Building Large-Scale Deep Learning System for Entity Recognition in E-Commerce Search

Background/aimsTo apply deep learning technology to develop an artificial intelligence (AI) system that can identify vision-threatening conditions in high myopia patients based on optical coherence tomography (OCT) macular images.MethodsIn this cross-sectional, prospective study, a total of 5505 qualified OCT macular images obtained from 1048 high myopia patients admitted to Zhongshan Ophthalmic Centre (ZOC) from 2012 to 2017 were selected for the development of the AI system. The independent test dataset included 412 images obtained from 91 high myopia patients recruited at ZOC from January 2019 to May 2019. We adopted the InceptionResnetV2 architecture to train four independent convolutional neural network (CNN) models to identify the following four vision-threatening conditions in high myopia: retinoschisis, macular hole, retinal detachment and pathological myopic choroidal neovascularisation. Focal Loss was used to address class imbalance, and optimal operating thresholds were determined according to the Youden Index.ResultsIn the independent test dataset, the areas under the receiver operating characteristic curves were high for all conditions (0.961 to 0.999). Our AI system achieved sensitivities equal to or even better than those of retina specialists as well as high specificities (greater than 90%). Moreover, our AI system provided a transparent and interpretable diagnosis with heatmaps.ConclusionsWe used OCT macular images for the development of CNN models to identify vision-threatening conditions in high myopia patients. Our models achieved reliable sensitivities and high specificities, comparable to those of retina specialists and may be applied for large-scale high myopia screening and patient follow-up.

Download Full-text

Deep Learning-Based Named Entity Recognition and Knowledge Graph Construction for Geological Hazards

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi9010015 ◽

2019 ◽

Vol 9 (1) ◽

pp. 15 ◽

Cited By ~ 2

Author(s):

Runyu Fan ◽

Lizhe Wang ◽

Jining Yan ◽

Weijing Song ◽

Yingqian Zhu ◽

...

Keyword(s):

Deep Learning ◽

Large Scale ◽

Conditional Random Field ◽

Named Entity Recognition ◽

Entity Recognition ◽

Knowledge Graph ◽

Geological Hazard ◽

Geological Hazards ◽

Named Entity ◽

Corpus Construction

Constructing a knowledge graph of geological hazards literature can facilitate the reuse of geological hazards literature and provide a reference for geological hazard governance. Named entity recognition (NER), as a core technology for constructing a geological hazard knowledge graph, has to face the challenges that named entities in geological hazard literature are diverse in form, ambiguous in semantics, and uncertain in context. This can introduce difficulties in designing practical features during the NER classification. To address the above problem, this paper proposes a deep learning-based NER model; namely, the deep, multi-branch BiGRU-CRF model, which combines a multi-branch bidirectional gated recurrent unit (BiGRU) layer and a conditional random field (CRF) model. In an end-to-end and supervised process, the proposed model automatically learns and transforms features by a multi-branch bidirectional GRU layer and enhances the output with a CRF layer. Besides the deep, multi-branch BiGRU-CRF model, we also proposed a pattern-based corpus construction method to construct the corpus needed for the deep, multi-branch BiGRU-CRF model. Experimental results indicated the proposed deep, multi-branch BiGRU-CRF model outperformed state-of-the-art models. The proposed deep, multi-branch BiGRU-CRF model constructed a large-scale geological hazard literature knowledge graph containing 34,457 entities nodes and 84,561 relations.

Download Full-text

SDA: Software-defined accelerator for large-scale deep learning system

2016 International Symposium on VLSI Design, Automation and Test (VLSI-DAT) ◽

10.1109/vlsi-dat.2016.7482563 ◽

2016 ◽

Cited By ~ 4

Author(s):

Jian Ouyang

Keyword(s):

Deep Learning ◽

Large Scale ◽

Learning System

Download Full-text

A Hybrid Deep Learning System for Real-World Mobile User Authentication Using Motion Sensors

Sensors ◽

10.3390/s20143876 ◽

2020 ◽

Vol 20 (14) ◽

pp. 3876 ◽

Cited By ~ 3

Author(s):

Tiantian Zhu ◽

Zhengqiu Weng ◽

Guolang Chen ◽

Lei Fu

Keyword(s):

Feature Extraction ◽

Deep Learning ◽

Mobile Devices ◽

Real World ◽

Large Scale ◽

User Authentication ◽

Mobile User ◽

Learning System ◽

Support Vector ◽

Motion Sensors

With the popularity of smartphones and the development of hardware, mobile devices are widely used by people. To ensure availability and security, how to protect private data in mobile devices without disturbing users has become a key issue. Mobile user authentication methods based on motion sensors have been proposed by many works, but the existing methods have a series of problems such as poor de-noising ability, insufficient availability, and low coverage of feature extraction. Based on the shortcomings of existing methods, this paper proposes a hybrid deep learning system for complex real-world mobile authentication. The system includes: (1) a variational mode decomposition (VMD) based de-noising method to enhance the singular value of sensors, such as discontinuities and mutations, and increase the extraction range of the feature; (2) semi-supervised collaborative training (Tri-Training) methods to effectively deal with mislabeling problems in complex real-world situations; and (3) a combined convolutional neural network (CNN) and support vector machine (SVM) model for effective hybrid feature extraction and training. The training results under large-scale, real-world data show that the proposed system can achieve 95.01% authentication accuracy, and the effect is better than the existing frontier methods.

Download Full-text

Deep learning for detecting retinal detachment and discerning macular status using ultra-widefield fundus images

Communications Biology ◽

10.1038/s42003-019-0730-x ◽

2020 ◽

Vol 3 (1) ◽

Cited By ~ 4

Author(s):

Zhongwen Li ◽

Chong Guo ◽

Danyao Nie ◽

Duoru Lin ◽

Yi Zhu ◽

...

Keyword(s):

Deep Learning ◽

Retinal Detachment ◽

Large Scale ◽

Global Scale ◽

Clinical Applications ◽

Learning System ◽

Vision Impairment ◽

Fundus Images ◽

Severe Visual Loss ◽

Visual Results

AbstractRetinal detachment can lead to severe visual loss if not treated timely. The early diagnosis of retinal detachment can improve the rate of successful reattachment and the visual results, especially before macular involvement. Manual retinal detachment screening is time-consuming and labour-intensive, which is difficult for large-scale clinical applications. In this study, we developed a cascaded deep learning system based on the ultra-widefield fundus images for automated retinal detachment detection and macula-on/off retinal detachment discerning. The performance of this system is reliable and comparable to an experienced ophthalmologist. In addition, this system can automatically provide guidance to patients regarding appropriate preoperative posturing to reduce retinal detachment progression and the urgency of retinal detachment repair. The implementation of this system on a global scale may drastically reduce the extent of vision impairment resulting from retinal detachment by providing timely identification and referral.

Download Full-text

Predicting statistics of asynchronous SGD parameters for a large-scale distributed deep learning system on GPU supercomputers

2016 IEEE International Conference on Big Data (Big Data) ◽

10.1109/bigdata.2016.7840590 ◽

2016 ◽

Cited By ~ 8

Author(s):

Yosuke Oyama ◽

Akihiro Nomura ◽

Ikuro Sato ◽

Hiroki Nishimura ◽

Yukimasa Tamatsu ◽

...

Keyword(s):

Deep Learning ◽

Large Scale ◽

Learning System

Download Full-text

Deep learning and multimodal target recognition of complex and ambiguous words in automated English learning system

Journal of Intelligent & Fuzzy Systems ◽

10.3233/jifs-189543 ◽

2020 ◽

pp. 1-12

Author(s):

Lijing Diao ◽

Ping Hu

Keyword(s):

Neural Network ◽

Deep Learning ◽

Natural Language ◽

Large Scale ◽

English Education ◽

Target Recognition ◽

Learning System ◽

Practical Significance ◽

Teaching Mode ◽

College English

On the basis of convolution neural network, deep learning algorithm can make the convolution layer convolute the input image to complete the hierarchical expression of feature information, which makes pattern recognition more simple and accurate. Now, in the theory of multimodal discourse analysis, the nonverbal features in communication are studied as a symbol system similar to language. In this paper, the author analyzes the deep learning complexity and multimodal target recognition application in English education system. Multimodal teaching gradually has its practical significance in the process of rich teaching resources. The large-scale application of multimedia technology in college English classroom is conducive to the construction of a real language environment. The simulation results show that the multi-layer and one-dimensional convolution structure of the product neural network can effectively complete many natural language problems, including the tagging of lexical and semantic roles, and thus effectively improve the accuracy of natural language processing. Multimodal teaching mode helps to memorize vocabulary images more deeply. 84% of students think that multi-modal teaching mode is closer to life. Meanwhile, multimedia teaching display is more acceptable. College English teachers should renew their teaching concepts and adapt themselves to the new teaching mode.

Download Full-text

A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis

npj Digital Medicine ◽

10.1038/s41746-021-00417-4 ◽

2021 ◽

Vol 4 (1) ◽

Author(s):

Yongli Xu ◽

Man Hu ◽

Hanruo Liu ◽

Hao Yang ◽

Huaizhou Wang ◽

...

Keyword(s):

Deep Learning ◽

Large Scale ◽

Learning Algorithms ◽

High Sensitivity ◽

Learning System ◽

Small Samples ◽

Fundus Images ◽

Fiber Layer ◽

Glaucoma Diagnosis ◽

Diagnostic Thinking

AbstractThe application of deep learning algorithms for medical diagnosis in the real world faces challenges with transparency and interpretability. The labeling of large-scale samples leads to costly investment in developing deep learning algorithms. The application of human prior knowledge is an effective way to solve these problems. Previously, we developed a deep learning system for glaucoma diagnosis based on a large number of samples that had high sensitivity and specificity. However, it is a black box and the specific analytic methods cannot be elucidated. Here, we establish a hierarchical deep learning system based on a small number of samples that comprehensively simulates the diagnostic thinking of human experts. This system can extract the anatomical characteristics of the fundus images, including the optic disc, optic cup, and appearance of the retinal nerve fiber layer to realize automatic diagnosis of glaucoma. In addition, this system is transparent and interpretable, and the intermediate process of prediction can be visualized. Applying this system to three validation datasets of fundus images, we demonstrate performance comparable to that of human experts in diagnosing glaucoma. Moreover, it markedly improves the diagnostic accuracy of ophthalmologists. This system may expedite the screening and diagnosis of glaucoma, resulting in improved clinical outcomes.

Download Full-text

Mining Microbe-disease Interactions From Literature via a Transfer Learning Model

10.21203/rs.3.rs-558906/v1 ◽

2021 ◽

Author(s):

Chengkun Wu ◽

Xinyi Xiao ◽

Canqun Yang ◽

JinXiang Chen ◽

Jiacai Yi ◽

...

Keyword(s):

Deep Learning ◽

Text Mining ◽

Large Scale ◽

Named Entity Recognition ◽

Learning Model ◽

Entity Recognition ◽

Interaction Extraction ◽

Golden Standard ◽

Data Browsing ◽

Deep Learning Model

Abstract Background: Interactions of microbes and diseases are of great importance for biomedical research. However, large-scale curated databases for microbe-disease interactions are missing, as the amount of related literature is enormous and the curation process is costly and time-consuming. In this paper, we aim to construct a large-scale database for microbe-disease interactions automatically. We attained this goal via applying text mining methods based on a deep learning model with a moderate curation cost. We also built a user-friendly web interface to allow researchers navigate and query desired information. Results: For curation, we manually constructed a golden-standard corpora (GSC) and a sliver-standard corpora (SSC) for microbe-disease interactions. Then we proposed a text mining framework for microbe-disease interaction extraction without having to build a model from scratch. Firstly, we applied named entity recognition (NER) tools to detect microbe and disease mentions from texts. Then we transferred a deep learning model BERE to recognize relations between entities, which was originally built for drug-target interactions or drug-drug interactions. The introduction of SSC for model ne-tuning greatly improves the performance of detection for microbe-disease interactions, with an average reduction in error of approximately 10%. The resulting MDIDB website offers data browsing, custom search for specific diseases or microbes as well as batch download. Conclusions: Evaluation results demonstrate that our method outperform the baseline model (rule-based PKDE4J) with an average F1-score of 73.81%. For further validation, we randomly sampled nearly 1,000 predicted interactions by our model, and manually checked the correctness of each interaction, which gives a 73% accuracy. The MDIDB webiste is freely avaliable throuth http://dbmdi.com/index/

Download Full-text

DeepADEMiner: A Deep Learning Pharmacovigilance Pipeline for Extraction and Normalization of Adverse Drug Effect Mentions on Twitter

10.1101/2020.12.15.20248229 ◽

2020 ◽

Author(s):

Arjun Magge ◽

Elena Tutubalina ◽

Zulfat Miftahutdinov ◽

Ilseyar Alimova ◽

Anne Dirkson ◽

...

Keyword(s):

Social Media ◽

Deep Learning ◽

Large Scale ◽

Named Entity Recognition ◽

Drug Effects ◽

Entity Recognition ◽

Pipeline Architecture ◽

Social Media Data ◽

End To End ◽

Media Data

Objective: Research on pharmacovigilance from social media data has focused on mining adverse drug effects (ADEs) using annotated datasets, with publications generally focusing on one of three tasks: (i) ADE classification, (ii) named entity recognition (NER) for identifying the span of an ADE mentions, and (iii) ADE mention normalization to standardized vocabularies. While the common goal of such systems is to detect ADE signals that can be used to inform public policy, it has been impeded largely by limited end-to-end solutions to the three tasks for large-scale analysis of social media reports for different drugs. Materials and Methods: We present a dataset for training and evaluation of ADE pipelines where the ADE distribution is closer to the average `natural balance' with ADEs present in about 7% of the Tweets. The deep learning architecture involves an ADE extraction pipeline with individual components for all three tasks. Results: The system presented achieved a classification performance of F1 = 0.63, span detection performance of F1 = 0.44 and an end-to-end entity resolution performance of F1 = 0.34 on the presented dataset. Discussion: The performance of the models continue to highlight multiple challenges when deploying pharmacovigilance systems that use social media data. We discuss the implications of such models in the downstream tasks of signal detection and suggest future enhancements. Conclusion: Mining ADEs from Twitter posts using a pipeline architecture requires the different components to be trained and tuned based on input data imbalance in order to ensure optimal performance on the end-to-end resolution task.

Download Full-text