This task provides you with opportunities to understand and apply predictive analytics techniques in real- world situations (ULO2), as outlined below. By completing this task, you will demonstrate your specialised and integrated knowledge of business analytics (GLO1), your ability to evaluate complex business information to advance critical and analytical thinking and judgement (GLO4), and your ability to solve ill- defined real world problems, by drawing on your analytical skills to interpret and analyse business data to develop solutions (GLO5).
The business context for this assignment is the domestic tourism sector, focusing on providers of tourist accommodation. Organisations such as AirBnB provide a digital platform that tourists can use to rent properties in particular locations around the world. The properties are owned by private individuals (property hosts), and AirBnB takes a commission for bookings via their digital platform.
AirBnB approached you again to develop RapidMiner process(es) capable of analysing and predicting customer feedback about their stay at Melbourne Airbnb rental properties. AirBnB provided you with a sample dataset of approximately 1,000 rental listings. This sample dataset for this assignment is available on the unit website, as part of the assessment resources for this assignment.
The provided dataset includes a variety of numerical, nominal and text attributes, and descriptions of these attributes. Refer to the accompanying data dictionary for details.
AirbnbAI would like you to use RapidMiner to address the following tasks:
Task A: AirBnB is seeking insights into the prices of properties that are more than 10 kilometres from the Melbourne CBD. Refer to the notes accompanying this assignment on how to calculate approximate distances between two points using geolocated data. Develop a process model to determine if a correlation exists for such properties between:
Task B: Develop a predictive model to estimate the review score ratings of all properties located more than 10 kilometres from the Melbourne CBD, using relevant predictor attributes in the data set. Again, refer to the accompanying notes for the assignment on how to calculate approximate distances between two points using geolocated data.
Explain how the predictive model could be used to estimate a review score rating.
Task C: AirBNB becomes concerned when any of the six review score attributes (accuracy, checkin, cleanliness, communication, location and value) drops below 10 for listed properties across the full dataset. For this reason, you’re asked to develop a process model to identify such properties of concern. The model should identify 3 clusters (groupings) within these properties of concern and enable comparisons of the clusters according to the six review score attributes.
Indicate which of the 3 identified property clusters is of most concern. Specify the number of properties in the cluster that is of most concern.