Search

Copyright of Journal of Xinjiang University (Natural Science Edition) is the property of Xinjiang University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Acta Agriculturae Zhejiangensis is the property of Acta Agriculturae Zhejiangensis Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Objective: Deep convolutional neural network (DCNN) is widely used in automatic road extraction from high-resolution remote sensing images (HRSIs). However, the existing methods make it difficult to model the context relationship of pixels in the predicted results. To solve this problem, some studies have used fully connected conditional random field (FullCRF) to perform secondary optimization of semantic segmentation results combined with context information, but the discontinuity problem of road structure cannot be effectively improved. Methods: In order to improve the integrity of road structure, this paper proposes a short range conditional random filed (SRCRF) model combined with DCNN. SRCRF mainly includes a unary potential function based on road pre-segmentation, a binary potential function based on spectral spatial features, and a K-neighborhood mean field inference algorithm. First, the prior knowledge of road pre-segmentation results is obtained by using the powerful feature extraction capability of DCNN as the unary potential function of SRCRF. Then, the dependence of the binary potential function defined by the linear combination of Gaussian kernel functions on the surrounding nodes is modeled. The binary potential function enables the classification results to have local consistency,that is, adjacent pixels with similar spectral features have the same label. Finally, the K-neighborhood mean field inference algorithm based on the mean-field approximation inference algorithm optimizes the inference range to make full use of the spatial context information and spectral feature context information of the road, and then calculates the optimal label corresponding to each pixel based on the space and spectral feature to optimize the road accurately. The convolution method is adopted to control the inference range of SRCRF within the radius of K in order to improve the proportion of feature vectors. Results: The experimental results show that SRCRF not only alleviates the transition smoothness of FullCRF, but also alleviates the structural discontinuity and incompleteness in the road acquisition results of HRSIs. In Zimbawe-Roads dataset and Cheng-Roads dataset, F1-scores of SRCRF are increased by about 4.01% and 3.73% respectively compared with DCNN, and about 3.25% and 2.28% respectively compared with FullCRF. Conclusions: The proposed SRCRF combines the advantages of DCNN, optimizes the fully connected structure of traditional conditional random fields into the K-neighborhood structure, reduces the inference scope, and improves the proportion of feature vectors. Compared with FullCRF, SRCRF can make better use of image color features and spatial features to accurately optimize the road extraction results of deep learning output. The performance of SRCRF is improved compared with DCNN and FullCRF, and the time is shortened by one order of magnitude compared with FullCRF. In the future, we will further investigate the potential for learning Gaussian features and investigate more complex architecture of conditional random field to better capture global context information. Additionally, we are particularly interested in exploring the application potential of SRCRF in other fields, such as building extraction, vehicle extraction, lake extraction, etc. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Named entity recognition is a key step in constructing a knowledge graph in the clock domain. However, the current clock domain has problems such as the small number of labeled samples, which leads to the low accuracy of named entity recognition in the clock domain. To this end, this paper uses the pre-trained language model BERT to extract the features of the text in the clock domain, and then uses the linear chain conditional random field (Linear-CRF) method for sequence labeling, and proposes a BERT-LCRF named entity recognition model. The results of comparative experiments show that the model can fully learn the feature information of the clock domain, improve the accuracy of sequence labeling, and then improve the effect of named entity recognition in the clock domain. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Objectives: In the field of human-related geographic information systems (GIS), the spatiotemporal analysis of character information has received increasingly more attention. It is important in that it helps GIS users to generate various thematic maps and achieve the visualization of human geographic con⁃ tent. For adaptation to the development direction of GIS intellectualization, it is of great significance to combine GIS requirements with natural language processing (NLP) methods and build a character information model. Methods: Firstly, we expound the research status of character information models in GIS and NLP and put forward the concept of life-track, which is mainly composed of a series of character event mentions. Secondly, considering the feasibility of the existing information extraction methods, a conceptual character life-track information model is determined. This model focuses on event information to highlight character spatiotemporal elements and also includes character attribute and relationship information. Finally, a complete information extraction process is designed for the model with online character encyclopedia pages as the data source. This paper focuses on two sub-tasks in the process: One is to use time features and Open How Net semantic calculations to identify event mentions, and the other is to use linguistics features and the conditional random field (CRF) model to extract location information. Results: Experiment results show that the method of event mention identification has an accuracy of 91.8%. Although the average F1 value of location information extraction is only 78% under the condition of a limited labeling corpus, some valuable experimental conclusions have been obtained by analyzing the weight of the transmit matrix of the CRF model: (1) The location phrase and its adjacent words have obvious characteristic effects. (2) The dependency syntactic parsing and the relative position of the word in the sentence have little influence on the extraction. (3) The target of location information extraction is the place where the event occurred, but in a few cases, some location phrases are irrelevant to the location of the event. This is the main reason for the low accuracy. Conclusions: Combining GIS with NLP, intelligent GIS development will be promising. The character life-track information model provides an example of the large-scale use of ubiquitous internet information. Improving methods applied in the extraction process and applying those methods to more online text types are the focus of our team's subsequent research. [ABSTRACT FROM AUTHOR]

Copyright of Telecommunication Engineering is the property of Telecommunication Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The paper proposed a high-resolution remote sensing image change detection framework, it combined adaptive space and conditional random field ( CRF) . The framework could reduce the noise caused by registration error and over-smoothing phenomenon caused by conditional random field in dual-phase image change detection. The proposed framework constructed a weighting adaptive space to eliminate the noise caused by the registration error. It used the fuzzy C-means clustering method based on morphological reconstruction to construct the CRF unary potential which effectively reduced speckle noise, and it constructed the conditional random field pairwise potential with spectral and spatial constrained model, which improved the anti-noise performance of the algorithm and could prevent over-smoothing. Compared with several existing change detection methods, the proposed method shows advantage in the overall accuracy,false rate and miss rate, and can keep boundary better in the same time. [ABSTRACT FROM AUTHOR]

Aiming at the problem of automatic identification of existing concepts and discovering new concepts in a specific field, this paper proposed a new words discovery method based on conditional random field ( CRF) and information entropy. This method used CRF to predict the boundary of conceptual words in text, selected the candidates of the new concept with the comparison to the existing concepts in the dictionary and found the probably location in text. Then it used the mutual information and the left and right entropy to judge the internal integration degree and the boundary freedom of the concept in the concept window for discovering new professional concepts. Experiments show that the concept discovery using the proposed method has a better effect than the method of using CRF alone. The accuracy of the concept discovery based on word and words model is respectively improved by 20. 06% and 46. 54% . [ABSTRACT FROM AUTHOR]

Copyright of Journal of Chongqing University of Posts & Telecommunications (Natural Science Edition) is the property of Chongqing University of Posts & Telecommunications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Currently the mainstream methods for Chinese word segmentation exploit statistical machine learning models. These methods usually require manual-annotated segmented sentences as training corpus, yet have neglected the annotated large-scale lexicon resources which have been built before, where these resources can be highly valuable when cross-domain evaluation is conducted, as the gold-standard sentence-level annotations arerare. Recently, the integration of lexicon formation into word segmentation models has gained increasing interest. As a whole, the integration methods can be classified into two categories:one being based on character-based models that cast word segmentation problem as sequence labeling, and the other being based on word-basedmodels that use beam-search to decode. This paper compared these two models, and combined them. Experimental results on benchmark datasetsshow that lexicon information can be more fully explored after combination, and finally the combined model can achieve better performances with both in-and cross-domain settings. [ABSTRACT FROM AUTHOR]

Copyright of Oil Drilling & Production Technology / Shiyou Zuancai Gongyi is the property of Shiyou Zuancai Gongyi Bianjibu and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In order to improve the effect of Tibetan name recognition,this paper designed the algorithm based on three levels of hierarchical features. It proposed a three-layer feature,which was based on the Tibetan character name without word- segmentation. The three-layer feature included internal features, the context information and the parallel relations feature. It used the conditional random fields (CRF)algorithm to identify the Tibetan name research. First,it considered the internal and context characteristics of the name as a CRF feature,and then considered the relationship between names as the rule to further improved the recognition effect. The recall was increased 10.43% and the F-value will reach 95. 02% . Experiment shows that the method achieves a very good effect for recognition of Tibetan names. [ABSTRACT FROM AUTHOR]

Chinese place name recognition is a research topic in named entity recognition, and a key to improve the application level of the geographic information systems in China. The traditional place name recognition method is based on the element characteristics of a place name and the part of speech of words, and employs limited features. This paper proposes a method of Chinese place name recognition method using syntactic features, and mines the syntactic characteristics of place names in natural language. The design employs four syntactic features, class, path, distance, and number, in conditional random fields (CRF) to train and recognize Chinese place names based on place name element s, position of speech (POS) and syntactic features. Comparative experiments with composite features and traditional features for Chinese place name show that with the help of the three composite feature,s Chinese place name recognition accuracy and recall rate can be improved effectively and with good results for complex place names. [ABSTRACT FROM AUTHOR]

With the rapid development of the World Wide Web，a huge quantity of geographic information resources are hidden as unstructured texts. Toponym recognition is the foundation of mining the potential geographic information from these texts. In traditional toponym recognition methods based on the natural language processing，the structure of Chinese toponym and features of user customs are ignored，which results in the low recall and precision. In this paper， linguistic know ledge is introduced to analyze Chinese toponym ，and the more specific morpheme categories are recognized. T hen the process of toponym recognition is transformed into an equivalent sequence labeling problem based on the conditional random tield. A proper labeling schema for Chinese toponym is also designed to improve the recognition accuracy. In the experiments， the 1.7 million tagged corpus of The People’s Daily are used to test the proposed method. The recall， precision and F value of the resuit are 92. 69% ，96. 73% and 94. 67% respectively，which are better than other machine learning models. It is proven that the proposed method is effective to recognize Chinese toponym . This research can provide more precise Toponym services for geographic information applications. [ABSTRACT FROM AUTHOR]

Traditional methods of natural language processing input a good deal of handmade features into statistics learning models to process the text. At present, conditional random field model performed well in many natural language processing tasks, but the way of handmaking features and huge number of features made it more difficult to build the model, reduced the speed of model computing, and generated overfitting phenomenon. To solve these problems mentioned above, this paper introduced a novel conditional random field model extended by tensor. It designed complicated features automatically, reduced the workloads of handmaking features, used the Tucker decomposition algorithm to speed the model, and could be applied in many natural language processing tasks. Experiments show that the model can perform better in many natural language processing tasks than traditional conditional random field model if using the same elementary features. [ABSTRACT FROM AUTHOR]

Web has become an important information source for most people. However, the information on the web is not always true and credible. How to help people quickly judge the credibility of information on the web was required. A new method based on content trust was proposed in this paper, which aim to evaluate the credibility of web information. Firstly, the method utilized the conditional random field model to get the theme of web information content, which can be used to a clue to collect the candidate evidence on the web. Secondly, the truthfulness of evidence can be evaluated by the characteristic of the concurrency and relevancy of the evidence and web information theme and other characteristic. Finally, the credibility value of web information was calculated. The experiment results show the method proposed in the credibility evaluation of web information content is reasonable and effective. [ABSTRACT FROM AUTHOR]

The task of aspect-level opinion mining usually include the extraction of product entities from consumer reviews, the identification of opinion words that are associated with the entities, and the determination of these opinion's polarities. Aiming at realizing aspect-level opinion mining for Chinese reviews, the paper proposes the four major steps: pre-processing; preparing the training set to learn the model; defining learning functions for conditional random field model; and applying the model to label new review data. At the same time, our experiments on the real Chinese reviews of five types of products show that the conditional random field based method can achieve 80% in most of performance indicators of extracted different types of review opinion elements. In order to verify the effectiveness of the proposed method, a test of the significance of difference is involved. Experiments report that there is scarcely difference of performance on conditional random field based method for both Chinese reviews and English reviews. Finally, we compare the precision of aspect extraction and the accuracy of sentiment classification based on three different methods, and the result shows that CRF-based method outperforms the other two such as lexicalized hidden markov model and association rule mining. [ABSTRACT FROM AUTHOR]

Tibetan word segmentation (TWS) is the process of combination the syllables sequence into words sequence, and which is the basic problem for Tibetan natural language processing. To solve the special problems in TWS, this paper reformulated the segmentation as a syllable tagging problem, and conducted comparative experiments at the same condition in different sequence label models. Experimental results show that the TWS system with conditional random fields achieves the best performance in thecondition of four-tag, other models achieve good resultsat the same time, all the above show that, the segmentation as a syllable tagging problem is a better approach to deal with TWS. [ABSTRACT FROM AUTHOR]

Named entity recognition is a fundamental technology in natural language processing(NLP). In recent years, rapid development of social network platforms such as microblog presents new challenges to the traditional named entity recognition(NER) technology because of the unique form. In this paper, an improved method based on the conditional random field(CRF) model is proposed for microblog texts. Due to the short texts and semantic ambiguity, external data resources are introduced to generate the topic feature and word representation feature for training the model. Due to the large-scale of microblog data and the high cost of manual standardization, an active learning algorithm based on least confidence is adopted to enhance the training effect at a lower cost of labor. Experiments on a Sina weibo data set show that this method improves the F-score by 4.54% compared to the traditional CRF methods.