Weka

(
IRIS ANALYSIS - WEKA (SEPTEMBER 2012)
Acuña Frances, Cárdenas Flavia, Ruiz Teresa, Vasquez Gabriela, euroresidentes

Abstract—


The purpose of this paper is to provide an analysis of the model in Weka for flower iris an it’s classification. We will use some tools, such as J48 to evaluate the variables, because it is the most accurate and trustful model for managing information.In addition J48 is the only tool that provides the three analysis, which is beneficial to have a complete evaluation criteria. That is why, we will also present a comparison of the results obtained with j48 and the other classicators.


Index Terms


J48 Decision Trees: A decision tree is a predictive machine-learning model that decides the target value (dependent variable) of a newsample based on various attribute values of the available data. The internal nodes of a decision tree denote the different attributes, the branches between the nodes tell us the possible values that these attributes can have in the observed samples, while the terminal nodes tell us the final value (classification) of the dependent variable.


WEKA: Weka (Waikato Environment for Knowledge Analysis- Environment for Knowledge Analysis Waikato University) is a software platform for machine learning and data mining written in Java and developed at the University of Waikato. Weka is Free Software released under the GNU-GPL. The Weka workbench contains a collection of visualization tools and algorithms for data analysis and predictive modeling, together with graphical user interfaces for easyaccess to this functionality.

INTRODUCTION

T
his paper is based on “Classification” using decision trees based on Dr. Polczynski’s Lecture”.
The general objectives of this sample are: Know attributes and classes for instances, specifically, and know sepal and petal dimensions and species for samples.
The final objective of the model is to determine a Weka decision tree toclassify unknown iris samples.
The paper of Dr. Polczynski’s is a a tutorial that shows the functions and results of the program Weka. It explain us step by step how to analyze the information that we put in the program, and also the problems that it can generate.





Description of iris species


1 Iris setosa: Iris setosa, the Arctic Iris, is native to the northern parts of NorthAmerican and into Asia. It is a perennial growing to 0.6 meter by 1 meter.

The flowers are consisting of 3 sepals and 3 petals. The blue flowers have darker veins. They are borned on a small erect stem, about 18 inches high. It is in flower in June, and the seeds ripen in August. The flowers are hermaphrodite, have both male and female organs. The plant is self-fertile.




2 Irisversicolor: is versicolor is a species belonging to the family (Iridaceae), is a native of North America and grows in swampy and marshy places. Lily is also commonly referred to any species of the genus Iris.It is a herbaceous plant with fleshy root horizontal, cylindrical and very similar to the pen. Reaches 100 cm with flexible branches. The leaves are larger than 30 cm long straight sword shaped andgrooved. The flowers are grouped in 2-6 and are blue or purple. The sepals are spatulate or lanceolate. The fruit is a capsule containing numerous flat see


3 Iris virginica: Virginia iris is a perennial plant. The slightly fragrant flowers (4 cm long, 7 cm across) consist of 3 horizontal sepals, or “falls,” and 3 erect petals. The petals and sepals can vary in color from dark-violet topinkish-white. The sepals have a splash of yellow to yellow-orange at the crest. Each plant has 2 to 6 flowers that bloom from April to May upon a single, erect, 3-9 dm tall stalk. The stalk is sometimes branched and has a slight zigzag appearance. The plant has 2 to 4 erect or arching, bright green, lance-shaped leaves that are flattened into one plane at the base. Leaves are 1 – 3 cm wide and are...