0, EDITORIAL words I personally have always liked algorithm for a class of things, in my opinion is the essence of human intelligence algorithm, which contains unparalleled beauty. And every time the learned algorithms to practice, and to solve practical problems, the kind of pleasure is elsewhere I can not Air Jordan Outlet understand. Always wanted to write about algorithms Bowen, he has written two scattered, but perhaps compared to the engineering of the article is too small for the Nike Zoom Kobe congregation, and did not arouse interest. Recently faced graduate looking for work, in order to give themselves a bargaining chip, decided to review the terms of knowledge algorithm again, I decided to take advantage of this opportunity to write a series of articles about the algorithm. To do so, mainly to strengthen their Nike Air Jordan 6 Women review of the results, I think, if able to review things with their own understanding of written in the article, it is bound to do more than simply reading the title master firmer and better able to trigger their thinking. If you can have friends interested gain from that natural is better. I named this series 'algorithm grocery store', the reason is a major feature of these articles LeBron James Shoes is the 'miscellaneous', I will not be devoted to the stack, linked lists, binary trees, search, sort and any other one will be talking about data structures textbook the basics, I would from a 'special' departure, such as probability algorithm, classification algorithms, NP problem, genetic algorithms, and then do a extended, may involve algorithms and data structures, discrete mathematics, probability theory, statistics, operations many aspects of learning, data mining, in the form of language and automatic machines, so the content structure like a grocery store. Of Nike Basketball course, I will do my best to try to make the contents of 'miscellaneous and not chaos.' 1.1 Summary of Bayesian classification is a general term for class classification algorithm, these algorithms are based on Bayes' theorem, it is referred to as the Bayesian classifier. In this paper, the first chapter as the classification algorithm will first introduce the classification, the classification issue a formal definition. Then, introduce basic Bayesian classification algorithm - Bayes theorem. Finally, we discuss Bayesian classification easiest one by way of example: Naive Bayes classifier. 1.2 Summary of the classification for the classification problem, in fact, who are no strangers to say every one of 378037 010 Original?CBlack-True Red-White Air Jordan 11 Retro Nike Kids Sneakers Factory Outlet us every day in the implementation of classification operation is not an exaggeration, but we did not realize nothing. For example, when you see a stranger, your brain subconsciously judge TA is male or female; you may often go on the road beside the friend said, 'This man is very rich look, over there a non-mainstream 'kind of thing, in fact, this is a classified operation. From a mathematical point of view, the classification problem do the following definition: the known collections: and, to determine the mapping rules, so there is one and only one makes any establishment. (Irrespective of fuzzy mathematics in the case of fuzzy sets) where C is called the set of categories, where each element is a category, and I called the collection of items, where each element is an item to be 2015 Nike Free 5.0 classified, f called a classifier. Classification algorithm task is to construct a classifier f. Here we must emphasize, classification problems are often constructed using empirical methods mapping rules, namely the classification of the general case of a lack of sufficient information to construct 100% correct mapping rules, but by learning to empirical data in order to achieve a certain probability sense correct classification, so the trained classifier is not necessarily able to accurately map each item to be classified with a classifier constructor quality classification, classification, features and the number of training samples and other data to be classified many factors ʱ?? For example, doctors diagnose the patient is a typical classification process, any doctor can not directly see the patient's condition can only observe the patient exhibits symptoms and a variety of laboratory testing data to infer the condition, then the doctor is like a classifier, and the accuracy of diagnosis of the doctor, and he had to be educated (the constructor), the patient's symptoms are prominent (data to be classified characteristics) and how much experience the doctor (training samples) are closely related. 1.3 basis of Bayesian classification - each mention of Bayes 'Theorem Bayes' theorem, reverence my mind are spontaneously, not because of this theorem more profound, but because it is particularly useful. This theorem to solve the problem often encountered in real life: a condition known probability, how to get the probability Nike Air Max of two events after the exchange, which is known Nike Air Max 2011 Men P (A | B) Nike Air Max 2011 Men under the circumstances how to obtain P (B | A). Here to explain what is the conditional probability: that the premise of the event B has occurred, the probability of an event A occurs, the conditional probability of event A is called the next event B occurs. Solving the basic formula is:. Bayes' theorem reason useful because we often encounter such a situation in life: we can easily be directly obtained P (A | B), P (B | A) it is difficult to directly draw, but we more concerned with P (B | A), Bayes' theorem is for us to get through from P (A | B) to obtain P (B | A) of the road. Below is given Lebron Slide 2 Elite without proof directly Bayes' theorem: 1.4, 1.4.1 Naive Bayes classifier, principles and processes Naive Mens Nike Free 3.0 V2 Shoes Black Blue Bayes Naive Bayes classifier is a very simple classification algorithm, called it simple Bayesian classification because the idea of this method is really very simple, naive Bayes ideological foundation is this: For a given entry to be classified, to solve the probability of each category appear under the conditions of this emerging, which the largest, considered this to be classified items belong to which category. Popular, like such a reason, you see a black man in the street, I ask you, where do you think this man come, you probably guessed Africa. why? Because the highest rate of African blacks, of course, they were also likely to be American or Asian, but no other information is available, we will choose the largest category of conditional probability, which is the ideological basis Naive Bayes. The formal definition of Bayesian classifier as follows: 1, set an item to be classified, and each a is a characteristic property of x. 2, there are categories of collections. 3. Calculate. 4, if, then. So the key now is how to calculate the conditional probability of each step 3. We can do: 1, to find a collection of items Nike Air Max 2011 to be classified a known classification, this set is called the training sample set. 2. Statistical obtain the conditional probability of each feature property is estimated in each category. which is. 3, if the individual characteristic attributes are independent conditions, the following is derived based on Bayes' theorem: because the denominator is constant for all categories, because we simply maximize molecule can be. And because each Air Max 2011 Dark Black Blue feature is independent of the condition of the property, so there: Based on the above analysis, process Naive Bayes classifier can be represented by the following diagram (not to consider verification): You can see the whole Bayesian classifier is divided into three Stage: The first stage - preparatory work phase, the phase of the mission is to make the necessary preparations for the Bayesian classifier, the main work is to determine the characteristic properties depending on the circumstances, and the appropriate division of the attributes of each feature, and then manually items to be classified on the part of the classification, the formation of the training sample set. Enter this stage is that all the data to be classified, the output is characteristic attributes and training samples. This stage is the whole naive Bayes classifier only need to manually complete the stage, its quality will have an important impact on the entire process, the quality of classifiers largely characteristic property division and the training sample quality is determined by the characteristics of property. The second stage - classifier training phase, the phase of the mission is to generate a classifier, the main work is to calculate the Mens Nike Free 3.0 V2 Shoes Grey Green frequency of occurrence for each category and each feature property divided in training samples for the conditional probability estimate for each category, and Results recording. Its input is a feature attribute and training samples, output is classifier. This stage is the mechanical stage, it can be automatically calculated by the program is completed according to the formula discussed earlier. The third stage - application stage. This phase of the mission is to treat classified using a classifier to classify items, whose input is to be classified items classifier and the output is to be classified item and category mappings. This stage is the mechanical phase, completed by the program. 1.4.2, under the category of conditional probabilities estimated characteristic attribute division and Laplace calibration