Page 103 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2018 Vol. 42
P. 103

102 Journal of Library Science in China, Vol. 8, 2016


            extent, were the specialization, supplement and modification of CT knowledge framework.
            Consequently, knowledge systems in later thesauri are good succession of CT, as well as their
            macrostructure and description patterns of microstructure. It can be seen that CT is enormously
            large and universally applicable. It covers a wide range of disciplines in social sciences, natural
            sciences, engineering, and technologies, and it has the largest term collections. Thus, CT is capable
            of covering existing thesauri in magnitude and specifics of concept schemes.
              (2) CT has been maintained and utilized better
              From a holistic perspective, the Chinese thesauri have not been maintained properly and not truly
            used widely. Most thesauri have never been revised, and their knowledge systems are expected
            to be updated. However, it is very unlikely that the copyright owner and original developer can
            construct and maintain the thesauri. Therefore, these thesauri can be considered and used as
            supplements to CT, instead of providing services and being used independently. Comparatively,
            CT has been maintained properly. The first edition CT was published in 1980, which contained 109
            thousand terms in total. In 1991, a partially revised version of CT was published, which covered
            natural sciences, engineering and technologies, and contained 81 thousand terms. In 2014, a new
            edition of CT was published focusing on engineering and technology. In order to meet the demands
            of automatic information processing, this newest version has increased the amount of terms, to a
            total of 360 thousands (Institute of Scientific and Technical Information of China, 2014). In 2015,
            a revise project of natural science section of CT was initiated.
              For the reasons above, NTW adopts the integration mode of using CT as a mapping center,
            which other thesauri are mapped to it. As shown in Figure 1, the architecture of NTW has three
            layers: data acquisition and transformation layer, storage and semantic integration layer, and
            service and application layer.


                            %BUB "DRVJTJUJPO    4UPSBHF   4FNBOUJD *OUFHSBUJPO  4FSWJDF   "QQMJDBUJPO
                            5SBOTGPSNBUJPO

                             6OJGPSN %FTD   5PQ $MBTTJGJDBUJPO 5PQ 0OUPMPHZ  UIF 5IJSE 1BSUZ
                  4PVSDF "   SJQUJPO   'PSNBU                            *OWPLJOH 4FSWJDF
                             5SBOTGPSNBUJPO
                                                                         5IFTBVSVT $VTUP
                                             5IFTBVSVT                   NJ[BUJPO 4FSWJDF
                                               "          $IJOFTF
                             %BUB *NQPSUJOH              5IFTBVSVT       $PODFQU   $PODFQU
                  4PVSDF #     7FSJGJDBUJPO   5IFTBVSVT                    (SPVQ TFSWJDF
                                                 #                       5IFTBVSVT .FUBEBUB
                                                        5IFTBVSVT
                                                           $                 4FSWJDF
                  4PVSDF $     .FUBEBUB
                              3FHJTUSBUJPO                               4FBSDI #SPXTF &EJU
                                                                            %PXOMPBE
                                                  #BTJD -FYJDPO
                                         Figure 1. Architecture of NTW.
   98   99   100   101   102   103   104   105   106   107   108