查新检索式逻辑运算符选择系统

A Boolean Operator Selection System for Sci-tech Novelty Searching

[目的/意义]在科技查新实践中,部分查新点所对应的检索式包含大量检索词,若检索结果中缺少合适文献,对检索式的进一步调整十分考验查新员的检索技巧。[方法/过程]查新检索式逻辑运算符选择系统以文献与查新点共同涉及内容的信息量来描述文献对查新点新颖性的否定程度,即两者的相关程度,并结合检索、分析功能,采集关键词在数据库中的分布特征计算信息量,评估文献相关性,评估结果可基于查新员反馈实时修正,最终反馈密切相关文献对应的检索式。[结果/结论]实践表明,该系统可辅助多种类型查新点的检索策略制定,尤其适用于涉及关键词较多的查新点。

[Purpose/Significance]In the practice of sci-tech novelty search,the search formulas for some novelty search points contain a large number of search terms.When the search results do not include suitable literature,further adjustments to the search strategy will be difficult,and influenced by the retrieval skills of the searcher.[Method/Process]The Boolean Operator Selection System for Sci-tech Novelty Searching uses the amount of information of the content of novelty search points that is also reported by the literature to quantify the quality of novelty denied by the literature,the relevance of a literature to a novelty search point.And this system has retrieval,analysis,and interaction functions to counts the distribution characteristics of the search terms in each database,calculates the amount of information of the overlapping part in the literature and novelty search points,and evaluates the literature relevance.The evaluation will be automatically corrected based on feedback from the searcher.Finally,the system will generate a search formula based on maximizing the correlation of the results.[Result/Conclusion]Practical use has proven that the system can assist in making retrieval strategies of various types of novelty points,especially for novelty search points that involve a large number of search terms.