Task 1:What is the role of IT management and why is it important. Respond to at least on other classmate about their thoughts.Task 2:Please read below student posts and reply each (2 posts)in 150 words.

Naresh :lassifier Performance

Pros and Cons

Naïve Bayes

Prose. Predicting the class of the test data set is simple and quick. It also succeeds well in

predicting multi-levels. A Naive Bayes algorithm performs better compared too many other

systems, including infrastructural regression while assuming individuality holds, and

customer needs less training data.

Cons. If a numerical component has a classification not identified in the training data set,

otherwise the model will give a possibility of 0 (zero) and would not be capable of making a

forecast. It is often called Zero Frequency. The Optimistic Bayes limitation is the presumption

of independent determinants (Rajalakshmi & Aravindan, 2018, pp. 363-396).

Logistic Regression

Prose. For observations the appropriate probability scores

Across tools the efficient operations accessible

Cons. It does not perform well across too big a feature room. It does not do a very well

significant number of numerical functionality/variables. For nonlinear features, relies on

transformations (Nakamura, et al., 2020, pp. 1-12).

Decision Trees

Prose. Fit for visual display, simple to understand, and to perceive. It is an illustration of a white-box

model that accurately simulates the process of making individual decisions. It can function with

elements categorical and numerical.

Cons. Restrictive tree depth, the only axis-aligned split of data.

KNN

Prose. No expectations regarding data — valuable, for instance, for nonlinear data, General

algorithm — to define and concentrate/understand, Multipurpose — valuable for classification or

deterioration

Cons. Computationally exclusive — for the stores of the algorithm and all of the training data, the

requirement of the high memory, Delicate to inappropriate structures as well as the measure of the

data

Support Vector Machine

Prose. SVM operates reasonably well if there is a substantial margin of category separation. SVM is

fairly efficient with memory. SVM seems to be more efficient in environments of large dimensions.

Cons. For large data sets, the SVM algorithm is also not suitable. The SVM can perform poorly in

situations where the number of functions for each data point higher than the number of model

training data.

Random Forest

Prose. The statistical efficiency can compare with both the best algorithms of supervised learning.

They include a realistic estimate of the importance of functionality.

Cons. An ensemble model is generally less open to interpretation than a major decision tree

Training a huge number of complex forests could have high processing costs and use a great deal of

memory (Wang, Wang, & Ma, 2019, pp. 1-16).

Dimensionality Reduction

Issues with classification in machine learning sometimes require so many variables on both the

basis from which the ultimate identification is made. Such considerations are primarily such

designated features parameters. Therefore higher the level of functions, the simpler the training set

becomes to envision and then work. Most of these characteristics are almost always linked, and

therefore redundant (Zhang & Luo, 2020, pp. 1-9). That is where algorithms for the reduction of

dimensional space come into effect. Minimization of dimensional space is the method of reducing

the number of explanatory variables under examination, by acquiring a collection of key variables. It

could be categorized into the collection of functions and the collection of features (Elhenawy,

Masoud, Glaser, & Rakotonirainy, 2020, pp. 33898-33908).

In this work that people demonstrate that the identification efficiency of close to the grounddimensional structural MRI data is enhanced through the use of dimension reduction approaches

with nothing but a small number of training examples. People tested two various versions of

dimension reduction: ANOVA F-test selection of dimensions, and PCA transition of features. The

user applied popular training algorithms on the decreased datasets utilizing 5-fold cross-validation.

Testing, the configuration of the maximum parameters including performance evaluation of the

algorithms were carried out and use two different technical measurements: precision, as well as the

Operating Characteristic Curve (AUC) of the transmitter (Lin, Mukherjee, & Kannan, 2020, pp. 1-11).

Ravi – Review the pros and cons of the following algorithms:

Naïve Bayes

Pros: It performs well if there should arise an occurrence of all out-information factors contrasted

with numerical variable.

Cons: If all out factor has a classification, which was not seen in preparing informational index, at

that point model will dole out a (zero) likelihood and will be not able to make an expectation.

Logistic Regression

pros: it is progressively powerful: the free factors don’t need to be typically appropriated, or have

equivalent change in each gathering

Cons: it requires significantly more information to accomplish steady, important outcomes. With

standard relapse, and DA, ordinarily 20 information focuses per indicator is viewed as the lower

bound (ArchanaH. & Sachin, 2015).

Decision Trees

Pros: A decision tree does not necessitate standardization of data.

Cons: A trivial modification in the data can source a great alteration in the edifice of the decision tree

instigating unpredictability.

KNN

Pros: K-NN is pretty spontaneous and meek, it has no conventions and No Working out Step

Cons: K-NN slow procedure and desires similar topographies.

Support Vector Machine

Pros: SVM works comparatively well when there is vibrant boundary of separation between classes.

Cons: SVM process is not appropriate for large data sets (ArchanaH. & Sachin, 2015).

Random Forest

Pros: Irregular Forest depends on the sacking calculation and utilizations Ensemble Learning

procedure. It makes the same number of trees on the subset of the information and joins the yield of

the considerable number of trees. Right now, decreases overfitting issue in choice trees and

furthermore diminishes the change and along these lines improves the exactness.

Cons: Complexity: Random Forest makes a ton of trees and joins their yields. As a matter of course,

it makes 100 trees in Python sklearn library. To do as such, this calculation requires significantly

more computational force and assets. Then again, choice tree is straightforward and doesn’t require

such a lot of computational asset (“Advantages and Disadvantages of Quantitative and Qualitative

Information Risk Approaches”, 2011).

Explain how Dimensionality Reduction helps improve the classifier performance

Dimensionality decrease assumes a significant job in grouping execution. An acknowledgment

framework is structured utilizing a limited arrangement of information sources. While the exhibition of

this framework increments on the off chance that we include extra highlights, sooner or later a

further incorporation prompts a presentation debasement. Head Components Analysis are one of

the top dimensionalities decrease calculations, it isn’t difficult to comprehend and utilize it in genuine

undertakings. This procedure, notwithstanding making crafted by highlight control simpler, it despite

everything assists with improving the consequences of the classifier (ArchanaH. & Sachin, 2015).

Purchase answer to see full

attachment

#### Why Choose Us

- 100% non-plagiarized Papers
- 24/7 /365 Service Available
- Affordable Prices
- Any Paper, Urgency, and Subject
- Will complete your papers in 6 hours
- On-time Delivery
- Money-back and Privacy guarantees
- Unlimited Amendments upon request
- Satisfaction guarantee

#### How it Works

- Click on the “Place Order” tab at the top menu or “Order Now” icon at the bottom and a new page will appear with an order form to be filled.
- Fill in your paper’s requirements in the "
**PAPER DETAILS**" section. - Fill in your paper’s academic level, deadline, and the required number of pages from the drop-down menus.
- Click “
**CREATE ACCOUNT & SIGN IN**” to enter your registration details and get an account with us for record-keeping and then, click on “PROCEED TO CHECKOUT” at the bottom of the page. - From there, the payment sections will show, follow the guided payment process and your order will be available for our writing team to work on it.