Purpose:This paper introduces an analysis framework for tracking the evolution of research topics at the selected topics level,covering a research topic’s evolution trend,evolution path and its content changes over t...Purpose:This paper introduces an analysis framework for tracking the evolution of research topics at the selected topics level,covering a research topic’s evolution trend,evolution path and its content changes over time.Design/methodology/approach:After the topics were recovered by the author-topic model,we first built the keyword-topic co-occurrence network to track the dynamics of topic trends.Then a single-mode network was constructed with each node representing a topic and edge indicating the relationship between topics.It was used to illustrate the evolution path and content changes of research topics.A case study was conducted on the digital library research in China to verify the effectiveness of the analysis framework.Findings:The experimental results show that this analysis framework can be used to track evolution of research topics at a micro level and using social network analysis method can help understand research topics’evolution paths and content changes with the passage of time.Research limitations:Using the analysis framework will produce limited results when examining unstructured data such as social media data.In addition,the effectiveness of the framework introduced in this paper needs to be verified with more research topics in information science and in more scientific fields.Practical implications:This analysis framework can help scholars and researchers map research topics’evolution process and gain insights into how a field’s topics have evolved over time.Originality/value:Tbe analysis framework used in this study can help reveal more micro evolution details.The index to measure topic association strength defined in this paper reflects both similarity and dissimilarity between topics,which belps better understand research topics’evolution paths and content changes.展开更多
Induction of common knowledge or regularities from large-scale clinical data is a vital task for Chinese medicine(CM).In this paper,we propose a data mining method,called the Symptom-Herb-Diagnosis topic(SHDT) mod...Induction of common knowledge or regularities from large-scale clinical data is a vital task for Chinese medicine(CM).In this paper,we propose a data mining method,called the Symptom-Herb-Diagnosis topic(SHDT) model,to automatically extract the common relationships among symptoms,herb combinations and diagnoses from large-scale CM clinical data.The SHDT model is one of the multi-relational extensions of the latent topic model,which can acquire topic structure from discrete corpora(such as document collection) by capturing the semantic relations among words.We applied the SHDT model to discover the common CM diagnosis and treatment knowledge for type 2 diabetes mellitus(T2DM) using 3 238 inpatient cases.We obtained meaningful diagnosis and treatment topics(clusters) from the data,which clinically indicated some important medical groups corresponding to comorbidity diseases(e.g.,heart disease and diabetic kidney diseases in T2DM inpatients).The results show that manifestation sub-categories actually exist in T2DM patients that need specific,individualised CM therapies.Furthermore,the results demonstrate that this method is helpful for generating CM clinical guidelines for T2DM based on structured collected clinical data.展开更多
摘要Purpose:This paper introduces an analysis framework for tracking the evolution of research topics at the selected topics level,covering a research topic’s evolution trend,evolution path and its content changes over time.Design/methodology/approach:After the topics were recovered by the author-topic model,we first built the keyword-topic co-occurrence network to track the dynamics of topic trends.Then a single-mode network was constructed with each node representing a topic and edge indicating the relationship between topics.It was used to illustrate the evolution path and content changes of research topics.A case study was conducted on the digital library research in China to verify the effectiveness of the analysis framework.Findings:The experimental results show that this analysis framework can be used to track evolution of research topics at a micro level and using social network analysis method can help understand research topics’evolution paths and content changes with the passage of time.Research limitations:Using the analysis framework will produce limited results when examining unstructured data such as social media data.In addition,the effectiveness of the framework introduced in this paper needs to be verified with more research topics in information science and in more scientific fields.Practical implications:This analysis framework can help scholars and researchers map research topics’evolution process and gain insights into how a field’s topics have evolved over time.Originality/value:Tbe analysis framework used in this study can help reveal more micro evolution details.The index to measure topic association strength defined in this paper reflects both similarity and dissimilarity between topics,which belps better understand research topics’evolution paths and content changes.
基金Supported by Scientific Breakthrough Program of Beijing Municipal Science & Technology Commission,China(No. D08050703020803,No.D08050703020804)China Key Technologies R&D Programme(No.2007BA110B06-01)+2 种基金Major State Basic Research Development Program of China (973 Program,No.2006CB504601)National Nature Science Foundation of China(No.90709006)National Science and Technology Major Project of the Ministry of Science and Technology of China(No.2009ZX10005-019)
摘要Induction of common knowledge or regularities from large-scale clinical data is a vital task for Chinese medicine(CM).In this paper,we propose a data mining method,called the Symptom-Herb-Diagnosis topic(SHDT) model,to automatically extract the common relationships among symptoms,herb combinations and diagnoses from large-scale CM clinical data.The SHDT model is one of the multi-relational extensions of the latent topic model,which can acquire topic structure from discrete corpora(such as document collection) by capturing the semantic relations among words.We applied the SHDT model to discover the common CM diagnosis and treatment knowledge for type 2 diabetes mellitus(T2DM) using 3 238 inpatient cases.We obtained meaningful diagnosis and treatment topics(clusters) from the data,which clinically indicated some important medical groups corresponding to comorbidity diseases(e.g.,heart disease and diabetic kidney diseases in T2DM inpatients).The results show that manifestation sub-categories actually exist in T2DM patients that need specific,individualised CM therapies.Furthermore,the results demonstrate that this method is helpful for generating CM clinical guidelines for T2DM based on structured collected clinical data.