subjects and sessions, resulting in poor generalization of the domain-specific classifier. Although the collection of a large number of recordings may alleviate this issue, it is often impractical and not user-friendly. This study proposes the integration of deep domain adaptation with few-shot learning to address the challenge by leveraging the knowledge from multiple source subjects to enhance the performance of a single target subject. The framework incorporated 3 modules: a feature extractor, domain discriminator, and classifier. The feature extractor utilized the available labeled samples with supervised contrastive loss to map the discriminate features onto a deep representation space, where the features from the same class were more similar than those from different classes. The domain discriminator was used to reduce domain drift, through adversarial training. The classifier predicted the user motor intention, based on EEG features. The framework was extensively evaluated through the BCI Competition IV Datasets 2a and 2b. The results of this study indicate that the framework is capable of enhancing the BCI performance and potentially decreases the calibration effort compared to the traditional approach, but the major limitation of this method is that it requires meticulous selection of source subjects.

Learning Target-Domain-Specific Classifier for Partial Domain Adaptation.

Unsupervised domain adaptation (UDA) aims at reducing the distribution discrepancy when transferring knowledge from a labeled source domain to an unlabeled target domain. Previous UDA methods assume that the source and target domains share an identical label space, which is unrealistic in practice since the label information of the target domain is agnostic. This article focuses on a more realistic UDA scenario, i.e., partial domain adaptation (PDA), where the target label space is subsumed to the source label space. In the PDA scenario, the source outliers that are absent in the target domain may be wrongly matched to the target domain (technically named negative transfer), leading to performance degradation of UDA methods. This article proposes a novel target-domain-specific classifier learning-based domain adaptation (TSCDA) method. TSCDA presents a soft-weighed maximum mean discrepancy criterion to partially align feature distributions and alleviate negative transfer. Also, it learns a target-specific classifier for the target domain with pseudolabels and multiple auxiliary classifiers to further address the classifier shift. A module named peers-assisted learning is used to minimize the prediction difference between multiple target-specific classifiers, which makes the classifiers more discriminant for the target domain. Extensive experiments conducted on three PDA benchmark data sets show that TSCDA outperforms other state-of-the-art methods with a large margin, e.g., 4% and 5.6% averagely on Office-31 and Office-Home, respectively.

Automated classification of cancer morphology from Italian pathology reports using Natural Language Processing techniques: A rule-based approach

Pathology reports represent a primary source of information for cancer registries. Hospitals routinely process high volumes of free-text reports, a valuable source of information regarding cancer diagnosis for improving clinical care and supporting research. Information extraction and coding of textual unstructured data is typically a manual, labour-intensive process. There is a need to develop automated approaches to extract meaningful information from such texts in a reliable and accurate way. In this scenario, Natural Language Processing (NLP) algorithms offer a unique opportunity to automatically encode the unstructured reports into structured data, thus representing a potential powerful alternative to expensive manual processing. However, notwithstanding the increasing interest in this area, there is still limited availability of NLP approaches for pathology reports in languages other than English, including Italian, to date. The aim of our work was to develop an automated algorithm based on NLP techniques, able to identify and classify the morphological content of pathology reports in the Italian language with micro-averaged performance scores higher than 95%. Specifically, a novel, domain-specific classifier that uses linguistic rules was developed and tested on 27,239 pathology reports from a single Italian oncological centre, following the International Classification of Diseases for Oncology morphology classification standard (ICD-O-M). The proposed classification algorithm achieved successful results with a micro-F1 score of 98.14% on 9594 pathology reports in the test dataset. This algorithm relies on rules defined on data from a single hospital that is specifically dedicated to cancer, but it is based on general processing steps which can be applied to different datasets. Further research will be important to demonstrate the generalizability of the proposed approach on a larger corpus from different hospitals.

Fault Diagnosis of Rolling Bearings of Different Working Conditions Based on Multi-Feature Spatial Domain Adaptation

The running state of rolling bearings is complex in operation, and the data are generally collected under different working conditions. However, when single-source domain adaptation is used to model the heterogeneously distributed data obtained under different working conditions, the domain-invariant representations can hardly be used for representation, which directly affects the fault diagnosis rate. To this end, a method for the fault diagnosis of rolling bearings under different working conditions based on multi-feature spatial domain adaptation is proposed. Firstly, all the data from source and target domains are mapped into a feature space to learn the common representations of all domains. Secondly, the data for each pair of source and target domains are mapped into different feature spaces to get the fault feature representations under various working conditions. And the multi-domain adaptation network is used for the domain-specific distribution alignment to learn multiple domain-invariant representations. Thirdly, these representations are used to train multiple domain-specific classifiers, thus obtaining the recognition result for each domain-invariant representation. Finally, the domain-specific decision boundaries predicted by multiple classifiers are employed to align the classifiers’ output of target samples and thus to reduce the influence from different classifiers. The effectiveness and feasibility of this proposed method have been verified by diagnostic experiments conducted according to the rolling bearing data from Case Western Reserve University and Laboratory, respectively.

Unsupervised Domain Adaptation via Structured Prediction Based Selective Pseudo-Labeling

Unsupervised domain adaptation aims to address the problem of classifying unlabeled samples from the target domain whilst labeled samples are only available from the source domain and the data distributions are different in these two domains. As a result, classifiers trained from labeled samples in the source domain suffer from significant performance drop when directly applied to the samples from the target domain. To address this issue, different approaches have been proposed to learn domain-invariant features or domain-specific classifiers. In either case, the lack of labeled samples in the target domain can be an issue which is usually overcome by pseudo-labeling. Inaccurate pseudo-labeling, however, could result in catastrophic error accumulation during learning. In this paper, we propose a novel selective pseudo-labeling strategy based on structured prediction. The idea of structured prediction is inspired by the fact that samples in the target domain are well clustered within the deep feature space so that unsupervised clustering analysis can be used to facilitate accurate pseudo-labeling. Experimental results on four datasets (i.e. Office-Caltech, Office31, ImageCLEF-DA and Office-Home) validate our approach outperforms contemporary state-of-the-art methods.

Combining Lexical and Syntactic Features for Detecting Content-Dense Texts in News

Content-dense news report important factual information about an event in direct, succinct manner. Information seeking applications such as information extraction, question answering and summarization normally assume all text they deal with is content-dense. Here we empirically test this assumption on news articles from the business, U.S. international relations, sports and science journalism domains. Our findings clearly indicate that about half of the news texts in our study are in fact not content-dense and motivate the development of a supervised content-density detector. We heuristically label a large training corpus for the task and train a two-layer classifying model based on lexical and unlexicalized syntactic features. On manually annotated data, we compare the performance of domain-specific classifiers, trained on data only from a given news domain and a general classifier in which data from all four domains is pooled together. Our annotation and prediction experiments demonstrate that the concept of content density varies depending on the domain and that naive annotators provide judgement biased toward the stereotypical domain label. Domain-specific classifiers are more accurate for domains in which content-dense texts are typically fewer. Domain independent classifiers reproduce better naive crowdsourced judgements. Classification prediction is high across all conditions, around 80%.