This dataset contains information on 50 restaurant ratings p…
This dataset contains information on 50 restaurant ratings provided by 49 customers. Each row represents an individual customer’s review of a restaurant, and each column corresponds to a specific restaurant. Ratings are recorded on a numerical scale from –3 to 3, where higher values indicate more positive experiences and lower values reflect more negative experiences. 1. Write the formulas for calculating the correlation and the cosine similarity between customer AD (row 1) and customer BW (row 10). You only need to provide the formulas based on the dataset, and no numerical calculations are required. (5 pts) 2. Apply user‐based collaborative filtering to this dataset and based on the results, recommend a restaurant to the first customer AD. (8 pts) 3. Apply item‐based collaborative filtering to this dataset and based on the results, recommend a restaurant to the first customer AD. (7 pts)
Read DetailsThis dataset represents customer transaction records. Each r…
This dataset represents customer transaction records. Each row corresponds to a single transaction, and each column indicates whether a specific item was included in that transaction. The data is presented in binary form: a value of 1 means the item was purchased, while a value of 0 means it was not. 1. Remove the first column (transaction ID) and convert the data to a matrix. Then convert the binary incidence matrix into a transactions database suitable for association rule mining. Create a frequency plot of the items purchased. (5 pts) 2. Generate an association rule to the data to find all the rules that result in Milk. Set the parameter for supp=0.03, conf=0.8, target=“rules”. Name these as rules1. Explain the first rule in rules1 in your own words. (5 pts)
Read Details