摘要：

【目的】 构建期刊热词评估指标，并通过热词排行榜服务，帮助用户快速直观地了解期刊研究的前沿领域和方向。【方法】 利用网络文献调研法分析当前智能分词词库析出词条特征；利用参数化、标准化唯一标记等方法提取与热词关联的计算参数，并进行统计分析，构建数学模型，进行多维度排序。【结果】 构建期刊有效词条析出模型，对有效词条进行筛选和标准化标记，对带有参数信息的有效词条通过逻辑计算构建热词评估指标数学模型，并以排行榜的形式给出热词指标知识服务的应用示例。【结论】 标准化唯一标记方法可提升分词词库的词条识别能力，使分词结果更加专业、可靠；期刊热词排行榜服务可帮助用户快速、直观地了解期刊研究的前沿领域和方向。

关键词: 科技期刊, 评估指标, 知识服务, 智能分词

Abstract:

[Purposes] This paper aims to construct evaluation indicators for hot words in journals and help users quickly and intuitively understand the cutting-edge fields and directions of journal research through the hot word ranking service. [Methods] The online literature research method was used to analyze the features of extracted entries in the current intelligent word segmentation lexicon; methods such as parameterized and standardized unique labeling were used to extract computational parameters associated with hot words, and statistical analysis was conducted. Mathematical models were built, and multidimensional sorting was performed. [Findings] A journal effective entry extraction model is constructed, and effective entries are screened and labeled in a standardized manner. A mathematical model for hot word evaluation indicators is constructed through logical calculation for effective entries with parameter information. An application example of knowledge service of hot word indicator is presented in the form of a ranking list. [Conclusions] The standardized unique labeling method can improve the entry recognition ability of the segmentation lexicon, making the segmentation results more professional and reliable. The journal hot word ranking service can help users quickly and intuitively understand the cutting-edge fields and directions of journal research.

Key words: Scientific journal, Evaluation indicator, Knowledge service, Intelligent word segmentation