需要金幣： 1000 個(gè)金幣 |
資料包括：完整論文 |  |
 |
| 轉(zhuǎn)換比率：金額 X 10=金幣數(shù)量， 例100元=1000金幣 | 論文字?jǐn)?shù)：12580 |  | |
| 折扣與優(yōu)惠：團(tuán)購最低可5折優(yōu)惠 - 了解詳情 | 論文格式：Word格式(*.doc) |  |
摘要：貝葉斯分類器的分類原理是通過某對(duì)象的先驗(yàn)概率，利用貝葉斯公式計(jì)算出其后驗(yàn)概率，即該對(duì)象屬于某一類的概率，選擇具有最大后驗(yàn)概率的類作為該對(duì)象所屬的類。目前研究較多的貝葉斯分類器主要有四種，分別是：Naive Bayes、TAN、BAN和GBN。 貝葉斯網(wǎng)絡(luò)是一個(gè)帶有概率注釋的有向無環(huán)圖，圖中的每一個(gè)結(jié)點(diǎn)均表示一個(gè)隨機(jī)變量,圖中兩結(jié)點(diǎn) 間若存在著一條弧，則表示這兩結(jié)點(diǎn)相對(duì)應(yīng)的隨機(jī)變量是概率相依的，反之則說明這兩個(gè)隨機(jī)變量是條件獨(dú)立的。網(wǎng)絡(luò)中任意一個(gè)結(jié)點(diǎn)X 均有一個(gè)相應(yīng)的條件概率表(Conditional Probability Table，CPT)，用以表示結(jié)點(diǎn)X 在其父結(jié)點(diǎn)取各可能值時(shí)的條件概率。若結(jié)點(diǎn)X 無父結(jié)點(diǎn),則X 的CPT 為其先驗(yàn)概率分布。貝葉斯網(wǎng)絡(luò)的結(jié)構(gòu)及各結(jié)點(diǎn)的CPT 定義了網(wǎng)絡(luò)中各變量的概率分布。 基于貝葉斯理論的樸素貝葉斯分類（Naive Bayes，NB）方法是一種簡單而有效的分類方法，它也是機(jī)器學(xué)習(xí)領(lǐng)域中應(yīng)用廣泛的分類算法之一。本文介紹了樸素貝葉斯分類算法的基本原理，研究了基于樸素貝葉斯算法的數(shù)據(jù)分類。實(shí)際應(yīng)用表明了樸素貝葉斯算法是一種有效的分類算法。 關(guān)鍵詞：樸素貝葉斯算法；文本分類；數(shù)據(jù)
Abstract：Bayesian classifier principle is the prior probability of an object by using the Bayesian formula to calculate the probability of subsequent experience, that the object belongs to a class of probability, choose a maximum a posteriori probability of the class as the object belongs to the class. Currently studied in Bayesian classifier, there are four, namely: Naive Bayes, TAN, BAN and GBN. Bayesian network is annotated with a probability of a directed acyclic graph, each node in a random variable indicated the figure between two nodes if there is an arc, then the corresponding two nodes the probability of a random variable is dependent, and vice versa indicated that the two random variables are independent conditions. Any network node X has a corresponding conditional probability table (Conditional Probability Table, CPT), to indicate the node in its parent node X to take all possible values ??of the conditional probability. If no parent node node X, then X, the prior probability distribution for CPT. Bayesian network structure and the nodes of the network in the CPT definition of the probability distribution of each variable. Based on Bayesian theory Bayesian classification (Naive Bayes, NB) is a simple and effective classification method, which is widely used in the field of machine learning classification algorithms. This article describes the simple basic principle of Bayesian classification algorithm, Naive Bayes algorithm is studied based on data classification. The application shows that the Naive Bayes algorithm is an effective classification algorithm. Keywords：Naive Bayes algorithm；Text Classification；Data
|
