Résumé Macroeconomie S2 by: #Reda_ecoplus van 9. Download. Samenvatting – Micro en macro economie te kennen leerstof. Vak: +0Mq2HKI3I0XJsjd2b6vNE2t0Br0zfcIWZ6TJLENXgp04LFCZ/s2/ . Macro-economie (BY). Week 2: Homework. II.1 Average propensity to consume. Macroeconomic consumption theory partly developed because the.

Author: | Mozshura Shakajora |

Country: | Burkina Faso |

Language: | English (Spanish) |

Genre: | Medical |

Published (Last): | 21 May 2013 |

Pages: | 453 |

PDF File Size: | 19.45 Mb |

ePub File Size: | 5.59 Mb |

ISBN: | 303-7-34343-547-2 |

Downloads: | 84422 |

Price: | Free* [*Free Regsitration Required] |

Uploader: | Goltigami |

Skip to main content. Log In Sign Up. Indonesian government has made a shift of macro economie pohay from the industrial secta with suphisheated technology to the development of industries based on natural resources.

## Macroéconomie L1 S2 (Cours Complet)

One of the from the point of view of accuracy. We make fold cross main functions of data mining is the classification that is used validation for the evaluation from which we got a to predict the class and generate information based on confusion matrix. The model evaluation method is fold historical data. In the mmacroconomie, there is a lot of cross validation. Evaluation result is a confusion matrix for algorithms that can be used to process the input into the desired output, thus it is very important to observe the assessment of precision, recall, F-measure, macroconomir success performance of each algorithm.

## Protected: Macroéconomie – IEP SGEL

The purpose of this research rate. By using the confusion matrix, the accuracy of the is to analyze and compare the performance i.

The evaluation method used is fold cross validation. Evaluation result is a confusion matrix for This article is organized as follows: In section two, the measuring accuracy in precision, recall, F-measure, and related work for accuracy comparison is covered. Based on the comparative analysis, the decision tree, k-NN, data sets, accuracy measurement, fold amcroconomie tree algorithm gains the accuracy better by variation of validation, WEKA, and confusion matrix are discussed in 2.

In section four, comparative analysis of the results are given. In section five, we present the conclusion Keywords-Classification; Decision Tree; k-NN; fold and the future works. Cross Validation; Confusion Matrix; Accuracy.

The They are, C4. Among them, the classification algorithms are decision tree C4. In the classification, there is a lot of algorithms that can be used to process the input into the desired output, Galathiya, et al. Using the proposed system, Based Algorithms, Neural Network-Based Algorithms, dan accuracy is gained and classification error rate is reduced Rule-Based Algorithms [2]. The model evaluation macdoconomie is C4. The measurements of accuracy are based on correctly classified instances, incorrectly classified instances, and time taken.

K-NN Classifier The divide-and-conquer approach to decision tree The k-nearest neighbor algorithm k-NN is the macrconomie induction, sometimes called top-down induction of often algorithm used for classification, although it can also decision trees, was developed and refined over many years be used for estimation and prediction.

Ross Quinlan of the University a2 Sydney, Australia. The method that has been records in the training set [9]. A series of improvements Figure 2 shows the example in the medical field.

### Listening comprehension – univ-kag

Now suppose there is the concept buys computer, that is, it predicts whether a a new patient record, without a drug classification, and customer at AllElectronics is likely to purchase a would like to classify which drug should be prescribed for computer. Internal nodes are denoted by rectangles, and the patient based on which drug was prescribed for other leaf nodes are denoted by ovals. Some decision tree patients with similar attributes. In k-NN, if k is set to 3, algorithms produce only binary trees where each internal then it can be concluded that A is the nearest neighbor to node branches to exactly two other nodeswhereas others the new patient based on the closest distance Euclidean can produce nonbinary trees [4].

Thus, the drug from patient A should be prescribed for the new patient, as we can see in Fig. Decision tree model [4] The pseudocode of C4. Example, Target Attribute, Attribute Output: Gain the accuracy in the measurements of The pseudocode of k-NN algorithm [2] is as follow.

Select one of the UCI data sets. Gain the confusion matrix. Gain the accuracy in the measurements of end precision, recall, F-measure, and success rate. Table I shows the Training and testing is performed k times. In iteration i, description of the data sets which are used in the partition Di is reserved as the test set, and the remaining experiments.

That is, in the first iteration, subsets D2,: Unlike the holdout and random subsampling methods above, here, Car 1. Ionosphere 35 2 Iris 5 3 In this research, the number of folds is It divides the data into ten segments: Thus, there are 10 iterations in performing training and Table II shows the classes of each data set which is testing.

BreastCancer no-recurrence-events, recurrence-events Weka was developed at the University of Waikato in New Car Zealand, and the name stands for Waikato Environment for unacc, acc, good, v-good Knowledge Analysis. It provides a uniform interface to many different iris-setosa, iris-versicolor, learning algorithms, along with methods for pre- and Iris iris-virginica postprocessing and for evaluating the result of learning schemes on any given data set [11].

Accuracy Measurement – Accuracy Measurement of Decision Tree In this research, WEKA is used to do training, testing, The following steps are carried out to measure the and gaining the confusion matrix. Given m classes, a 3.

Apply the training and testing by performing confusion matrix is an m x m matrix where entry ci,j folds cross validation evaluation method. Obviously, the best 5.

Calculate the accuracy for each class based on solutions will have only zero values outside the diagonal confusion matrix. Given m classes, a confusion matrix is a table of at IV. In this section, the results of the research on 5 data sets are presented in some tables and figures. Table III shows a confusion matrix for height classification. From the Table V and Fig. K-NN performs better precision in data sets which have a large number of is actually in it.

The formulas of accuracy measurements are Decision Tree k-NN given below. The system is built using Visual C The purposes are to observe the confusion matrix, Figure 4.

Comparison of accuracy in recall investigate the table of confusion, and measure the accuracy of each algorithm in precision, recall, F-measure, From the Table VI and Fig. K-NN performs better recall having 1. K-NN performs in data sets which have a large number of instances and better success rate in data sets which have a large number classes. Thus, it can also be used for the purpose of Car 0.

In this research, the success rate of Diabetes 0. In comparative analysis, decision tree is having quite better results Ionosphere 0. It is giving accuracy with a variation of Iris 0. For the further scope, it is necessary to observe and investigate the value of k in k-fold cross validation, and also make a comparison among the model evaluation methods.

Comparison of accuracy in F-measure [2] D. Larose, Discovering Knowledge in Data: K-NN performs [4] I. Kamber, Data Mining Concepts and Techniques. Bramer, Principles of Data Mining: Undergraduate Jacroconomie in BreastCancer Comparison of mqcroconomie in success rate [12] X.

Springer-Verlag, decision tree gives better success rate compared to k-NN, He obtained his S. He obtained his in He macroconomue his M. He got his Ph. His area of interest include: Herman Mawengkang is a professor in operations research at the department of mathematics, the University of Sumatera Utara, Medan, Indonesia.

He macrkconomie produced many papers published in several journals. He supervised many Masters and PhD students.

She obtained Master in Industrial Computing and Ph. Major field of study is macroconomei intelligence. Inwhile obtaining her Master at the Universiti Kebangsaan Malaysia, she worked as a Graduate Research Assistant at the same university.

After completing her Master, she continued her study to Ph. D and still worked as a Graduate Research Assistant under her supervisor. She has published some papers, both in Macroconomif Conference and International Journal. Most of it related to artificial intelligence, such as: Remember me on this computer. Enter the email address you signed up with and we’ll email you a reset link.

Click here to sign up. Help Center Find new research papers in: