10.07.2015 Views

Web Mining and Social Networking: Techniques and ... - tud.ttu.ee

Web Mining and Social Networking: Techniques and ... - tud.ttu.ee

Web Mining and Social Networking: Techniques and ... - tud.ttu.ee

SHOW MORE
SHOW LESS
  • No tags were found...

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

50 3 Algorithms <strong>and</strong> <strong>Techniques</strong>Naive Bayesian classi f ier is one basic type of the Bayesian learners with an assumptionthat all the attributes are conditionally independent given the value of the class C. By independencewe mean probabilistic independence, that is, A is independent of B given C wheneverPr(A|B,C) = Pr(A|C) for all possible values of A, B <strong>and</strong> C, whenever Pr(C)>0.The pseudo code of Naive Bayesian classifier is shown in Algorithm 3.5 [182].Algorithm 3.5: Naive Bayesian learning algorithmNaive Bayes Learn(samples)For each target value v j doPr(v ˆ j ) ← estimate Pr(v ˆ j );For each attribute value a i of each attribute a doPr(a ˆ i |v j ) ← estimate Pr(a ˆ i |v j );endendClassify New Instance(x))v NB = argmax v j ∈V ˆ Pr(v j )∏ ai ∈x ˆ Pr(a i |v j );3.2.4 Neural Networks ClassifierNeural network (NN) classifier, or arti f icial neural network (ANN) classifier, is one of theclassic methods used for supervising learning [182, 33]. A Neural network can be commonlyrepresented as a format of graph, which consists of densely interconnected elements. Th<strong>ee</strong>lements are called neurons [1].X1X2W2W1X3W3WiXiYXnWnFig. 3.13. An example neural networkFigure 3.13 illustrates one example of neural network. The input is a number of valuesX 1 ,X 2 ,...,X n at the left side <strong>and</strong> the output is Y at the right side, both of which are in thecontinuous space (i.e., commonly is betw<strong>ee</strong>n 0 <strong>and</strong> 1). The neuron in the middle of the networkfirst counts the weighted sum of the inputs, then adjusts by subtracting some threshold θ,finally transfers the result to a non-linear function f (e.g., sigmoid) to compute <strong>and</strong> output. Insummary, the process can be modeled as follows.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!