【摘要】： “金盾工程”建設(shè)在全國進入全面推進的新階段，城市公安信息化基礎(chǔ)工作建設(shè)取得了巨大成績，同時也存在一些問題，，如公安各類信息資源沒有得到充分挖掘和合理利用，不能以簡便、靈活的應(yīng)用手段為領(lǐng)導(dǎo)和一線民警開展工作提供綜合信息和科學(xué)決策支持。近年來發(fā)展的數(shù)據(jù)挖掘技術(shù)能夠發(fā)現(xiàn)數(shù)據(jù)中隱藏的規(guī)律，起到輔助決策的作用。本文研究的主要內(nèi)容是數(shù)據(jù)挖掘技術(shù)在城市公安應(yīng)用研究中一個重要部分——刑事案件的關(guān)聯(lián)分析模型。 本文首先提出了城市公安刑事案件關(guān)聯(lián)分析模型的總體框架，然后研究了模型各步驟中使用的方法與算法，主要有以下幾個方面：在數(shù)據(jù)預(yù)處理過程中，給出了適合公安數(shù)據(jù)提取與清理的策略，在此基礎(chǔ)上，建立刑事案件多維數(shù)據(jù)模型，包括選取星型數(shù)據(jù)模式，使用分箱等方法對數(shù)據(jù)進行離散化與概念分層處理，建立數(shù)據(jù)立方體。然后進行多維關(guān)聯(lián)規(guī)則挖掘，利用改進的Apriori算法尋找頻繁謂詞集，按照最小支持度與最小置信度的要求在頻繁謂詞集中產(chǎn)生強關(guān)聯(lián)規(guī)則。在進行多層關(guān)聯(lián)規(guī)則挖掘時，本文分析選取了層交叉單項過濾策略，通過修改多維頻繁謂詞集算法來發(fā)現(xiàn)不同概念層中的頻繁謂詞集，產(chǎn)生強關(guān)聯(lián)規(guī)則，最后利用檢驗冗余規(guī)則原則去除冗余結(jié)果。在本文的最后，利用大連市公安局提供的刑事案件數(shù)據(jù)，按照模型的方法與步驟，選擇了適當(dāng)?shù)墓ぞ�，完成了大連市甘井子區(qū)刑事案件的關(guān)聯(lián)分析，驗證了模型的正確性與有效性。 本文的研究為今后城市公安數(shù)據(jù)挖掘應(yīng)用研究提供了參考，對輔助公安決策具有理論意義與現(xiàn)實意義。
[Abstract]:The construction of the "Golden Shield Project" has entered a new stage of comprehensive promotion throughout the country, and great achievements have been made in the construction of the basic work of urban public security informatization. At the same time, there are also some problems, such as the insufficient exploitation and rational utilization of all kinds of public security information resources. The simple and flexible application means can not provide comprehensive information and scientific decision support for leaders and frontline police. The data mining technology developed in recent years can discover the hidden rules in the data and play an assistant role in decision-making. The main content of this paper is the association analysis model of criminal cases, which is an important part of the application of data mining technology in urban public security. This paper first puts forward the general framework of the criminal case association analysis model of the city public security, and then studies the methods and algorithms used in each step of the model. The main aspects are as follows: in the process of data preprocessing, On the basis of this, the multi-dimensional data model of criminal case is established, including selecting star data model, using the method of dividing boxes and so on, to discretize the data and deal with the concept stratification. Create a data cube. Then the multi-dimension association rule mining is carried out and the improved Apriori algorithm is used to find the frequent predicate set. Strong association rules are generated in the frequent predicate set according to the requirement of minimum support degree and minimum confidence degree. In the mining of multi-level association rules, this paper analyzes and selects the layer cross single item filtering strategy, by modifying the multi-dimensional frequent predicate set algorithm to find the frequent predicate sets in different concept layers, and produces strong association rules. Finally, the redundancy results are removed by using the principle of checking redundancy rules. At the end of this paper, by using the criminal case data provided by Dalian Public Security Bureau, according to the method and steps of the model, the appropriate tools are selected, and the correlation analysis of criminal cases in Ganjingzi District of Dalian City is completed. The correctness and validity of the model are verified. The research in this paper provides a reference for the application of urban public security data mining in the future, and has theoretical and practical significance to assist public security decision-making.
【學(xué)位授予單位】：大連理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2007
【分類號】：D631.2

【引證文獻】

相關(guān)期刊論文 前1條

1 王春雨;王雪華;郭發(fā)志;葉鑫;;大連市社會治安防控警務(wù)模式研究[J];大連理工大學(xué)學(xué)報(社會科學(xué)版);2009年04期

相關(guān)博士學(xué)位論文 前1條

1 王春雨;刑事案件關(guān)聯(lián)分析與防控警務(wù)模式研究[D];大連理工大學(xué);2010年

相關(guān)碩士學(xué)位論文 前1條

1 王娜;一種冗余規(guī)則刪減方法及其應(yīng)用[D];大連理工大學(xué);2011年

本文編號：2255818