ترجمه تخصص کامپیوتر-داده کاوی

---

شرح پروژه

نیاز به ترجمه روان و جمله بندی قابل فهم و درک و تخصص داده کاوی برای مقاله است

ترجمه توضیحی به زبان ساده و روان مورد نیاز است

صفحه اخر و رفرنس نیاز به ترجمه ندارد

تصاویر و جداول نیاز به ترجمه ندارد

ترجمه باید 2 روزه تحویل داده شود

فرمول ها با توضیحات کامل نیاز است

 به عنوان نمونه پاراگراف زیر را ترجمه و ارسال کنید

SYSTEMATIC CONSTRUCTION OF A FOREST My method is another example of improving accuracy through combining the power of multiple classifiers [15]. Here, each classifier is a decision tree, and I call the combined classifier a decision forest. Some previous attempts to construct multiple trees rely on certain heuristic procedures or manual intervention [19], [30], [31]. The number of different trees they can obtain is often severely limited. I emphasize that arbitrarily introduced differences between trees, such as using randomized initial conditions or perturbations in the search procedures [9], do not guarantee good combination performance. Here, good combination performance is defined as 100 percent correct rate on training data and a monotonic decrease in generalization error as the forest grows in the number of trees, provided that the data do not have any intrinsic ambiguity. Also, it is important to have a systematic procedure that can produce a large number of sufficiently different trees. My method relies on an autonomous, pseudorandom procedure to select a small number of dimensions from a given feature space. In each pass, such a selection is made and a subspace is fixed where all points have a constant value (say, zero) in the unselected dimensions. All samples are projected to this subspace, and a decision tree is constructed using the projected training samples. In classification a sample of an unknown class is projected to the same subspace and classified using the corresponding tree. For a given feature space of n dimensions, there are 2n such selections that can be made, and with each selection a decision tree can be constructed. More different trees can be constructed if the subspace changes within the trees, that is, if different feature dimensions are selected at each split. The use of randomization in selecting the dimensions is merely a convenient way to explore the possibilities. The trees constructed in each selected subspace are fully split using all training data. They are, hence, perfectly correct on the training set by construction assuming no intrinsic ambiguities in the samples. When two samples cannot be distinguished by the selected features, there will be ambiguities, but if no decision is forced in such cases, this does not introduce errors. For each tree, the classification is invariant for points that are different from the training points only in the unselected dimensions. Thus, each tree generalizes its classification in a different way. The vast number of subspaces in highdimensional feature spaces provides more choices than needed in practice. Hence, while most other classification methods suffer from the curse of dimensionality, this method can take advantage of high dimensionality. And contrary to the Occam’s Razor, our classifier improves on generalization accuracy as it grows in complexity.

این پروژه شامل 1 فایل مهم است، لطفا قبل از ارسال پیشنهاد حتما نسبت به بررسی این فایل اقدام فرمایید.

مهارت ها و تخصص های مورد نیاز

---

---