Information Extraction Method Research Articles

A Fine-Tuned Bidirectional Encoder Representations From Transformers Model for Food Named-Entity Recognition: Algorithm Development and Validation.

BackgroundRecently, food science has been garnering a lot of attention. There are many open research questions on food interactions, as one of the main environmental factors, with other health-related entities such as diseases, treatments, and drugs. In the last 2 decades, a large amount of work has been done in natural language processing and machine learning to enable biomedical information extraction. However, machine learning in food science domains remains inadequately resourced, which brings to attention the problem of developing methods for food information extraction. There are only few food semantic resources and few rule-based methods for food information extraction, which often depend on some external resources. However, an annotated corpus with food entities along with their normalization was published in 2019 by using several food semantic resources.ObjectiveIn this study, we investigated how the recently published bidirectional encoder representations from transformers (BERT) model, which provides state-of-the-art results in information extraction, can be fine-tuned for food information extraction.MethodsWe introduce FoodNER, which is a collection of corpus-based food named-entity recognition methods. It consists of 15 different models obtained by fine-tuning 3 pretrained BERT models on 5 groups of semantic resources: food versus nonfood entity, 2 subsets of Hansard food semantic tags, FoodOn semantic tags, and Systematized Nomenclature of Medicine Clinical Terms food semantic tags.ResultsAll BERT models provided very promising results with 93.30% to 94.31% macro F1 scores in the task of distinguishing food versus nonfood entity, which represents the new state-of-the-art technology in food information extraction. Considering the tasks where semantic tags are predicted, all BERT models obtained very promising results once again, with their macro F1 scores ranging from 73.39% to 78.96%.ConclusionsFoodNER can be used to extract and annotate food entities in 5 different tasks: food versus nonfood entities and distinguishing food entities on the level of food groups by using the closest Hansard semantic tags, the parent Hansard semantic tags, the FoodOn semantic tags, or the Systematized Nomenclature of Medicine Clinical Terms semantic tags.

RETRACTED ARTICLE: Application of land ecological environment risk assessment based on SAR image

The ecological environment of land is an important part of the ecosystem. Extracting land eco-environmental information is very important to realize environmental change monitoring and sustainable development. In the past, people used surface measurement and optical remote sensing technology to extract land ecological environment information. According to the data measured by the surface survey method, the accuracy is relatively low, and it is difficult to meet the application needs of people. However, because of its own characteristics, optical remote sensing is easy to be disturbed by the weather and cannot extract the land ecological environment information timely and accurately. The development of SAR technology provides a new way to extract land ecological environment information. This technology cannot be affected by cloud, rain, and fog and can be used for all day monitoring of land ecological environment risk assessment. In addition, the full polarimetric SAR image directly opens up a new method of land ecological information extraction. According to Wishart H/α, the classification algorithm extracts the land eco-environmental cover information from the full polarimetric SAR image. The classification algorithm is applied to all the polarization data in a certain area, and the coverage information is extracted according to the scattering mechanism of the terrestrial ecological environment. After comparing the extraction accuracy with the optical image, it is found that the land ecological environment coverage information is extracted by different methods under different visual values of multi-view processing. Using error analysis to improve the classification algorithm, improve the accuracy of land eco-environmental cover information extraction, the past accuracy of 54.0% to 81.7%, and the accuracy of 73.1% to 88.0%. Based on the impact of the application of SAR image in land ecological environment risk assessment, a new concept assessment model is established. Through this method, the past evaluation model can be improved.

Multiple annotation for biodiversity: developing an annotation framework among biology, linguistics and text technology

Biodiversity information is contained in countless digitized and unprocessed scholarly texts. Although automated extraction of these data has been gaining momentum for years, there are still innumerable text sources that are poorly accessible and require a more advanced range of methods to extract relevant information. To improve the access to semantic biodiversity information, we have launched the BIOfid project (www.biofid.de) and have developed a portal to access the semantics of German language biodiversity texts, mainly from the 19th and 20th century. However, to make such a portal work, a couple of methods had to be developed or adapted first. In particular, text-technological information extraction methods were needed, which extract the required information from the texts. Such methods draw on machine learning techniques, which in turn are trained by learning data. To this end, among others, we gathered the biofid text corpus, which is a cooperatively built resource, developed by biologists, text technologists, and linguists. A special feature of biofid is its multiple annotation approach, which takes into account both general and biology-specific classifications, and by this means goes beyond previous, typically taxon- or ontology-driven proper name detection. We describe the design decisions and the genuine Annotation Hub Framework underlying the biofid annotations and present agreement results. The tools used to create the annotations are introduced, and the use of the data in the semantic portal is described. Finally, some general lessons, in particular with multiple annotation projects, are drawn.

Adaptive parameters optimization model with 3D information extraction for infrared small target detection based on particle swarm optimization algorithm

Traditional small target detection methods could not adjust the parameters adaptively to achieve small target detection in real scene applications. To solve this problem, a novel small target detection method named Adaptive parameters optimization model with three-dimensional (3D) Information Extraction (A3DIE) is proposed. In this method, a multi-objective particle swarm optimization algorithm is designed to optimize the parameters in the 3D Information Extraction (3DIE) method and realize the adaptability of detection method in different detection scenarios. In the optimization algorithm, an adaptive environment selection strategy is developed to enhance the evolutionary ability and obtain high-quality solution sets. Furthermore, a knee point selection strategy is designed to obtain the optimal small target detection method parameters. The experimental results indicate that our method can accurately and stably detect small targets in different real scenes compared with the baseline methods.

Research on Intelligent Inspection Method of Transmission Channel Satellite Remote Sensing Environmental Changes

With the shortage of transmission channel resources and the increase of human activities, the inspection of transmission channel environment changes is of great significance to the protection of line breakage, the management of transmission channel resources and the detection of hidden dangers in advance. In order to explore a fast and intelligent extraction method of environmental change information in the transmission channel, this paper takes a transmission line section in Tongzhou District of Beijing as the experimental area, takes two high-resolution satellite images of Beijing No.2 as the data source, and conducts principal component analysis and vegetation index processing analysis on the two satellite image data. An intelligent patrol method for transmission channel environmental change based on multi-source feature multi-scale segmentation technology and object-oriented classification algorithm is proposed. The results show that the satellite remote sensing environmental change intelligent patrol method proposed in this paper is sensitive and effective to the human activity change patrol of the transmission channel, and can be extended to the large-scale and normal intelligent patrol operation of the transmission channel against external breach.

Design of Networked Intelligent Translation System Based on Machine Learning Algorithm

China’s social and economic development is relatively rapid, and China’s opening to the outside world is further expanding. Under this background, the demand for foreign language professionals is also expanding. This paper designs a networked intelligent translation system based on machine learning algorithm, and proposes a cross-language information extraction method based on bilingual word representation. The learning of bilingual word representation can be divided into two stages: unsupervised and supervised. Capturing Chinese and English bilingual semantic information. The cross-language information extraction method based on machine learning algorithm reduces the impact of translation errors and language gaps on cross-language classification performance, obtains effective bilingual semantic information and annotation information, and improves the cross-language information extraction performance. The test results show that the designed automatic English translation system can effectively and quickly realize intelligent English translation of memory-assisted long characters, with high data recall rate, good accuracy and reliability, and good system compatibility, which is worthy of popularization and practical application.

A Graph Database Representation of Portuguese Criminal-Related Documents

Organizations have been challenged by the need to process an increasing amount of data, both structured and unstructured, retrieved from heterogeneous sources. Criminal investigation police are among these organizations, as they have to manually process a vast number of criminal reports, news articles related to crimes, occurrence and evidence reports, and other unstructured documents. Automatic extraction and representation of data and knowledge in such documents is an essential task to reduce the manual analysis burden and to automate the discovering of names and entities relationships that may exist in a case. This paper presents SEMCrime, a framework used to extract and classify named-entities and relations in Portuguese criminal reports and documents, and represent the data retrieved into a graph database. A 5WH1 (Who, What, Why, Where, When, and How) information extraction method was applied, and a graph database representation was used to store and visualize the relations extracted from the documents. Promising results were obtained with a prototype developed to evaluate the framework, namely a name-entity recognition with an F-Measure of 0.73, and a 5W1H information extraction performance with an F-Measure of 0.65.

Extraction of metamorphic minerals by multiscale segmentation combined with random forest

The identification of metamorphic minerals is the basis of metamorphic rock research. Extraction of mineral information by remote sensing technology has been widely used. Digital image processing technology is also effectively applied to remote sensing image processing. Results show that the band ratio of remote sensing images can enhance mineral information, while the variogram function can describe the spatial correlation and variability of image pixels and extract more detailed texture information. The metamorphic minerals are found to present a block or strip distribution. The object-oriented remote sensing image information extraction method can avoid the salt and pepper phenomenon based on pixel extraction. Meanwhile, the random forest classification method has a fast calculation speed and high parameter accuracy. It is not sensitive to the noise caused by more lithologic components and its classification effect is found to be stable. To improve the extraction accuracy of metamorphic minerals from remote sensing images and further improve the recognition effect of metamorphic zones, this paper combined the ratio operation, multiscale segmentation, and random forest classification to extract metamorphic mineral information from ASTER images in Beishan area in Gansu Province. Initially, the image was enhanced by the ratio formula of the characteristic spectral structure of the target mineral. Multiscale image segmentation was then performed based on the spectrum and variogram. Finally, the accuracy was evaluated by the thin film identification results of the field exploration samples after the extraction of the target mineral by random forest. Results show that biotite, muscovite, and amphibole have identification characteristics on the ASTER image with an extraction accuracy of 85.4088%, 84.7640%, and 85.7308%, respectively. The extraction accuracy of other metamorphic minerals with less content are found to reach more than 60%. Multiscale segmentation can make full use of the clustering features of minerals and the variogram texture can enhance the ability of morphological features to distinguish the minerals. Random forest is not sensitive to noise and the extraction results are observed to be stable.

LEARNING COMPACT DNN MODELS FOR BEHAVIOR PREDICTION FROM NEURAL ACTIVITY OF CALCIUM IMAGING.

In this paper, we develop methods for efficient and accurate information extraction from calcium-imaging-based neural signals. The particular form of information extraction we investigate involves predicting behavior variables linked to animals from which the calcium imaging signals are acquired. More specifically, we develop algorithms to systematically generate compact deep neural network (DNN) models for accurate and efficient calcium-imaging-based predictive modeling. We also develop a software tool, called NeuroGRS, to apply the proposed methods for compact DNN derivation with a high degree of automation. GRS stands for Greedy inter-layer order with Random Selection of intra-layer units, which describes the central algorithm developed in this work for deriving compact DNN structures. Through extensive experiments using NeuroGRS and calcium imaging data, we demonstrate that our methods enable highly streamlined information extraction from calcium images of the brain with minimal loss in accuracy compared to much more computationally expensive approaches.

Hybrid deep learning model for automating constraint modelling in advanced working packaging

Management of constraints (e.g. materials and labour) is a major challenge in construction projects. Advanced working packaging (AWP) is an effective constraint-management method. However, one prerequisite for AWP, i.e. constraint modelling, is generally performed manually. Information extraction methods in the industry cannot meet the demands for AWP, because they focus on entity extraction but ignore extraction of semantically rich relations. To address this problem, this study proposes a hybrid deep learning model. A bidirectional long short-term memory and conditional random field (Bi-LSTM-CRF) model and knowledge representation learning (KRL) model are developed to extract entities and relations among entities from text documents, respectively. To better apply the KRL model, the study maps domain classes of entities and then stacks class information in the model structure, while employing synonym mapping to handle entities unseen during training. The overall accuracies for extracting entities and relations can reach 0.936 and 0.884, respectively, and adding class information increases relation extraction performance metrics by 6.63%. In a scenario implementation, it is shown that the model can automate constraint modelling for ongoing projects. Therefore, the model is useful for AWP and can reduce delays and reworks by saving a significant amount of time for constraint monitoring and removal.

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.

Machine learning-based LIBS spectrum analysis of human blood plasma allows ovarian cancer diagnosis.

Early-stage screening and diagnosis of ovarian cancer represent an urgent need in medicine. Usual ultrasound imaging and cancer antigen CA-125 test when prescribed to a suspicious population still require reconfirmations. Spectroscopic analyses of blood, at the molecular and atomic levels, provide useful supplementary tests when coupled with effective information extraction methods. Laser-induced breakdown spectroscopy (LIBS) was employed in this work to record the elemental fingerprint of human blood plasma. A machine learning data treatment process was developed combining feature selection and regression with a back-propagation neural network, resulting in classification models for cancer detection among 176 blood plasma samples collected from patients, including also ovarian cyst and normal cases. Cancer diagnosis sensitivity and specificity of respectively 71.4% and 86.5% were obtained for randomly selected validation samples.

Structured information extraction for bone scan image report texts

Extracting structured information from the bone scan image report text plays a crucial role in supporting clinical analysis and research. This study summarized the structure and characteristics of 3608 bone scan image report text using dictionary-based information extraction method, including data cleaning, entity recognition, building dictionary and extraction rules. This method was used to obtain the structured data of bone scan image report text required for clinical research, and the effect evaluation was carried out on 1000 randomly selected report texts, with the precision rate and recall rate higher than 90%. The method proposed in this study is practical and could have good effect on structured results for bone scan imaging report text.

Lecture Information Service Based on Multiple Features Fusion

Information service is always a hot topic especially when the Web is accessible anywhere. In university, lecture information is very important for students and teachers who want to take part in academic meetings. Therefore, lecture news extraction is an important and imperative task. Many open information extraction methods have been proposed, but due to the high heterogeneity of websites, this task is still a challenge. In this paper, we propose a method based on fusing multiple features to locate lecture news on the university website. These features include the linked relationship between parent webpage and child webpages, the visual similarity, and the semantics of webpages. Additionally, this paper provides an information service based on a main content extraction algorithm for extracting the lecture information. Stable and invariant features enable the proposed method to adapt to various kinds of campus websites. The experiments conducted on 50 websites show the effectiveness and efficiency of the provided service.

A Novel Hierarchical Model in Ensemble Environment for Road Detection Application

As a popular research direction in the field of intelligent transportation, road detection has been extensively concerned by many researchers. However, there are still some key issues in specific applications that need to be further improved, such as the feature processing of road images, the optimal choice of information extraction and detection methods, and the inevitable limitations of detection schemes. In the existing research work, most of the image segmentation algorithms applied to road detection are sensitive to noise data and are prone to generate redundant information or over-segmentation, which makes the computation of segmentation process more complicated. In addition, the algorithm needs to overcome objective factors such as different road conditions and natural environments to ensure certain execution efficiency and segmentation accuracy. In order to improve these issues, we integrate the idea of shallow machine-learning model that clusters first and then classifies in this paper, and a hierarchical multifeature road image segmentation integration framework is proposed. The proposed model has been tested and evaluated on two sets of road datasets based on real scenes and compared with common detection methods, and its effectiveness and accuracy have been verified. Moreover, it demonstrates that the method opens up a new way to enhance the learning and detection capabilities of the model. Most importantly, it has certain potential for application in various practical fields such as intelligent transportation or assisted driving.

Measuring the Impact of Natural Hazards with Citizen Science: The Case of Flooded Area Estimation Using Twitter

Twitter has significant potential as a source of Volunteered Geographic Information (VGI), as its content is updated at high frequency, with high availability thanks to dedicated interfaces. However, the diversity of content types and the low average accuracy of geographic information attached to individual tweets remain obstacles in this context. The contributions in this paper relate to the general goal of extracting actionable information regarding the impact of natural hazards on a specific region from social platforms, such as Twitter. Specifically, our contributions describe the construction of a model classifying whether given spatio-temporal coordinates, materialized by raster cells in a remote sensing context, lie in a flooded area. For training, remotely sensed data are used as the target variable, and the input covariates are built on the sole basis of textual and spatial data extracted from a Twitter corpus. Our contributions enable the use of trained models for arbitrary new Twitter corpora collected for the same region, but at different times, allowing for the construction of a flooded area measurement proxy available at a higher temporal frequency. Experimental validation uses true data that were collected during Hurricane Harvey, which caused significant flooding in the Houston urban area between mid-August and mid-September 2017. Our experimental section compares several spatial information extraction methods, as well as various textual representation and aggregation techniques, which were applied to the collected Twitter data. The best configuration yields a F1 score of 0.425, boosted to 0.834 if restricted to the 10% most confident predictions.

Bunch-by-bunch three-dimensional position and charge measurement in a storage ring

Bunch-by-bunch beam parameters monitor is a very useful tool for accelerator operation and optimization. An in situ bunch-by-bunch three-dimensional position and charge measurement system has been developed at the Shanghai Synchrotron Radiation Facility. Signals from beam position monitors carry almost full beam information, a broadband digital receiver can retain most information. Therefore, a four-channel wideband oscilloscope is employed to capture the signals from all electrodes of the beam position monitor. Correlation coefficient-based methods are used to extract these bunch-by-bunch parameters synchronously. We evaluated the performance of this bunch-by-bunch information extraction method, the measurement uncertainty of longitudinal phase is less than 0.2 ps, the uncertainty of transverse position is less than $10\text{ }\text{ }\ensuremath{\mu}\mathrm{m}$ and the charge uncertainty is 0.3% under the condition that the bunch charge is 600 pC. Thanks to the great performance of the new system, it is accurate enough to study the transient state during injection. In addition, an in situ wakefield measurement system is planned to be established based on this system.

An Overview of Entity Relation Extraction

Plenty of digital text is generated and shared in nowadays life, which constitutes many unstructured text resources. An automatically information extraction method is highly in demand as a large amount of useful information can be stored and output through it. In this paper, several basic concepts about entity relation extraction are introduced, including word embedding, positional embedding and convolutional neutral networks. Several types of relation extraction are also discussed: supervised relation extraction, relation extraction using distant supervision and relation extraction using few-shot approach. In addition, the existing challenges and problems are discussed, like overlapped triples, wrong labeled data and the difficulties that few-shot learning approaches are now facing.

Disease Concept-Embedding Based on the Self-Supervised Method for Medical Information Extraction from Electronic Health Records and Disease Retrieval: Algorithm Development and Validation Study.

BackgroundThe electronic health record (EHR) contains a wealth of medical information. An organized EHR can greatly help doctors treat patients. In some cases, only limited patient information is collected to help doctors make treatment decisions. Because EHRs can serve as a reference for this limited information, doctors’ treatment capabilities can be enhanced. Natural language processing and deep learning methods can help organize and translate EHR information into medical knowledge and experience.ObjectiveIn this study, we aimed to create a model to extract concept embeddings from EHRs for disease pattern retrieval and further classification tasks.MethodsWe collected 1,040,989 emergency department visits from the National Taiwan University Hospital Integrated Medical Database and 305,897 samples from the National Hospital and Ambulatory Medical Care Survey Emergency Department data. After data cleansing and preprocessing, the data sets were divided into training, validation, and test sets. We proposed a Transformer-based model to embed EHRs and used Bidirectional Encoder Representations from Transformers (BERT) to extract features from free text and concatenate features with structural data as input to our proposed model. Then, Deep InfoMax (DIM) and Simple Contrastive Learning of Visual Representations (SimCLR) were used for the unsupervised embedding of the disease concept. The pretrained disease concept-embedding model, named EDisease, was further finetuned to adapt to the critical care outcome prediction task. We evaluated the performance of embedding using t-distributed stochastic neighbor embedding (t-SNE) to perform dimension reduction for visualization. The performance of the finetuned predictive model was evaluated against published models using the area under the receiver operating characteristic (AUROC).ResultsThe performance of our model on the outcome prediction had the highest AUROC of 0.876. In the ablation study, the use of a smaller data set or fewer unsupervised methods for pretraining deteriorated the prediction performance. The AUROCs were 0.857, 0.870, and 0.868 for the model without pretraining, the model pretrained by only SimCLR, and the model pretrained by only DIM, respectively. On the smaller finetuning set, the AUROC was 0.815 for the proposed model.ConclusionsThrough contrastive learning methods, disease concepts can be embedded meaningfully. Moreover, these methods can be used for disease retrieval tasks to enhance clinical practice capabilities. The disease concept model is also suitable as a pretrained model for subsequent prediction tasks.

A dynamic information extraction method for areal mean rainfall error and its application in basins of different scales for flood forecasting

In this study, we proposed and tested a method based on system response curve (SRC) to extract the error information of areal mean rainfall. These extracted rainfall error are then used to update the rainfall time series data and propagated through the model to update the stream-flow forecast. The method is evaluated via synthetic and real-data cases. After our statistical analysis in different application scenarios, we found that the SRC method can effectively improve the accuracy of real-time flood forecasting when an optimal updating width is selected (the average relative error was reduced from 1.65% to 0.86%). In addition, we benchmark our results against a more conventional AR (Auto-Regression) streamflow-updating method. The average accuracy improvement of SRC method is 6.3% higher than that of AR method. More importantly, we found that the optimal updating width of SRC method is highly correlated with the average lead time of the basin, it has guiding significance for selecting the optimal updating width in practical application.