本研究利用主題模型方法發現資訊計量學研究主題的結構，探討主題的逐年與發展以及比較相關期刊。本研究利用檢索自Web of Science資料庫的Journal of Informetrics和Scientometrics期刊2007到2013年間的論文資料，做為主題模型方法的輸入，推算主題結構的研究資料。研究結果發現主題數量設定為10的主題模型具有最小的混淆度。雖資料範圍與分析方法不同，本研究產生的主題仍可和由專家分析產生的結果相容。實務的案例研究與書目計量指標的測量方式在各年度都較受重視，且整個領域日趨穩定。兩種核心期刊都廣泛地涉及資訊計量學的所有主題，但Journal of Informetrics特別著重於書目計量指標的建構與應用，而Scientometrics則關注於國家、機構、領域及期刊上的學術生產力評鑑方式與影響因素的探討。

In this study, we used the approach of topic modeling to uncover the possible structure of research topics in the field of Informetrics, to explore the distribution of the topics over years, and to compare the core journals. In order to infer the structure of the topics in the field, the data of the papers published in the Journal of Informetrics and Scientometrics during 2007 to 2013 are retrieved from the database of the Web of Science as input of the approach of topic modeling. The results of this study show that when the number of topics was set to 10, the topic model has the smallest perplexity. Although data scopes and analysis methods are different to previous studies, the generating topics of this study are consistent with those results produced by analyses of experts. Empirical case studies and measurements of bibliometric indicators were concerned important in every year during the whole analytic period, and the field was increasing stability. Both the two core journals broadly paid more attention to all of the topics in the field of Informetrics. The Journal of Informetrics put particular emphasis on construction and applications of bibliometric indicators and Scientometrics focused on the evaluation and the factors of productivity of countries, institutions, domains, and journals.