Information Extraction Method Research Articles

Use of Natural Language Processing to Extract Clinical Cancer Phenotypes from Electronic Medical Records.

Current models for correlating electronic medical records with -omics data largely ignore clinical text, which is an important source of phenotype information for patients with cancer. This data convergence has the potential to reveal new insights about cancer initiation, progression, metastasis, and response to treatment. Insights from this real-world data will catalyze clinical care, research, and regulatory activities. Natural language processing (NLP) methods are needed to extract these rich cancer phenotypes from clinical text. Here, we review the advances of NLP and information extraction methods relevant to oncology based on publications from PubMed as well as NLP and machine learning conference proceedings in the last 3 years. Given the interdisciplinary nature of the fields of oncology and information extraction, this analysis serves as a critical trail marker on the path to higher fidelity oncology phenotypes from real-world data.

Feature extraction method of abrasive belt wear state for rail grinding

Abrasive belt has the advantages of high removal efficiency and high waste collection; furthermore, it does not easily damage the rail when applied in rail grinding. However, the poor wear consistency and short service life restrict its development in this field. It is important to explore the theory and technology of monitoring the abrasive belt wear condition to evaluate the means of improving the grinding performance of the belt. This article discusses the contact stress state between the belt and the rail surface. In addition, the study establishes the belt rail grinding force model during operation using the abrasive grain distribution function. The analysis of the force model proposes that the belt abrasion condition be monitored by using the force ratio parameter of the abrasive belt in the grinding process. The extraction method of the force ratio information is proposed in accordance with the grinding environment. The results of experiments successfully verified the validity of the method.

Fusion Method Evaluation and Classification Suitability Study of Wetland Satellite Imagery

Based on HJ-1A HSI data and Landsat-8 OLI data, RS image fusion experiments were carried out using three fusion methods: principal component (PC) transform, Gram Schimdt (GS) transform and nearest neighbor diffusion (NND) algorithm. Four evaluation indexes, namely mean, standard deviation, information entropy and average gradient, were selected to evaluate the fusion results from the aspects of image brightness, clarity and information content. Wetland vegetation was classified by spectral angle mapping (SAM) to find a suitable fusion method for wetland vegetation information extraction. The results show that PC fusion image contains the largest amount of information, GS fusion image has certain advantages in brightness and clarity maintenance, and NND fusion method can retain the spectral characteristics of the image to the maximum extent; Among the three fusion methods, PC transform is the most suitable for wetland information extraction. It can retain more spectral information while improving spatial resolution, with classification accuracy of 89.24% and Kappa coefficient of 0.86.

Random forest classification of Callicarpa nudiflora from WorldView-3 imagery based on optimized feature space

Taking the Xiushui township of Baisha county in Hainan province as the research area,the random forest algorithm with obvious advantages in feature selection and classification extraction was used to extract the information of the Callicarpa nudiflora planting in the study area. Firstly,four kinds of different characteristic variables were generated based on World View-3 data,including spectral features,principal component features,vegetation index and texture features. Secondly,the spatial distribution of the C. nudiflora in the study area was extracted by remote sensing by random forest classification algorithm. Finally,the feature space of the random forest classification algorithm was optimized based on the feature importance to obtain the best random forest classification results,and this result is compared with the classification result of the random forest algorithm of the unoptimized feature space. The results showed that:①The overall accuracy of the C. nudiflora extracted by World View-3 image was 89. 97%,and the Kappa coefficient was 0. 84,which indicates that the random forest algorithm had higher classification accuracy and better applicability in Hainan C. nudiflora recognition.② The overall accuracy of extracting C. nudiflora with the dimension reduction feature was 90. 4,and the Kappa coefficient was 0. 85,which indicates that the random forest algorithm can effectively select features. At the same time as the feature variable data mining,the precision of the information extraction of the C. nudiflora was still guaranteed,and the operation efficiency was improved. This study provides a new idea,method and technical means for information extraction of cultivated medicinal plant resources in terms of feature selection and method selection.

Study of extracting natural resources of Chinese medicinal materials planted area in Luoning of Henan province based on UAV of low altitude remote sensing technology and remote sensing image of satellite

The study is aimed to effectively obtain the planting area of traditional Chinese medicine resources. The herbs used as the material for traditional Chinese medicine are mostly planted in natural environment suitable mountainous areas. The UAV low altitude remote sensing data were used as the samples and the GF-2 remote sensing images were applied for the data source to extract the planting area of Salvia miltiorrhiza and Artemisia argyi in Luoning county combined with field investigation. Remote sensing satellite data of standard processing obtain specific remote sensing data coverage. The UAV data were pre-processed to visually interpret the species and distribution of traditional Chinese medicine resources in the sample quadrat. Support vector machine( SVM) was used to classify and estimate the area of traditional Chinese medicine resources in Luoning county,confusion matrix was used to determine the accuracy of spatial distribution of traditional Chinese medicine resources. The result showed that the application of UAV of low altitude remote sensing technology and remote sensing image of satellite in the extraction of S. miltiorrhiza and other varieties planting area was feasible,it also provides a scientific reference for poverty alleviation policies of the traditional Chinese medicine Industry in local areas.Meanwhile,research on remote sensing classification of Chinese medicinal materials based on multi-source and multi-phase high-resolution remote sensing images is actively carried out to explore more effective methods for information extraction of Chinese medicinal materials.

Unsupervised Extraction of Common Product Attributes From E-Commerce Websites by Considering Client Suggestion

Develop an unsupervised learning framework for extracting popular product attributes from product description pages originated from different E-commerce Web sites. Unlike existing information extraction methods that do not consider the popularity of product attributes, in this proposed framework is able to not only detect popular product features from a collection of customer reviews but also map these popular features to the related product attributes. Building an intelligent E-commerce systems typically involves a component that can automatically extract product attribute information from a variety of product description pages in different E-commerce Web sites. Web information extraction methods such as wrappers are able to automatically extract product attributes from the Web content One novelty in this framework is that it can bridge the vocabulary gap between the text in product description pages and the text in customer reviews. Technically,in this framework developed a discriminative graphical model based on hidden Conditional Random Fields. As an unsupervised model, this framework can be easily applied to a variety of new domains and Web sites without the need of labelling training samples. E-commerce is proposed for enhancing the capability. Covered by electronic commerce surroundings, facing therefore voluminous new recent business model, it's obligatory to conduct the analysis to the electronic commerce pattern analysis method and like this is often useful in North American nation uncover the new electronic commerce pattern as provide the approach for electronic commerce pattern modernization to be conjointly helpful within the enterprise outline the particular electronic commerce strategy and therefore the implementation step. Initiated from this encouragement, during this paper proposes the innovative construct of the E-commerce recent agricultural product selling supported the massive web knowledge platform later the rapid development of rebuilding and opening up, China's agriculture has entered a new historical stage of development. Evaluate the growth mechanism of agricultural production enterprises from the angle of resource dynamic provide. In the e-commerce environment, the enterprise data and economic information are relatively concentrated, so the economical accounting system can instantly grasp the current activities of the economical data, and quickly generate economical information.

Multi-adaptive Feature Extraction Method of Marine Plant Information in Internet of Things

Xing, J.; Wang, X., and Dong, J., 2019. Multi-adaptive feature extraction method of marine plant information in Internet of Things. In: Guido-Aldana, P.A. and Mulahasan, S. (eds.), Advances in Water Resources and Exploration. Journal of Coastal Research, Special Issue No. 93, pp. 495–500. Coconut Creek (Florida), ISSN 0749-0208.In ocean monitoring and development, underwater image processing and recognition has gradually become a hot topic in ocean research. Because of the complex shooting environment, the underwater image is often seriously distorted. The absorption of water will lead to the attenuation of light in the transmission process, and the scattering of water will make the whole image appear atomization effect. According to the characteristics of underwater light propagation and underwater imaging system, the characteristics of underwater noise are analyzed. Butterworth high pass filter is used to smooth the backscatter noise of low frequency underwater image. On the basis of the above, SIFT feature points are extracted from the improved gray image, and K-nearest neighbor algorithm is used to match the feature points extracted from the standard image data set to obtain the information features of marine plants. The experimental results show that this method can obtain more feature points of marine plant information, and at the same time guarantee the accuracy of feature extraction time.

Extracting accent information from Urdu speech for forensic speaker recognition

This paper presents a new method for extraction of accent information from Urdu speech signals. Accent is used in speaker recognition system especially in forensic cases and plays a vital role in discriminating people of different groups, communities and origins due to their different speaking styles. The proposed method is based on Gaussian mixture model-universal background model (GMM-UBM), mel-frequency cepstral coefficients (MFCC), and a data augmentation (DA) process. The DA process appends features to base MFCC features and improves the accent extraction and forensic speaker recognition performances of GMM-UBM. Experiments are performed on an Urdu forensic speaker corpus. The experimental results show that the proposed method improves the equal error rate and the accuracy of GMM-UBM by 2.5 % and 3.7 %, respectively.

A Reservoir Information Extraction Method Based on Improved Hilbert-Huang Transform

Hilbert-Huang transform (HHT) is a nonlinear non-stationary signal processing technique, which is more effective than traditional time-frequency analysis methods in complex seismic signal processing. However, this method has problems such as modal aliasing and end effect. The problem causes the accuracy of signal processing to drop. Therefore, this paper introduces the method of combining the Ensemble Empirical Mode Decomposition (EEMD) and the Normalized Hilbert transform (NHT) to extract the instantaneous properties. The specific process is as follows: First, the EEMD method is used to decompose the seismic signal to a series of Intrinsic Mode Functions (IMF), and then The IMFs is screened by using the relevant properties, and finally the NHT is performed on the IMF to obtain the instantaneous properties.

Modeling of the"thin" structure probability discrepancy distribution in the indications of ADS-B and SSR. Uncompensated latency

The upcoming transition to a new generation of the elements in the ATC system - AZN-B - has set a number of tasks related to confirming the integrity of the data received, and determining their accuracy characteristics. Radar surveillance systems which are used up to now have established a good reputation and the matter of succession is naturally associated with the comparison of these two recording systems data. Discrepancies (differences) in the AZN-B registrations and radar data are an additional source of information related to reliability and accuracy of their measurements. The problem lies in the method of information extraction. The article analyzes the stable features of discrepancy distribution (histograms) probability obtained in the "Air Navigation Research and Development Institute" branch of GosNII GA by analyzing the observations of aircraft operations in Moscow region. The presence of asymmetry in these histograms is supposed to be interpreted as the presence of uncompensated latency in the transmission of data from AZN-B, which is typical for a certain portion of aircraft in the airspace of Russia. The emphasis on the "territorial affiliation" of the effect is due to the fact that we don’t have any information about the works abroad devoted to the phenomenon under discussion. In accordance with the hypothesis stated, a mathematical model of the measurement procedure with the following parameters to be estimated was constructed: the average value of the uncompensated delay and the expected share of aircraft, characterized by this defect. The values of the sought-for parameters were discovered according to the condition of minimum "distance" between the experimental histogram and the corresponding model analytical dependence. The estimates made revealed that about 30% of the aircraft, out of the total observed number, report their coordinates with an additional error of 75 m. at an average.

A Study of Medical Problem Extraction for Better Disease Management.

This study focuses on the extraction of medical problems mentioned in electric health records to support disease management. We experimented with a variety of information extraction methods based on rules, on knowledge bases, and on machine learning, and combined them in an ensemble method approach. A new dataset drawn from cancer patient medical records at the University of Utah Healthcare was manually annotated for all mentions of a selection of the most frequent medical problems in this institution. Our experimental results show that a medical knowledge base can improve shallow and deep learning-based sequence labeling methods. The voting ensemble method combining information extraction models outperformed individual models and yielded more precise extraction of medical problems. As an example of applications benefiting from acurate medical problems extraction, we compared document-level cancer type classifiers and demonstrated that using only medical concepts yielded more accurate classification than using all the words in a clinical note.

Research on landslide hazard information extraction method based on SAR image

Research on landslide hazard information extraction method based on SAR image

Novel Technique for Growth Plate Analysis Based on the Superposition of T1- and T2-weighted MR Imaging of Adolescent Wrists.

In this article, a new method of information extraction on the basis of the differentiation of T1- and T2-weighted MR images is proposed. It relies on a technique of superposition of T1- and T2-weighted MR images with use of statistical dominance algorithm. On the basis of implemented image analysis, a reproducible extraction of growth zone of adolescent boys’ wrists is possible.

Assessing the Availability of Data on Social and Behavioral Determinants in Structured and Unstructured Electronic Health Records: A Retrospective Analysis of a Multilevel Health Care System.

BackgroundMost US health care providers have adopted electronic health records (EHRs) that facilitate the uniform collection of clinical information. However, standardized data formats to capture social and behavioral determinants of health (SBDH) in structured EHR fields are still evolving and not adopted widely. Consequently, at the point of care, SBDH data are often documented within unstructured EHR fields that require time-consuming and subjective methods to retrieve. Meanwhile, collecting SBDH data using traditional surveys on a large sample of patients is infeasible for health care providers attempting to rapidly incorporate SBDH data in their population health management efforts. A potential approach to facilitate targeted SBDH data collection is applying information extraction methods to EHR data to prescreen the population for identification of immediate social needs.ObjectiveOur aim was to examine the availability and characteristics of SBDH data captured in the EHR of a multilevel academic health care system that provides both inpatient and outpatient care to patients with varying SBDH across Maryland.MethodsWe measured the availability of selected patient-level SBDH in both structured and unstructured EHR data. We assessed various SBDH including demographics, preferred language, alcohol use, smoking status, social connection and/or isolation, housing issues, financial resource strains, and availability of a home address. EHR’s structured data were represented by information collected between January 2003 and June 2018 from 5,401,324 patients. EHR’s unstructured data represented information captured for 1,188,202 patients between July 2016 and May 2018 (a shorter time frame because of limited availability of consistent unstructured data). We used text-mining techniques to extract a subset of SBDH factors from EHR’s unstructured data.ResultsWe identified a valid address or zip code for 5.2 million (95.00%) of approximately 5.4 million patients. Ethnicity was captured for 2.7 million (50.00%), whereas race was documented for 4.9 million (90.00%) and a preferred language for 2.7 million (49.00%) patients. Information regarding alcohol use and smoking status was coded for 490,348 (9.08%) and 1,728,749 (32.01%) patients, respectively. Using the International Classification of Diseases–10th Revision diagnoses codes, we identified 35,171 (0.65%) patients with information related to social connection/isolation, 10,433 (0.19%) patients with housing issues, and 3543 (0.07%) patients with income/financial resource strain. Of approximately 1.2 million unique patients with unstructured data, 30,893 (2.60%) had at least one clinical note containing phrases referring to social connection/isolation, 35,646 (3.00%) included housing issues, and 11,882 (1.00%) had mentions of financial resource strain.ConclusionsApart from demographics, SBDH data are not regularly collected for patients. Health care providers should assess the availability and characteristics of SBDH data in EHRs. Evaluating the quality of SBDH data can potentially enable health care providers to modify underlying workflows to improve the documentation, collection, and extraction of SBDH data from EHRs.

The threshold determination methods of water body information extraction using GF-1 satellite image

The GF-1 satellite is the first satellite of China High-resolution Earth Observation System (referred to short as CHEOS) with the advantages of high spatial resolution, high temporal resolution and 4 multi-spectral channels, which can quickly extract inland water body information. At present, the methods of water body information extraction by remote sensing mainly include single-band threshold method, spectral relation method, normalized difference vegetation index (NDVI) method, normalized difference water body index (NDWI) method and shade water body index (SWI) method. The results of water body information extraction will vary with different methods; especially, the thresholds determination of each method is an important factor in the accuracy of water body information extraction. In this paper, the water body information of Qinghai Lake, Hongze Lake, and Hulun Lake were extracted by NDWI index, and the optimal thresholds of water body information extraction were determined by using Jeffries-Matusita (J-M) distance method and density slice method was in the range of -0.061~0.093. Finally, the accuracy of the extraction results was analyzed, and the overall accuracy is more than 88%.

A multi-parameter constrained potential underdamped stochastic resonance method and its application for weak fault diagnosis

Vibration information is widely used in the fault diagnosis of rotating machinery, but the weak components reflecting early faults are often difficult to extract because of strong background noise. Among the variety of weak information extraction methods, the stochastic resonance (SR) method has the unique advantage of transferring noise energy to weak signals and has great application prospect in weak signal extraction. The current underdamped SR method with two adjustment parameters cannot achieve an optimal potential model structure, so it is easy to fall prey to the local optimization problem without having a better feature extraction method. Therefore, in this paper, we present a multi-parameter constrained potential underdamped stochastic resonance method by investigating the potential model and signal-to-noise ratio of the system. Moreover, the weighted ant colony algorithm is introduced to optimize potential parameters and a damping factor is used to improve the ability of weak fault information extraction. Furthermore, several types of bearing failure experiments were implemented to verify the effectiveness of the proposed method. The results show that the proposed method has excellent weak signal extraction ability and is suitable for early fault diagnosis.

Incremental Updating Information Extraction and Topology Conflict Detection Method for Updating Road Network

Abstract. In recent years, Located Based Service (LBS) has become one of the hotspots on the application of geographic information to the government and the public. Providing a good performance map to the users is an important part of the location service system. Compared with the traditional web spatial information service system, service system of map information also put forward higher requirements of real time and veracity. The problem of updating electronic map in real-time cannot be effectively solved, hindering the updating of multi-scale electronic map and restricting the development of location service market. As a very important branch for the updating data field of geographic information science, incremental update has become a hot issue in the current international GIS research field. The extraction of incremental information includes two key technologies, namely road network matching and road network selection. After incremental updating roads were inserted into the database, coordinating topological conflicts caused by these inserted roads was the mainly task. However, the accuracy and algorithm efficiency of each technology need further improvement to provide the users with the latest and the most accurate basic geographic data. As the "skeleton" element of a map, the road network rapidly changes, carrying out more urgent requires for the incremental road update. Aiming at key technology problems of the updated road network, the paper focused on the following three parts:(1) To address the problems that the traditional probabilistic relaxation method only adopted geometric constraints as one of road matching criterions and could not respond to M: N matching pattern, an improved probabilistic relaxation method was proposed from the combined views of local optimization and global one, integrating geometric indicators with topology ones to achieve an effect with local optimization, as well as identifying M: N matching pattern by inserting virtual nodes to achieve a globally optimal effect. Then the matching strategies and corresponding implement algorisms were designed for different matching patterns. The case test showed that the overall matching accuracy of each evaluation indictor reached over 90%, increasing by 6%–12%; the evaluation indicators on both spatial and attribute properties increased by 4%–6%; the proper buffer threshold could be defined as twice the average value of the closest distances from all nodes in the candidate matching dataset.(2) Aiming at the low accuracy and irrational structural selection caused by only using linear pattern or areal pattern, an improved method was proposed combining linear method as well as areal ones. The proposed method improves the Stroke generation algorithm of linear pattern using OLS model taking the overall information from the roads to be connected. Meanwhile, it partitioned road network by weighted Voronoi diagrams. The roads were selected under the process of calculating the importance of Stroke and its road density threshold. The test demonstrated that the improved method added the accuracy of Stroke generation results, and the results of road selection could maintain the overall characteristics of the road network as well as its spatial distribution information. Compared with method only using one pattern, the combined method had a better accuracy of road selection.(3) According to the characteristics of detection topology for an updated road network, this paper proposed a topology conflict detection algorithm considering the incremental update of multi-scale network. The algorithm designed K-level topological neighbourhood to identify incremental neighbourhood road segments, built a topological refinement model based on geometric metrics, proposed checking rules based on comprehensive operation operator, and detected the conflict topology by using improved topological distance. The experimental results showed that 1) the accuracy and recall of the proposed method were more than 90%; 2) considering conflict topology caused by the generalization, the accuracy increased by 29.2%; 3) the average path length of a road network could be used as the reference of the best K-value in the K order incremental neighbourhood method.

Building Detection From Satellite Images for Urban Planning Using MATLAB-Based Pattern Matching Method

Remote sensing and satellite image processing has many applications and an emerging field of trending in recent research. Remote sensing methods for extraction of information are used to obtain information related to the Earth's resources and environment for most government-based research applications. Few enhanced hybrid algorithms are reported and in use for real-world applications in recent years by extensive research for digital image processing. The digital images and satellite images are now extensively used for its information, which have many advantages over present analog image processing methods. The satellite data are larger and carry enormous amounts of information for applications like land description, crop assessment, cultivation monitoring, meteorology, regional description, underwater analysis, sea water interpretation, and so on. This paper analyses the requirements of better interpretation methods of received/captured images and newer processing techniques required. This paper presents complete study and pattern matching research in image processing tools using MATLAB programming and its capabilities that could be used in remote sensing applications in the future.

Spatial pattern analysis of Haloxylon ammodendron using UAV imagery - A case study in the Gurbantunggut Desert

Spatial patterns are not only the foundation for the understanding of plant interactions, but also reflect the spatial processes among plant populations. The primary requirement of spatial pattern analysis is the collections of location information of individual plants. In this study, we used low-altitude Unmanned Aerial Vehicle (UAV) remote sensing technology to obtain regional high-precision remote sensing images for Haloxylon ammodendron (H. ammodendron) forest in the southwestern Gurbantunggut Desert, and extracted spatial position information to analyze the spatial patterns using Ripley's L(r) function. Applying and comparing seven spatial position extraction methods, this study showed that the index RGRI (Red-Green Ratio Index) made 83.46% accuracy in the spatial position extraction, and an accuracy of 79.48% was obtained using NGBDI (Normalized Green-Blue Difference Index), while other five location information extraction methods resulted relatively lower accuracy. Results from spatial pattern analysis indicated that the extraction by UAV remote sensing were consistent with those obtained by field measurements. The H. ammodendron population showed a random distribution within the scale of 0–15 m, which suggested that the dependence of mutual asylum between individuals was low and not important. This distribution may be caused by the intense competition of individual vegetation for soil moisture, nutrients and other resources in desert areas. This study with low-altitude UAV imagery index analysis provided an efficient approach to rapid monitoring of plant population distribution characteristics in desert areas.

A multimedia case-based reasoning framework for assembly sequence planning

PurposeAssembly sequence planning (ASP) is a crucial job during assembly process design. However, it is still difficult to reuse the existing solution to solve a new ASP problem. In particular, with the rapid development of digital technologies, the reusable assembly information of an existing solution is not concentrated in one multimedia but dispersed in multiple heterogeneous multimedia, e.g. text, three-dimensional graphics, even images and videos. This paper aims to propose a multimedia case (MC)-based reasoning framework to solve ASP by reusing the existing solution whose assembly information is dispersed in multimedia.Design/methodology/approachThe proposed framework is designed with the introduction of the MC. An MC seamlessly integrates the dispersed assembly information of an existing solution. Under the proposed concept and architecture, the assembly information of an existing solution is extracted to build assembly descriptors of multimedia. Therefore, the MC is captured by organizing the assembly descriptors of corresponding multimedia.FindingsBy means of the framework proposed, it is possible to reuse the existing solution whose assembly information is dispersed in multimedia to solve ASP. Moreover, the extraction method of assembly information can flexibly parse most of the multimedia. Finally, the MC has the capability to represent the existing solution by collecting dispersed assembly information.Originality/valueThe proposed framework can discover the similar existing solution and avoid the potential failures confronted in the past so that the feasibility of ASP result can be improved as much as possible.