Part 2. Finding the relationships hidden in the customer's online shopping data (10 points total) Visualizing correlations between customer's preferences on cameras and printers (2 points) Now that you understand a bit about the online shopping preference data, you want to check whether customer's preferences on buying cameras is correlated with their preferences on buying printers. *Make a joint plot to visualize the relationship of printer interest vs camera interest. * (camera on the x-axis and printer on the y-axis) Your plot should 1. include axis labels 2. use the seaborn plot style (you need to import the package and then "set" the style) 3. set the y-axis range between -5 and 50 # Put your code here Removing outliers to visualize detailed patterns of the correlations between camera and printer (3 points) As you can see in the joint plot, there is a cluster of data points with some pattern. But there are also some outliers with very high preferences on printers, which make it difficult to check the details within the cluster of data points. 1. Come up with your own criteria of classifying a data point as an outlier or not; 2. Use mask to filter out outliers; 3. Make a joint plot again on the subset of data, after removing outliers. Describe your criteria about outliers here: # Put your code here to remove outliers using masking: # Put your code here to make the joint plot: Visualize the relationship between customer's preferences on mathematics_books and coffee_maker (1 point) Make a joint plot to visualize the relationship between customer's preferences on mathematics_book and coffee_maker. (mathematics_book on the x- axis, coffee_maker on the y-axis) Your plot should 1. include axis labels 2. use the seaborn plot style # Put your code here: Question: based on the joint plot, do you observe any pattern/relationship between preferences on mathematics_books and coffee_makers? Your answer here: Conditioned on laptop preference, visualize correlations between mathematics_book and coffee_maker for subsets of customers (4 points) In data analysis, sometimes the pattern may not be immediately clear to see, such as the relationship between customer's preferences on mathematics_book and coffee_maker. In some cases, the power of big-data is to allow us to incorporate more entities into consideration, which can be helpful to delineate the subtle patterns hidden in the dataset Here, we want to have a closer look at the relationship between customer's preferences on mathematics_book vs. coffee_maker. But this time, we also include customer's preferences on laptops into our analysis. 1. Classify customers into 2 groups based on their preferences on laptops. Group 1: the customer's preference on laptop is higher than the mean preference on laptops. Group 2: all other customers. (Hint: use masking) 2. Make a joint plot to visualize the relationship between customer's preferences on mathematics_book and coffee_maker, only for customers in Group 1. 3. Make another joint plot to visualize the relationship between customer's preferences on mathematics_book and coffee_maker, only for customers in Group 2 4. Explain your observations and provide an interpretation on the observations, 5. If you are the seller, what is your strategy of sending coffee maker advertisements, based on customer's preference profiles? # Put your code here