K-NN is to find the distance between each variable (Attribute) in the data, then calculate out which way is suitable for data model numbers. But the variable is discrete, it can do.The algorithm K-NN process are as follows:
.1. Set the size of K (should be designated as odd)
2. Calculate the distance (Distance) of data to consider with a group of data sample
3.Sort order of distance. Select data set and consider approaching the point wants to consider according to the number of K defined
4. Considering the information of K series and observation group (class) where near point is considered as the most 5
.The class to points considering
.The algorithm is an algorithm K-NN Li algorithm used in data clustering. By organizing data are close to each other as the same group. By checking the number K which, if the conditions of the decision complexity.But the algorithm K-NN duration calculation for a long time. If variables (Attribute) have a lot of problems in the calculation of value. And quite use quantity to calculate very high on the computer. Because of the time used for calculation.So in order to speed up techniques for algorithm K-NN more, all the information is often used to is stored in a memory (Memory). By means of access to the memory. The basic reason (Memory-Based Reasoning).In the store the class of algorithms K-NN in memory, and if the desired information for words ตอบมี independent variables, only a few. To understand the model algorithm K-NN easier.Related to the type of information that is not standard, such as text (Text) but may have the standard measurement for the kind of information right. The efficiency of the algorithm K-NN depends on the number of distance.That can discriminate effectively between normal data and abnormal data. To explain the distance between the information is extremely challenging when the data is complex, for example, the chart data and information, respectively. Etc.
.
การแปล กรุณารอสักครู่..
