Select one of the datasets from UCI Machine Learning Repositories (http://archive.ics.uci.edu/ml/) OR ( https://www.kaggle.com/datasets )  OR use your own dataset if available.

1.5 Marks

Learning Outcome(s):1 Explain different data mining tasks, problems and the algorithms most appropriate for addressing them

Question One

Select one of the datasets from UCI Machine Learning Repositories

(http://archive.ics.uci.edu/ml/) OR ( https://www.kaggle.com/datasets )

OR use your own dataset if available.

 

Give a brief description about the topic of your dataset.

 

Answer:

The dataset topic is reviews for Hotel Worldwide (Booking) and briefly is talking about

review hotel rating and to see what guests thought of their stay overall. Moreover, look individual reviews to get a sense of what people liked and didn’t like about specific hotels. You can also use the review publication date and stay date columns to track changes in travelers’ opinions over time.

1 Marks

Learning Outcome(s): 2 Apply and evaluate data mining algorithms with respect to problems they are specifically designed for.

Question Two

The dataset may follow the following requirements (Data description)

• Number of instances: between 300-500
• Number of attributes: between 10 to 15

 

Describe some of the attributes of your data (meaning, type, range, etc.)

 

 

 

 

 

 

 

 

 

 

 

 

1 Marks

Learning Outcome(s):1 Explain different data mining tasks, problems and the algorithms most appropriate for addressing them.

Question Three

Prepare a CSV OR ARFF format data file of the data (screenshots are required).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1 Marks

Learning Outcome(s):1 Explain different data mining tasks, problems and the algorithms most appropriate for addressing them.

Question Four

Load the dataset in Weka (Screenshots are required).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1 Marks

Learning Outcome(s): 2 Apply and evaluate data mining algorithms with respect to problems they are specifically designed for.

Question Five

Describe your data using the statistical information given by Weka.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1 Marks

Learning Outcome(s): 2 Apply and evaluate data mining algorithms with respect to problems they are specifically designed for.

Question Six

In the preprocessing step, remove some of the values from an attribute and give us a solution to handle this problem using Weka.

 

 

 

 

 

 

 

 

 

 

 

 

 

2 Marks

Learning Outcome(s): 2 Apply and evaluate data mining algorithms with respect to problems they are specifically designed for.

Question Seven

Do a basic preprocessing to the dataset such Data Cleaning, Data reduction and Normalization etc.

2 Marks

Learning Outcome(s): LO3 Apply a wide range of clustering, estimation, prediction, and classification algorithms.   LO4 Carry out recent data mining techniques and applications.

Question Eight

Based on dataset run Apriori algorithm with different support and confidence values. Discuss the generated rules.

 

 

1 Marks

Learning Outcome(s): LO3 Apply a wide range of clustering, estimation, prediction, and classification algorithms.   LO4 Carry out recent data mining techniques and applications.

Question Nine

Based on your dataset selection, apply K- Nearest neighbor data mining algorithm. (take k = 5).

 

 

 

 

 

 

 

 

 

 

 

1.5 Marks

Learning Outcome(s): LO3 Apply a wide range of clustering, estimation, prediction, and classification algorithms.   LO4 Carry out recent data mining techniques and applications.

Question Ten

Based on your selection dataset, Apply the Decision tree data mining algorithm and record the accuracies (Provide a screenshot for the provided tree and a brief description for it).

 

 

 

 

 

 

 

 

 

 

 

 

1 Marks

Learning Outcome(s): LO3 Apply a wide range of clustering, estimation, prediction, and classification algorithms.   LO4 Carry out recent data mining techniques and applications.

Question Eleven

Apply the K-mean algorithm on the dataset (choose an arbitrary value for K) and study the formed clusters.