//can any one check my answers and solve the others, i will be thank.

CSBP 411: Machine Learning - Spring 2019

1. Suppose that you are employed as a data mining consultant for an Online Shopping Company. Describe how data mining can help the company by giving specific examples of how learning concepts, such as clustering, classification, and association rules can be applied.

A huge collection of data is present on any on-line shopping companies that have information regarding numerous product belongs to totally different classes. Analyzing them manually will result in wastage of your time. Therefore, in order to save lots of time and improves accuracy, we will use a conception of data mining. Data mining is extracting useful information from large datasets in order to discover models and patterns that are unknown. From given a database of many customers with their corresponding products purchased; we could identify between loyalty customers and normal customers this is one benefit of Datamining. Moreover,it Identifying customer behaviors, improve customer quality service and provide good transportation facilities.

Clustering is a process of grouping similar instance into clusters. In the case of online shopping company, the cluster will be used to group customers according to their reviews in same or different products. It can be used also for a recommendation system (tracking the user clickstreams).

Classification is predicting a discrete class. Using of classification algorithms will let to generate classification rule. For online shopping company it can used to classify the products according to customer reviews and feedback.

Associations are detecting association between features. Association can be used for discovery of great offers and what the customer wants to buy.

2.For each of the following data sets, describe at least 3 main attributes and their values.

e.Movies

Type	Year	Running Time
Action	2012	2Hrs
Comedy	2010	2:30Hrs
Horror	2011	3Hrs

6. Images from Earth-orbiting satellites

Near- Polar Orbits	Size	Metadata Type
Yes	Very big	Geospatial
Yes	Medium	Clime change
No	Small	Astronomical

7. Patients with blood-pressure issues.

Age range	Sex	Blood Pressure mm Hg
35-45	Female	130 ~ 155
50-60	Male	120 ~ 140
55-65	Male	115 ~ 150

8. Building window size and related air conditioning cost

Position	Sliding	Size
Top	Yes	Big
Down	No	Medium
Middle	No	Small

3. Describe how to deal with missing values in a decision tree.

The missing value is treated as an attribute value in its own right. the first solution for this is assigned the instance to the most frequent values. Another solution is to split the instance into pieces and send part of it to branch (most popular branch). Finally, to not classify it (ignore).

4. Describe how to deal with numeric attributes in a decision tree. Is it possible for a numeric attribute to appear more than once is tree branch? Explain your answer.

Predicts a numeric quantity, which called also a regression. By regression tree that each leaf would contain a numeric value that is the average of all the training set values to which the leaf applies, because a numeric quantity is what has predicted decision trees with averaged numeric values at the leaves.

Yes, Numeric attributes can appear more than once, but only with very different numeric conditions also if a test is not at the root of the tree, it can appear in different branches.

5. Explain the reasoning behind using support and confidence when constructing association rules.