The War Function Mining Based On GEP And Application In Military

Postgraduate: LuoSongTao Supervisor: Prof. TANG ChangjieEvery affect factor resulted in victory or defeat of the war is always the hot the statesman, military and historian focus on. But at present the research of the war's affected factor, for the most part, using traditional method. Data mining is a process that discovers knowledge from mass data and information. With this tool, military history mining aims at finding universal rules of military affairs from historical data, grasping the progress of future warfare, make great sense to military research.The expressional of the military history is text. But the traditional text's processing is not fit the military history. The data is fuzziness and not accurately show in military datum. So mining the data can't use general data mining method. Analyzing this class's data need preprocessing the data first, making it standardization, and then obtain relational knowledge by mining the military history.The main contributions of this paper include:(1) Analyzing the character of military history material and introducing data mining technology into the process of coordinating materials;(2) Advancing the method of disposal the correlated data with fuzzy function;(3) Integrate experiential knowledge and information entropy to selecting attribute dimensions, and wipe off independent or weak-correlated attributes;(4) Mining the warfare influence factor by Gene Expressing Programming from the pretreated data。At last this paper proving the effectiveness of the new algorithm by extensive experiments and analyzing the experimental results detailed. The paper is organized as follows. Chapter 1 summarizes data mining. Chapter 2 briefly introduces GEP. Chapter 3 introduces the fuzzy mathematics. Chapter 4 gives the pretreatment method for the military fuzziness data and the algorithm for GEP mining. Chapter 5 speaks of the military function mining's application. Chapter 6 come up with experiments, and evaluates the results. Chapter 7 draws a conclusion discusses prospects for the future works.