There are several researchers who

have focused on data mining techniques, particularly classification, But a few

papers about Lunar classification. In 2, the authors presented a Bayesian

network approach for short-term solar flare level prediction based on three sequences

of photospheric magnetic field parameters extracted from Solar and Heliospheric

Observatory/Michelson Doppler Imager longitudinal magnetograms. The subjective

network structure which depicts conditional independent connections among

magnetic field parameters and the quantitative conditional likelihood tables

which decide the probabilistic esteems for every factor are found out from the

informational index. They use dimensions reduction as preprocessing technique.

Two Bayesian network models are assembled using raw sequential data (BN_R) and

feature extracted data (BN_F), respectively. The clarifications of these models

are reliable with physical examinations of specialists. The performances of the

Bayesian network model are higher than the the2 performance of the naive Bayes

model.

Paper 3 presented how to Estimate

the availability of sunshine by Specific data mining techniques, the author

takes into consideration the requirement of the data for his paper and did

pre-process of data by selecting the feature(attribute) that needed for

clustering. so, the Number of years and Mean Rainfall attributes in millimeter

were reduced from the data set. in 3 the author uses Clustering method which

makes groups(clusters) of a similar type of data. There are different types of

clustering methods, the algorithms that 3 decide used is Simple K-Means and

Expectation Maximization algorithm which is under the partition method of

clustering. The partition method is based on the greedy heuristics in which

they are used in an iterative manner to obtain a local optimum solution. A simple k-means algorithm uses Euclidean

distance method for distance calculation. The simple k-means method is

described as first, Select the number of clusters (k) then assume k seeds as centroids

of the k clusters. The seeds could be chosen randomly by the user if the values

of data are unknown. after that, Compute the Euclidean distance of each object

of the dataset from each of the centroids. next, allocate all objects to the

cluster if the distance between the centroid of the cluster and the object is

small. after that, Compute the centroids of the clusters by calculating the

means of attribute values of the objects in the cluster. The last step is to

stop the algorithm if the stopping criterion is met or Compute the Euclidean

distance of each object again, and continue the procedure. In this paper, the

simple k-means algorithm is computed using the open source software tool Weka.

and they use Expectation maximization (em) algorithm which works in contrast to

the simple k-means algorithm. it on the concept of assuming that the objects in

the dataset have attributes whose values are distributed and we can describe it

as Assuming the initial values then, use the normal distributions and calculate

the probability of each object belonging to the two clusters. Then, calculate

the possibility of data coming from the two clusters. Finally, iterate the

process by re-assuming the parameters and go to the normal distributions and

calculate the probability again till the stopping criterion is met. The dataset

of the monthly mean maximum and a minimum temperature of Chennai, Coimbatore,

Madurai, and Kanyakumari were gathered and simple k-means an expectation

maximization algorithm were used. The clusters formed indicated that the

maximum monthly mean temperature was recorded in the month of May and June

which implied the maximum sunshine hours in these months. The maximum in

temperature implies the maximum daylight or sunshine hours. The sunshine hours

determine the amount of solar radiation that can be acquired. The maximum

amount of sunshine is recorded in the city of Chennai when compared to the

other cities in the dataset such as Coimbatore, Madurai, and Kanyakumari. In

3 the author uses Weka environment to load data.