Discussion 11: Recommender Systems

Forhad Akbar

11/05/2019

Problem Statement

Perform a Scenario Design analysis. Consider whether it makes sense for your selected recommender system to perform scenario design twice, once for the organization (e.g. Amazon.com) and once for the organization’s customers.
Attempt to reverse engineer what you can about the site, from the site interface and any available information that you can find on the Internet or elsewhere.
Include specific recommendations about how to improve the site’s recommendation capabilities going forward.

Scenario Design analysis

From discussion of week 09:

  1. Who are your target users?

The Travelers: This market segment involves mostly those who enjoy traveling and at the same time not spending all of their money in a hotel room. They would rather spend their in visiting touristic places while they travel because they will not be spending much time where they will be staying. There are some travelers who rather stay somewhere they can relax and spend most of their days without emptying their pockets. The Hosts:These include owners or renters who are willing to rent out their places. The reasons vary as well. They might want to make some money out of an unoccupied space or simply they just want to meet interest people. Whatever the reason is, all hosts are looking to list their current place on Airbnb because they have the ability to get to know who will be staying in their property.

  1. What are their key goals?

The key goals is to serve both travelers and the host. They help to find a place for travelers and help host to find a traveler that has interest at host place. They also provide prior review of the host to make it easy for travelers to decide. Air Anyone interested in traveling looking for spacious, comfortable accommodations that are also affordable and functional. Also, anyone has extra space for rent or host guest for extra income.

  1. How to accomplish those goals?

Here is a great site map of Airbnb website. We can see how Airbnb connect guest and host being acting as a media between them. They also provide lots of experiential stories to help people to navigate and get to know Airbnb. Airbnb provides a simple technological infrastructure, which hosts, and guests can use effortlessly to book and promote their own homes. By providing free access to book and list properties, Airbnb quickly generated a dedicated following of users and hosts and swiftly overcame the initial entry and mobilization barrier. The company also maximized its transaction by letting customers browse freely and with at no charge through all listings until reservations were secured. Once the host and guest Airbnb accept the transaction charges the guest an additional transaction fee of 6-12%, while hosts are charged a 3% transaction fee.

Reverse Engineering

Airbnb’s marketplace contains millions of diverse listings which potential guests explore through search results generated from a sophisticated Machine Learning model that uses more than hundred signals to decide how to rank a particular listing on the search page. Once a guest views a home they can continue their search by either returning to the results or by browsing the Similar Listing Carousel, where listing recommendations related to the current listing are shown. Together, Search Ranking and Similar Listings drive 99% of Airbnb’s booking conversions.
Airbnb uses Listing Embedding technique that they developed and real time Personalization in Search Ranking. The embeddings are vector representations of Airbnb homes learned from search sessions that allow to measure similarities between listings. They effectively encode many listing features, such as location, price, listing type, architecture and listing style, all using only 32 float numbers.

Recommendations

It’s clear that Airbnb’s recommendation engine has been tremendously successful in delivering customized travel experiences to millions of users. As data on guests becomes increasingly available, it’s worth pondering just how personalized these recommendations can become. For instance, would a guests appreciate it if Airbnb recommended a property based on its proximity to a particular store that she shops at? Or, its proximity to a location she has visited in the past? In other words, just how customized is too customized for the comfort of Airbnb guests? Users appreciate the personalization; we have come to expect customized offerings, but how will Airbnb (and similar recommendation search engines) find the balance between customization and privacy?

Airbnb should add interesting attraction nearby the location of host while recommanding using somekind of geolocation technique rather only depends on host input.

Inaddition, Airbnb can add a option of advanced search beside only haiving location, check-in, check-out date and number of guest. In advanced search they can add option like close to public transportation, close to spefic location like Airport or city. This advanced option should help recommandation system to be more precise.