Work to be handed in by: 17 May 2023 @2pm

Assignment Specification:

The module is assessed 100% through this coursework assignment.

This coursework aims to assess your knowledge of business intelligence and data mining, and your  ability  to  perform  data  mining  tasks  by  applying  suitable  concepts,  methods  and techniques learned during the lectures and practical sessions for business intelligence.

The coursework is carried out individually.  You are required to produce a report of 20 pages of A4 (+/- 10%), including tables and diagrams but excluding references and appendices,        based on the case as described in this coursework document.

An appendix can be used to include support materials to back up main body points where     necessary. You are also required to submit the supplementary materials of your work using SAS Enterprise Miner on blackboard by the specified deadline.

1.   The Case – Opportunities and Challenges of Sharing Economy: Airbnb and InsideAirbnb

Airbnb - Holiday Lets, Homes, Experiences & Places

Airbnb (Airbnb.co.uk) is an online marketplace for arranging or offering short-term rental/lodging i.e. temporary accommodation, primarily homestays, or tourism

experiences. It was founded in August 2008 by Brain Chesky and friends, and it currently has 6,300 employees as in 2021.

On December 10, 2020, Airbnb went public with a valuation of over $100 billion, making it one of the largest IPOs (Initial Public Offerings) of 2020.  It is reported that Airbnb capital market was  more than the top three  largest  hotel  chains  (Marriott,  Hilton,  and  Intercontinental) combined1 . Though some are calling over evaluation, the company lacks traditional mortgages, employee  fees,  and  maintenance  fees  which  burden  hotels.  Airbnb  hosts  pay  their  own mortgage and clean their apartments, leaving the company much freer of debt, thus making it far more valuable.

Airbnb service overview

Airbnb provides a platform for hosts to accommodate guests with short-term lodging and tourism-related activities. Guest can search for accommodation using filters such as location, price,  specific types  of  home.  Before  booking,  users  must  provide  personal  and  payment information.  Some  hosts  also  require  a  scan  of  government-issues  identification  before accepting a reservation. Hosts provide prices and other details for their rental or listing e.g. number of guests included in the price, type of property, type of room, number of bathrooms, number of bedrooms, number of beds and type of bed, and minimum number of nights for a reservation, and amenities.

In addition, Airbnb provides a guest review system where hosts and guests can leave reviews about  their  experience,  and  rate  each  other  after  a  stay.  However,  the truthfulness  and impartiality  of  reviews  may  be  adversely  affected  by  concerns  of  future  stays  because prospective hosts may refuse to host a user who generally leaves negative reviews. Besides, the company's policy requires users to forego anonymity, which may also detract from users' willingness to leave negative reviews .

Criticism of Airbnb

Airbnb has attracted criticism for increasing housing/residential rental prices in cities where it operates and creating nuisances and security issues etc for those living near leased properties and has negatively affects the quality of life in residential areas, and housing crisis around city in the UK, USA and Europe. The company has attracted regulatory attention from cities such as San Francisco, New York City, and the European Union over the past number years. It has also faced challenges from the hotel industry and other, similar companies.

Airbnb has made a quarter (25%) of its global workforce redundant in 2020 due to the global pandemic2 . But the news was welcome by some campaigners who were fighting for soaring rents in cities with large number of Airbnb hosts. The number of longer-term rental properties (i.e. residential as opposed to short-term/holiday lets)  in  central  Dublin was  up  71%  on comparable period last year, as landlords abandoned short-term lets through Airbnb3 .

 Inside Airbnb – Adding Data to the Debate (insideairbnb.com)

Inside Airbnb (insideairbnb.com) is an independent, non-commercial set of tools and data that allows individual to explore how Airbnb is really used in cities around the world. It was set up by Murray Cox and John Morries in 2016.

Airbnb claims to be part of the “sharing economy” and disrupting the

hotel industry by offering short-term rental/lodging. However, data shows that most Airbnb listings in most cities are entire homes, many of which are rented all year round (i.e. illegal short-term rentals)– disrupting housing and communities.

Most recently, New York City’s plans to crack down on illegal short-term rentals which could remove as many as 10,000 Airbnb listing is sparking fierce debates about housing, hotels, the tourist market and residents’ rights4 .

By  analysing  publicly  available  information  about  a  city’s  Airbnb’s  listings,  Inside  Airbnb provides filters and key metrics so users can see how Airbnb is being used to compete with the residential housing market. With Inside Airbnb, user can ask fundamental questions about Ainbnb in any neighbourhood, or across the city as a whole, such as:

•    how many listings are in my neighbourhood and where are they?

•    how many houses and apartments are being rented out frequently to tourists and not to long-term residents?

•    how  much  are  hosts  making from  renting to tourists  (compare that to  long-term residential rentals)?

•    which host are likely running a business with a multiple listings and where are they?

These questions (and the answers) get to the core of the debate for many cities around the world, with Airbnb claiming that their hosts only occasionally rent the homes in which they live. In addition, many cities or state legislation or ordinances that address residential housing, short term or vacation rentals, and zoning usually make reference to allowed use, including:

•    how many nights a dwelling is rented per year

•    minimum nights stay

•    whether the host is present

•    how many rooms are being rented in a building

•    the number of occupants allowed in a rental

•    whether the listing is licensed

The  Inside  Airbnb  tool  or  data  can  be  used  to  answer  some  of  these  questions.  Some understanding of how the Airbnb platform is being used will help clear up the laws as they change.

Further information of Airbnb, please visit:https://www.airbnb.co.uk/

Further information of Inside Airbnb, please visit:http://insideairbnb.com/index.html

2.   Coursework requirements:

The sharing economy has brought opportunities and challenges to homeowners, society,        residents, communities and governments.   One of the biggest issues with Airbnb is whether hosts are renting out residential properties permanently as hotels (i.e. illegal short-term      rentals) or sharing the primary residence in which they live "occasionally" (i.e. legal short-  term rentals).

Airbnb could easily answer this question but instead it is up to us to shape our communities and solve our urgent need to house tourists, housing shortage/crisis, and to address the

nuisances, security and safety issues etc for those living near leased properties by Airbnb.

In this assignment, you are required to carry out data mining tasks using data of Airbnb          listings of Edinburgh, Scotland, UK from InsideAirbnb, and to report your findings as a result your data mining/analysis to address the challenges and issues of the sharing economy of      Airbnb.

2.1 DATA

The Edinburgh data (16 December 2022) is available to download from InsideAirbnb and on Blackboard as shown in Figure 1 below.

http://insideairbnb.com/get-the-data

Figure 1: Edinburgh Data compiled by InisideAirbnb as on 16 December 2022 (-

http://insideairbnb.com/get-the-data

As shown in Figure 1, the data set includes:

1)   Listings.csv.gz contains detailed listing data of Edinburgh. The data was compiled

on 16 December 2022. Each row of the data represents a single listing and             contains information about the host of the property, the property’s                         characteristics and overall rating of the property and its features by guests. There are 7,390 listings and 67 variables in the data set. Listing can be deleted in the      Airbnb platform. The data presented is a snapshot of listings available at a             particular of time as on and up to 16 December 2022.

2)   Reviews.csv.gz contains the detailed reviews data for each listing. The data was used for a number of derived variables in the detailed listing data e.g.                    number_of_reviews, number_of_review_ltm, first_review, last_review, and         reviews_per_months.

3)   Calender.csv.gz contains detailed calendar data i.e. the availability calendar for

365 days in the future for each listing. In addition

4)   A data dictionary – can be viewed and downloaded from

https://docs.google.com/spreadsheets/d/1iWCNJcSutYqpULSQHlNyGInUvHg2Bo UGoNRIGa6Szc4/edit#gid=1322284596

2.2. Your tasks

You are required to use the detailed listing data (listings.csv) to find meaningful pattens and rules of whether hosts in Edinburgh are renting out residential properties as hotels/business (illegal short-term rentals) or genuine sharing the primary residence in which the live               “occasionally” (legal short-term rentals).

You are expected to conduct cluster analysis (unsupervised learning) to differentiate

hosts/listings that are likely to be genuine short term let (legal short-term rentals) versus         hosts/listings that are likely to be operating as a business (illegal short-term rentals) based on information about the host of the property, the property’s characteristics and overall rating    of the property and its features by guests etc,  and to build a classification model that could    differentiate hosts and/or listings that are for genuine (occasionally) short-term or vice versa. You are also expected to conduct further literature search on the issues and data exploration, data preparation for your clustering and classification tasks.

You may want to refer to the calendar (calendar.csv) and review data (reviews.csv) and derive further new variables where necessary.

2.2 What to deliver

You are required to produce a report of 20 pages of A4 (+/- 10%) including tables and              diagrams but excluding references and appendices. An appendix can be used to include           support materials to back up main body points where necessary. You are also required to       submit the supplementary materials of your work using SAS Enterprise Miner on blackboard by the specified deadline.

The total of 100 marks will be allocated to the following aspects of the report, which should also be used as a guideline to structure the report.

1.    Introduction (10%)

The introduction section should include the background and context, problems and issues of   sharing economy and Airbnb, and a clear statement of the data mining goal. You are expected to justify your statements using relevant literature and sources.

2.    Data understanding (10%)

In this section, you are expected to conduct exploratory data analysis e.g. summaries statistics and data visualisation techniques, using suitable and relevant techniques and methods and report  your  key  findings,  including  variables  and  measurement  identified  for  your  data preparation tasks.

3.    Data Preparation (20%)

In this section, you are expected to take the data identified in the previous step and prepare them for the model building. This should include:

a)   data cleaning e.g. missing data handling

b)   data transformation e.g. creating new derive variables

You are expected to create two new derive variables “Occupancy rate” – estimating how often an Airbnb listing is being rent out, and “listing income” i.e. approximate a listing’s income.  You are expected to justify the approaches taken backup by relevant sources or literature, and