本文選題：恐怖組織網(wǎng)絡 + 變化檢測��； 參考：《科技導報》2017年09期

【摘要】：恐怖組織的社會網(wǎng)絡結構變化與恐怖事件的發(fā)生具有一定的關聯(lián)性。基于此關聯(lián),通過監(jiān)測恐怖組織社會網(wǎng)絡的變化,可以實時、有效地識別恐怖事件。將基于社會網(wǎng)絡變化檢測的恐怖事件監(jiān)測與識別問題視為分類問題,并通過神經(jīng)網(wǎng)絡模型進行分類研究。以某一時刻是否發(fā)生恐怖事件為標準,對恐怖組織社會網(wǎng)絡進行分類;通過網(wǎng)絡分析技術,得出網(wǎng)絡的參數(shù)指標,建立混合算法改進的神經(jīng)網(wǎng)絡模型;將網(wǎng)絡的參數(shù)指標與恐怖事件發(fā)生情況分別作為輸入和輸出,對神經(jīng)網(wǎng)絡進行訓練與測試。案例分析和對比結果表明,基于神經(jīng)網(wǎng)絡模型的社會網(wǎng)絡變化檢測方法具備較好的恐怖事件監(jiān)測與識別能力;該方法可在一定程度上彌補現(xiàn)有方法正確率不高、通用性不強、檢測結果與恐怖事件實際發(fā)生的相關性不高等不足。
[Abstract]:The change of social network structure of terrorist organizations is related to the occurrence of terrorist events. Based on this correlation, terrorist events can be recognized in real time and effectively by monitoring the changes of social networks of terrorist organizations. The problem of terrorist event detection and identification based on social network change detection is considered as a classification problem, and the classification is studied by neural network model. The social network of terrorist organization is classified according to whether or not a terrorist event occurs at a certain time, the parameter index of network is obtained by network analysis technology, and the improved neural network model of hybrid algorithm is established. The parameters of the neural network and the occurrence of terrorist events are taken as input and output respectively to train and test the neural network. The results of case analysis and comparison show that the social network change detection method based on neural network model has a better ability to detect and identify terrorist events, and this method can make up for the lack of accuracy and universality of the existing methods to some extent. The correlation between the detection results and the actual occurrence of terrorist events is not high enough.
【作者單位】： 國防科技大學信息系統(tǒng)與管理學院;
【基金】：國家自然科學基金項目(71473263) 高等學校博士學科專項科研基金項目(20134307110020)
【分類號】：C912.3
，

本文編號：1942113