25 June 2013, Volume 29 Issue 6

    

• Select all
  |

• Select
  Reviews of the Open Data Metric Studies:An Alternative Metric (Altmetrics) for Calculating the Online User Behavior and the Scientific Community Impact

  Ku Liping

  New Technology of Library and Information Service. 2013, (6): 1-8. https://doi.org/10.11925/infotech.1003-3513.2013.06.01

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  This paper introduces what the advantages of Altmetrics that by using the social network usage data to recommend the information retrieval ranking order and with the other impact index to change the scholar evaluation approach. By using the case observation analysis, the author draws-up the open data framework of the Altmetrics.com, focuses on its specialized service for the Article-Level Metrics. For discussion on the library new knowledge service issue in the open repository, open publishing supporting and the new research-group-embed information service, the two core articles which can improve the empirical validation method are simplified as a brief operational workflow.
• Select
  A Knowledge Representation Method for Pharmaceutical Products in China

  Chen Ying, Li Jiao, Li Junlian

  New Technology of Library and Information Service. 2013, (6): 9-15. https://doi.org/10.11925/infotech.1003-3513.2013.06.02

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  To represent the semantics of pharmaceutical products and facilitate drug information integration, this paper develops a knowledge representation method for pharmaceutical products based on the frame work of concept, relationship and attribute. In evaluation, this method is applied to represent the drug products’ records which are initially organized in a relational database. In the case study, it shows that the method has the advances in pharmaceutical concept standardization and semantic relationship representation. At last, the perspectives of this method are discussed including its applications in health informatization and drug information dissemination online.
• Select
  Overview on Construction of Ontology in Social Tagging System

  He Jinjing, Dou Yongxiang

  New Technology of Library and Information Service. 2013, (6): 16-22. https://doi.org/10.11925/infotech.1003-3513.2013.06.03

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  The research background and related research work about the Ontology in social tagging system are given in this paper. And then two categories are provided: the Ontology based on the concept of “tagging behavior” and the Ontology based on the concept of “tag”.It also shows the advantages and existing problems of Ontology based on the social tagging system in improving semantic organization. Finally, based on the characteristic of the Ontology, the improvement way is explored in this paper.
• Select
  Study on the Model of Automatic Extraction and Annotation of Trail Cases

  She Guiqing, Zhang Yongan

  New Technology of Library and Information Service. 2013, (6): 23-29. https://doi.org/10.11925/infotech.1003-3513.2013.06.04

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  This paper constructs an Ontology-based automatic extraction and annotation model for the massive texts of criminal judgments combined with the case-Ontology. It uses regular expressions to construct extraction rules and templates for the semi-structured characteristics of the texts of legal cases, according to the structure of the documents and the clue words. Besides, it applies natural language processing techniques for the accurate information extraction, then gives semantic annotation of the results of extraction for building an Ontology knowledge base of legal cases, to realize the transformation of case texts to semantic information Web, for the further similar case retrieval and judge recommendation. And the experiment shows a good result.
• Select
  Comparative Analysis of Centrality Indices in Extracting Concepts from Semantic Predication Network——Based on Disease Treatment Research

  Zhang Han, Liu Shuangmei

  New Technology of Library and Information Service. 2013, (6): 30-35. https://doi.org/10.11925/infotech.1003-3513.2013.06.05

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  The aim of the study is to compare the validity of four node centrality indices in extracting crucial nodes from semantic predication network. Depending on Unified Medical Language System (UMLS) and SemRep, this paper first constructs a semantic predication network for biomedical literature, in which nodes represent UMLS concepts and edges semantic relations between nodes. Relying on the semantic type of the concepts and the semantic relations, schemas related to disease treatment are defined and used to extract disease treatment related predications. Then four centrality indices including degree centrality, betweenness centrality, closeness centrality and eigenvector centrality are used to extract crucial concepts related to four aspects of disease treatment (therapeutic drugs, therapeutic procedures, body location of the disease and disease comorbidities). The extracted concepts are compared to a reference standard produced by domain experts. The results show that centrality combined with semantic schema can effectively extract crucial nodes of the users interest. Among four centrality indices, degree centrality performs best (F-score is 0.72) and eigenvector centrality performs secondly best (F-score is 0.66).
• Select
  Sentence Alignment and Re-Alignment for Environmental Protection Texts in English-Chinese Parallel Corpus

  Xiong Wenxin

  New Technology of Library and Information Service. 2013, (6): 36-41. https://doi.org/10.11925/infotech.1003-3513.2013.06.06

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  Sentence alignment is a crucial step for building parallel corpus. There are plenty of such tools available for constructing a language repository for machine translation systems. Based on the evaluation regarding user-friendly design and alignment quality, the performance of Champollion is superior to other mainstream open source tools in aligning English-Chinese parallel texts. Inspired by “transformation-based error-driven” strategy, the author makes a thorough linguistic analysis on the error output produced by Champollion, and proposes an error correction strategy which improves the precision rate dramatically. The realignment approach as a module attached to Champollion’s output can reach a precision rate 93.91% from baseline 88.74%, in the case of alignment of English-Chinese texts in the area of environmental protection. This alignment and realignment strategy combined statistics-based method with linguistic insights can be applied to other domains.
• Select
  A New Method of Keywords Extraction for Chinese Short-text Classification

  Hu Yongjun, Jiang Jiaxin, Chang Huiyou

  New Technology of Library and Information Service. 2013, (6): 42-48. https://doi.org/10.11925/infotech.1003-3513.2013.06.07

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  Short texts are different from traditional documents in their shortness and sparseness. Feature extension can ease the problem of high sparse in the vector space model, but feature extension inevitably introduces noise. To resolve the problem, this paper proposes a high-frequency words expansion method based on LDA. By extracting high-frequency words from each category as the feature space, using LDA to derive latent topics from the corpus, it extends the topic words into the short-text. Extensive experiments conducted on Chinese short messages and news titles show that the new method proposed for Chinese short-text classification can obtain a higher classification performance comparing with the conventional classification methods.
• Select
  Sentiment Analysis of Product Reviews by means of Cross-domain Transfer Learning

  Zhang Zhiwu

  New Technology of Library and Information Service. 2013, (6): 49-54. https://doi.org/10.11925/infotech.1003-3513.2013.06.08

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  Aiming at the problem of sentiment analysis of incomplete product reviews data, this paper proposes a cross-domain sentiment analysis method based on spectral clustering and transfer learning. With the help of domain-independent words as a bridge, using spectral clustering algorithm to align domain-specific words from different domains into unified clusters, it can reduce the gap between domain-specific words of the two domains, and can improve the accuracy of sentiment classifiers in the target domain. Experiments studies are carried out to show the efficiency and superiority of the proposed approach in solving the problem of cross-domain sentiment analysis of product reviews.
• Select
  The Identification and Analysis of Micro-blogging Opinion Leaders in the Network of Retweet Relationship

  Xiong Tao, He Yue

  New Technology of Library and Information Service. 2013, (6): 55-62. https://doi.org/10.11925/infotech.1003-3513.2013.06.09

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  This paper builds the adjacent matrix based on the retweet relationship in micro-blogging to find the opinion leaders through improved HITS algorithm. Then a net of opinion leaders is built based on the relationship of retweet to prove the efficiency of the improved algorithm and to analyze the function of opinion leaders. The research shows that the improved HITS algorithm can find the opinion leaders effectively. The hub of an opinion leader is highly positive correlated with the amount of his fans. According to the analysis of the net of opinion leader, the authors find that the opinion leaders play important roles in the key nodes,and their function is not weakened by the increasing of information sources in the micro-blogging.
• Select
  Object Recognition of Network Comments Based on Conditional Random Fields

  Lin Chen, Wang Lancheng

  New Technology of Library and Information Service. 2013, (6): 63-67. https://doi.org/10.11925/infotech.1003-3513.2013.06.10

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  Combined with the characteristic of comment object, this paper gives an identification method based on conditional random fields. Without domain knowledge, the new method introduces characteristics word and clues word, then transforms comment object recognition problem into solving maximum probability sequence. The experimental results show that this method can completely, effectively extract comment objects from network comments.
• Select
  Extracting Information Method Term from Chinese Academic Literature

  Hua Bolin

  New Technology of Library and Information Service. 2013, (6): 68-75. https://doi.org/10.11925/infotech.1003-3513.2013.06.11

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  This paper identifies sentences on method from academic literature using rules, then extracts method terminology from these sentences using lexicon and rules, among which synonymous terminology is merged. The author makes an experiment to extract method knowledge from full text of papers published on Journal of the China Society for Scientific and Technical Information, then builds a set of information method system by a statistical analysis on experiment result, which testifies the method is effective.
• Select
  Analyzing the Demand of Online Product Review System’ s Features Using Kano Model: An Empirical Study of Chinese Online Shops

  Sun Xiaoling, Zhao Yuxiang, Zhu Qinghua

  New Technology of Library and Information Service. 2013, (6): 76-84. https://doi.org/10.11925/infotech.1003-3513.2013.06.12

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  Based on the impact of electronic word-of-mouth and critical issues of online information system design, this paper conducts a feature package of online product review system by a survey of mainstream Chinese online shops. Kano model then employs to classify customer demand about these features. The result indicates that among various features most of them are dispensable except of features which making deep mining on argument quality and valence such as tag clouds based on text mining as well as multiple valences. It can be important reference when dealing with system design and improvement.
• Select
  The Application of Aliyun Search Cloud Service to Build Search Engine for Library Sites

  Wang Shuang, Chen Junjie, Xiao Zheng, Huang Guofan

  New Technology of Library and Information Service. 2013, (6): 85-89. https://doi.org/10.11925/infotech.1003-3513.2013.06.13

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  The application of cloud search service becomes a new search technical direction. Xiamen University Library chooses Aliyun service to rebuild site search, and packages the cloud search as an independent search engine. After pre-processing, the site data is submitted and index files are generated, then the search strings are passed to the cloud search engine. Base on the cloud search service, searching finished and results displayed. Evaluation results show that Aliyun cloud search, compared with the old search engine, is significantly improved in search efficiency and functionality,etc.
• Select
  Application of WebGIS in Collection Spatial Information Visualization

  Bao Jie, Zhu Shiping

  New Technology of Library and Information Service. 2013, (6): 90-95. https://doi.org/10.11925/infotech.1003-3513.2013.06.14

  Abstract ( ) Download PDF ( ) HTML ( )   Knowledge map   Save

  Aiming at the present situation of university library document obtained difficultly, this paper realizes the visualization of collection spatial information based on WebGIS technology application in the library. It introduces the design idea, functional partitioning, choice of open source software, development platform and so on, and expounds the realization process of the key technologies of spatial data expression way, map service and map search.