基于改進SVM算法的思政教育動態預警系統研究

席衛華

摘 要： 針對目前高職院校思政教育工作存在的信息反饋滯后、預警干預機制不完善、風險管控能力較弱等問題，開發了一款基于改進SVM算法的思政教育動態預警系統。通過構建體量較大的基于用戶行為的日志數據庫，引入改進過的支持向量機（SVM）算法并融入預警分類器建立面向思政教育動態預警模型，在Matlab2016b環境下進行模型效能仿真驗證，較好解決了多維應用背景下的高校思政教育動態預警過程中人力耗費與實際效能失衡、信息反饋滯后等問題，具有動態預警精準、泛化預警能力強、風險管控變化趨勢預估效率高等優勢。以我國東部某高職院校為效能評價載體，利用VS2012平臺開發了驗證環境并對模型進行了實證分析，分析結果表明所提模型可以實現全方位的高校思政教育動態預警，在預警適應性、模型擬合度、信息過載處理效率等方面具有明顯優勢。

關鍵詞： 思政教育; 動態預警模型; 改進支持向量機算法; 預警分類器; 系統開發

中圖分類號： TP 315 ? ? ?文獻標志碼： A

Abstract： Aiming at the problems of lag in information feedback， imperfect early warning intervention mechanism and weak risk management and control ability in the ideological and political education work of higher vocational colleges， a dynamic early warning system based on improved SVM algorithm was developed. By constructing a large user database based on user behavior， an improved support vector machine （SVM） algorithm is introduced and an early warning classifier is built to establish a dynamic early warning model for ideological and political education. The model performance simulation is validated in Matlab 2016b environment. It solves the problems of imbalance between manpower consumption and actual efficiency and lag of information feedback in the dynamic early warning process of ideological and political education in colleges and universities under the background of multi-dimensional application. It has the advantages of dynamic early warning accuracy， strong generalized early warning ability， and high risk estimation performance. Taking a higher vocational college in eastern China as the effectiveness evaluation carrier， the verification environment is developed by using VS2012 platform， and the model is empirically analyzed. The analysis results show that the model proposed in this paper can realize the dynamic early warning of ideological and political education in colleges and universities. Model fit， information overload processing efficiency and other aspects have obvious advantages.

Key words： ideological and political education; dynamic early warning model; improved support vector machine algorithm; early warning classifier; system development

0 引言

思政教育是高等教育的重要組成部分，貫穿高等教育教學全過程，是高職院校培養社會主義建設者和接班人的重要制度載體，在強化高職院校在校生的價值引領、加強愛國奉獻主旋律宣傳教育，建立全員全過程全方位育人機制等領域發揮著基礎性作用。高職院校思政教育涉及面廣、受眾層次多樣、未形成統一的標準模式[1]，屬于復雜的系統工程，必須采用系統化的方法加以分析，借鑒協同理論，搭建協同平臺、完善協同機制、營造協同文化、建立動態預警模型是實現高職院校思政教育全局協同管控的必經之路?！?br>