[目的/意義]挖掘網絡輿情相關信息,構建基于語義隸屬度模糊推理的網絡輿情監測預警評估方法,對實現網絡輿情危機監測和預警具有重要意義。[方法/過程]采用扎根理論提取、篩選指標要素,使用改進的AHP法確定指標權重,通過二級模糊綜合評判方法判定輿情信息的語義隸屬度,建立基于語義隸屬度模糊推理的網絡輿情監測預警模型。[結果/結論]能夠對網絡輿情事件進行有效監測和預警,為網絡輿情預警評估的實現和應用提供一個整體、系統的科學依據,提高網絡輿情實時監控能力和管理能力。[局限]諸多平臺數據和隱性指標數據未能予以完全考慮。

[Purpose/significance] It is very important to realize network public opinion crisis monitoring and early warning by mining relevant information and constructing network public opinion monitoring and early warning evaluation method based on fuzzy reasoning of semantic membership. [Method/process] First,indexes are chosen by grounded theory. Then,the AHP method is used to determine the index weight. Last,through the two-level fuzzy comprehensive evaluation method to determine the semantic membership degree of public opinion information,the paper establishes a network public opinion monitoring and early-warning model based on the fuzzy membership of the semantic membership degree. [Result/conclusion] The model can effectively monitor and warn the network public opinion crisis,and provide a systematic and scientific basis for the realization and application of network public opinion early warning assessment,thus enhancing the network public opinion real-time monitoring ability and management ability. [Limitations] Many platform data and implicit indicator data have not been fully considered.