| 滕广青.Folksonomy模式中紧密型领域知识群落动态演化研究[J].中国图书馆学报,2016,42(4):51~63 |
| Folksonomy模式中紧密型领域知识群落动态演化研究 |
| Dynamic Evolution of the Close-knit Domain Knowledge Communities in Folksonomy |
| 投稿时间：2016-03-14 修订日期：2016-04-18 |
| DOI：10.13530/j.cnki.jlis.164004 |
| 中文关键词: 知识组织 Folksonomy 领域知识 知识群落 复杂网络 |
| 英文关键词: Knowledge organization Folksonomy Domain knowledge Knowledge community Complex network |
| 基金项目:本文系国家自然科学基金项目“基于网络结构演化的Folksonomy模式中社群知识组织与知识涌现研究”(编号:71473035)和教育部人文社会科学研究规划基金项目“基于后结构主义网络分析的Folksonomy模式中社群知识非线性自组织研究”(编号:14YJA870010)的研究成果之一 |
|
| 摘要点击次数: 2906 |
| 全文下载次数: 1155 |
| 中文摘要: |
| 具有高度感知易用性的Folksonomy知识组织模式,在实践应用中却呈现离散纷杂的外在表象。对Folksonomy知识组织模式中领域知识群落及其结构关系的研究,对于洞悉领域知识发展过程中的衍生、交叉、融合等现象具有重要意义。本研究以复杂网络分析中的层次派系分析技术为主要研究方法,基于标签共同标注关系构建领域知识网络。从标签间关联关系出发,对特定领域知识网络中紧密型知识群落的发展过程进行时间序列的动态跟踪与分析。研究结果表明:标签数量在领域成熟期趋于饱和,但标签间的连线数量却持续增长,紧密型领域知识群落规模逐渐扩大；紧密型领域知识群落数量总体递增的同时也存在波动,这种波动与知识群落自身的扩张、衰减、派生、融合的演化过程有关；随着领域知识的发展,紧密型知识群落之间的交叠密度呈上升趋势,并基于交叠关系形成了一个更大的知识群落,成为领域知识网络的部分骨架结构,进而能够展示出多个主要发展方向。对领域知识群落演化规律的揭示,有助于把握领域知识演进的发展脉络,并揭示Folksonomy知识组织模式中领域知识的发展模式与规律。图4。表4。参考文献32。 |
| 英文摘要: |
For its high level of perceived ease of use，Folksonomy presents a discrete and perplexing appearance in its practical application. Studies about domain knowledge communities and their structural relationships within Folksonomy （knowledge organization mode） make up an important research topic in library and information academia and are crucial for a thorough understanding of numerous phenomena，such as deviation，crossover and fusion，in the development of domain knowledge.
This study makes an extraction of 2231 related articles in specific knowledge domains with 647 tags and the inter-tag-relationships number is 11749 in total. The tagging-time span is from 2006 to 2015 and the timer shaft is divided into 10 units of time-window. The study adopts as the major research approach the clique analysis technique which is commonly used for complex network analysis，and domain knowledge networks are constructed via co-tagging relations of the tags. The cliques at level are constructed via dynamic thresholds based on the association relationships between tags and the close-knit domain knowledge communities are identified by the frequency and completeness of the association relationships. The dynamic tracking and analysis of the close-knit knowledge communities in specific domain knowledge are executed with the time series as clues. Finally，the skeleton structure of domain knowledge is extracted in perspective of overlapping cliques.
The present study concludes that，in the development of domain knowledge based on Folksonomy，tag population in domain mature stage tends to become saturated while the association relationships between the tags （lines） continue to yield a sustainable increase and the scale of close-knit domain knowledge communities presents a tendency of fluctuating expansion. Two factors account for this fluctuating expansion. Firstly，there are remarkable dynamic changes in the development of domain knowledge；some significant relationships between tags in the initial developmental stage become no longer significant in the follow up stage due to the rapid development of other relationships，and some insignificant relationships in the initial stage become increasingly significant in the follow up period due to the rapid development of their own. Secondly，expansion，attenuation，derivation and fusion of domain knowledge communities co-exist in the domain knowledge development process，and this multi-mode-coexisting evolutionary process also leads to fluctuations in the formation of close-knit domain knowledge communities. With the development of domain knowledge，overlap density among close-knit domain knowledge communities is on the rise and gradually forms a larger knowledge community based on overlapping relationships which become part of the skeleton structure of domain knowledge network and comes to display a number of major multi-dimensional development directions. The idea and approach of clique at level and overlapping cliques employed in this study expand the application of complex network analysis technique in the field of library and information science，promote the crossing of disciplinary knowledge，and provide a new perspective for the study of knowledge organization.
This study proposes close-knit domain knowledge communities （CDKC） and is helpful for the grasp of the core content and factors of domain knowledge from complicated tag sets. The tracking and analysis of close-knit domain knowledge communities in Folksonomy via time series are dynamic in nature and capable of further development，and the transformation of static analyses into dynamic ones helps to penetratingly reveal the mode and pattern of domain knowledge development. It is virtually impossible for the present study to exhaust the previous literature and studies in this area，but the revelation of the evolutionary patterns and rules concerning domain knowledge contribute to the grasp of the developing venation of domain knowledge evolution，and also facilitate the studies aiming to reveal the development patterns and rules of domain knowledge in Folksonomy. 4 figs. 4 tabs. 32 refs.
|
|
查看全文
查看/发表评论 下载PDF阅读器 |
