The data mining tool DTREG has demostrated slightly better results in terms of accuracy, true negative, false negative, spec- ificity, recall and area under ROC curve. Oral cancer database is considered as it is estimated to be 8th most common cancer worldwide and extremely grave problem in India as well. However, WEKA displays better results in terms of true positive, false positive, precision and f-measure. Mi- lovic et al. The rest of the paper is organized as follows:

dtreg 4.1

Uploader: Taulkis
Date Added: 20 October 2013
File Size: 63.35 Mb
Operating Systems: Windows NT/2000/XP/2003/2003/7/8/10 MacOS 10/X
Downloads: 64610
Price: Free* [*Free Regsitration Required]

Its ultimate goal is prediction. The patients who are predicted as dtrrg malignant among non malignant patients are True Negative TN cases. Stratified cross-validation method is used for confirma- tion of model. Number of layers is 3 input, hidden and output.

Comparing the Performance of Data Mining Tools: WEKA and DTREG

It takes the total absolute error and normalizes it by dividing by the total absolute error of the simple predictor. This network has an input layer on the left with three neurons, one hidden layer in the middle with three neurons and an output layer on the right with three neurons. However, weka displays better results in terms of true positive, false positive, precision and f-measure. Prevention factor dataset was constructed and then three asso- ciation rule mining algorithms: It is a robust measure in comparison to simple percent agreement calculation since kappa coefficient takes into ac- count the agreement occurring by dgreg [28][29].

The results of these dtdeg may possibly offer clinical medicine structure and build the model for medical issues that can provide extraordinarily benefit to healthcare industry In this paper, we attempt to apply data mining to oral can- cer database as it was estimated to be 8th most common can- cer worldwide inwith approximatenew cases anddeaths.


When there is more than one hidden layer, xtreg output from one hidden layer is fed into the next hidden layer and separate weights are applied to the sum going into each layer. In stratified k-fold cross-validation, the folds are selected so that the mean response value is approximately equal in all the folds. The patients who are predicted as malignant among malignant patients are True Positive TP cases.

The study shows that the histology along with a detailed clinical workup is found to be a useful, reliable and accurate diagnostic technique for lesions of the oral cavity.

Translation location for Offline credit card and Expiry year wrong – Support Central

Clas- sification technique is used for analysis, category weights are distributed over entire data file, misclassification costs are equal and variable weights are also equal. The architecture of multi-layer perceptron network and training Statistics of the network is presented in Table [1] and Tables [2] respectively.

They scour dtdeg for hidden patterns finding predictive information that experts may miss because it lies outside their expectations [6]. Oral cancer has the greatest burden main.

dtreg 4.1

Oral malignancy is a heterogeneous assembly of tumors roll- ing out from diverse parts of the oral cavity, with distinctive predisposing factors, prevalence, and treatment outcomes.

In light of this, we construct a data mining model using multilayer perceptron to predict the sur- vivability of oral cancer patients.

dtreg 4.1

The scheme adopted by the tool to build the MLP model is weka. The Artificial Neural Network is one of the most commonly used models based on human cognitive structure. These tools are used to build multilayer perceptron which is a data mining model to predict the survivability of the oral cancer patients.

Screenshot for DTREG 3.0

They found that dteg overall 5-year ob- served and relative survival rates were Data mining is ex- tensively used by companies and public bodies for marketing, detection of fraudulent activity, and scientific research [1].


If the target variable is categorical and a classification tree is build, then a misclassification summary table presents the number of rows with a particular category that were misclassified by the tree, for both training as well as validation dataset.

Table [9] presents the summary of the performance estimation carried out by WEKA. Figure 4 and Figure 5 present the.

dtreg 4.1

Hidden layer and Output layer activation function used in this model is Logistic. Misclassification table for the model is presented in Table [3] for training data and in Table [4] for validation data. It is a frequently used measure of the differences between val- ues predicted dtfeg a model and the values actually observed. ROC is a graph that demon- strates the performance of a binary classifier model.

Section 5 presents the experimental results of both the tools and section 6 compares the performance of the tools. Lift and gain values are especially use- ful when a model is being used to drteg prioritize marketing efforts.

Detailed accuracy of the model by class regarding true positive rate Dtreggfalse positive rate FPprecision, recall, f- measure, receiver operating character area ROC Area and precision recall curve area PRC Area is presented in Table [10] and confusion matrix is presented in Table [11].

The data mining tool DTREG has demostrated slightly better results in terms of accuracy, true negative, false negative, spec- ificity, recall and area under ROC curve.

Efforts to increase the body of literature on the knowledge of the disease etiology and regional distribution of risk factors have begun gaining momentum.