【摘要】：大數(shù)據(jù)時代下,移動互聯(lián)網(wǎng)發(fā)展與移動終端的普及形成了海量移動對象軌跡數(shù)據(jù).軌跡數(shù)據(jù)含有豐富的時空特征信息,通過軌跡數(shù)據(jù)處理技術(shù),可以挖掘人類活動規(guī)律與行為特征、城市車輛移動特征、大氣環(huán)境變化規(guī)律等信息.海量的軌跡數(shù)據(jù)也潛在性地暴露出移動對象行為特征、興趣愛好和社會習(xí)慣等隱私信息,攻擊者可以根據(jù)軌跡數(shù)據(jù)挖掘出移動對象的活動場景、位置等屬性信息.另外,量子計算因其強大的存儲和計算能力成為大數(shù)據(jù)挖掘重要的理論研究方向,用量子計算技術(shù)處理軌跡大數(shù)據(jù),可以使一些復(fù)雜的問題得到解決并實現(xiàn)更高的效率.對軌跡大數(shù)據(jù)中數(shù)據(jù)處理關(guān)鍵技術(shù)進行了綜述.首先,介紹軌跡數(shù)據(jù)概念和特征,并且總結(jié)了軌跡數(shù)據(jù)預(yù)處理方法,包括噪聲濾波、軌跡壓縮等;其次,歸納軌跡索引與查詢技術(shù)以及軌跡數(shù)據(jù)挖掘已有的研究成果,包括模式挖掘、軌跡分類等;總結(jié)了軌跡數(shù)據(jù)隱私保護技術(shù)基本原理和特點,介紹了軌跡大數(shù)據(jù)支撐技術(shù),如處理框架、數(shù)據(jù)可視化;也討論了軌跡數(shù)據(jù)處理中應(yīng)用量子計算的可能方式,并且介紹了目前軌跡數(shù)據(jù)處理中所使用的核心算法所對應(yīng)的量子算法實現(xiàn);最后,對軌跡數(shù)據(jù)處理面臨的挑戰(zhàn)與未來研究方向進行了總結(jié)與展望.
[Abstract]:Under big data era, the development of mobile Internet and the popularity of mobile terminals have formed massive mobile object trajectory data. Trajectory data contain abundant space-time characteristic information. By using track data processing technology, we can mine the characteristics of human activity and behavior, the characteristics of urban vehicle movement, the law of atmospheric environment change and so on. Huge amounts of trajectory data also potentially expose the behavior characteristics interests and social habits of moving objects and other privacy information. An attacker can mine the moving objects' activity scene location and other attribute information based on the trajectory data. In addition, because of its powerful storage and computing ability, quantum computing has become an important theoretical research direction for big data. The use of quantum computing technology to deal with the trajectory big data can solve some complex problems and achieve higher efficiency. The key techniques of data processing in trajectory big data are reviewed. Firstly, the concept and characteristics of trajectory data are introduced, and the methods of trajectory data preprocessing are summarized, including noise filtering, trajectory compression and so on. Secondly, the research achievements of trajectory index and query technology and track data mining are summarized, including pattern mining, trajectory classification and so on. This paper summarizes the basic principles and characteristics of trajectory data privacy protection technology, and introduces the trajectory big data support technology, such as processing framework, data visualization; The possible ways of applying quantum computation in trajectory data processing are also discussed, and the implementation of quantum algorithms corresponding to the core algorithms used in trajectory data processing is also introduced. Finally, the challenges and future research directions of trajectory data processing are summarized and prospected.
【作者單位】： 電子科技大學(xué)信息與軟件工程學(xué)院;Department
【基金】：國家自然科學(xué)基金(61602097,61272527) 四川省科技廳計劃(2015JY0178) 四川省科技支撐計劃(2016GZ0065,2016GZ0063) 中央高�；究蒲袠I(yè)務(wù)費(ZYGX2014J051,ZYGX2011J066,ZYGX2015J072) 中國博士后基金(2015M572464)~~
【分類號】：TP311.13

【參考文獻】

相關(guān)期刊論文 前7條

1 許佳捷;鄭凱;池明e，

本文編號：2301588