# US 2021/0279648 A1 · Database systems for reservation of inventory to substitute for host cancelation

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0279648 A1
- Document ID: US20210279648A1
- Status: Published application
- Kind: A1
- Application number: 16/812,083
- Filed: 2020-03-06
- Published: 2021-09-09
- Earliest claimed: 2020-03-06
- Assignees: Airbnb, Inc.
- Applicants: Airbnb, Inc.
- Inventors: Navin Sivanandam; Christian Aurelio Goldammer; Michael Andrew Hinckley
- Web page: https://patentgrep.com/patent/US20210279648A1

## CPC

- G06Q 10/02: Reservations, e.g. for tickets, services or events
- G06F 16/951: Indexing; Web crawling techniques
- G06F 16/9537: Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N 20/00: Machine learning
- G06Q 10/0285: {for lodging}
- G06Q 30/0205: {based on location or geographical consideration}
- G06Q 50/14: Travel agencies
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A system and a method are disclosed for replacing a canceled accommodation for a guest user using reserved accommodations. In an embodiment, an accommodation management system retrieves information describing historical bookings and listings for accommodations in a region. The system determines common attributes of the listings and a historical cancelation rate of bookings. The system determines an amount of listings to reserve for a time period based on the historical cancelation rate and moves, from an available listing database to a reserved listing database, the amount of listings for the region, each listing including common attributes. The system sends an indication to each host user of the moved listings that the listing has become unavailable for the time period. The system receives an indication that a booking for a guest user has been canceled and books one of the listings in the reserved listing database for the guest user.

## Claims

### Claim 1

A method for replacing a canceled accommodation for a guest user using reserved accommodations, the method comprising:
retrieving, by a homestay application, information describing historical bookings and listings for accommodations in a region;
determining one or more common attributes of the listings in the region;
determining, based on the historical bookings, a historical cancelation rate of bookings for the region;
determining an amount of listings to reserve for a time period based at least in part on the historical cancelation rate;
moving, from an available listing database to a reserved listing database, the amount of listings for the region, each moved listing including one or more of the common attributes;
sending, to each host user of the moved listings, an indication that the listing has become unavailable for the time period;
receiving an indication that a booking by a guest user has been canceled by a host user, the canceled booking specifying the region and a subset of the time period; and
booking one of the listings in the reserved listing database for the guest user for the time period by populating a booking data structure with a connection between the guest user and the one of the listings in the reserved listing database.

### Claim 2 (depends on 1)

The method of claim 1, wherein determining the amount of listings to reserve for the time period based at least in part on the historical cancelation rate comprises:
comparing the historical cancelation rate to a threshold value;
responsive to the historical cancelation rate exceeding or equaling the threshold value, determining the amount of listings to reserve based at least in part on the historical cancelation rate and an amount of available listings in the region during the time period; and
responsive to the historical cancelation rate being less than the threshold value, determining the amount of listings to reserve to be a baseline value.

### Claim 3 (depends on 1)

The method of claim 1, further comprising:
inputting, to a machine learning model, available listings for a region over a time period, each available listing associated with a set of attributes;
receiving, from the machine learning model, a percentage match for each of the available listings indicating a likelihood of matching an ideal listing for the region;
ranking each of the available listings based on the percentage match; and
moving, from the available listing database to the reserved listing database, a top percentage of the ranked available listings equaling the amount of listings for the region.

### Claim 4 (depends on 3)

The method of claim 3, further comprising:
retrieving, from a booking database, historical bookings for the region;
determining similar attributes for each of the historical bookings;
labelling each historical booking with a booking type and a cancellation indication;
training the machine learning model for the region using the labelled historical bookings.

### Claim 5 (depends on 4)

The method of claim 4, wherein determining similar attributes for each of the historical bookings comprises:
retrieving, for each historical booking, attributes for the listing associated with the historical booking;
mapping, each attribute to a category in an equivalence table; and
determining the similar attributes based on other attributes in the categories of the mapped attributes in the equivalence table.

### Claim 6 (depends on 5)

The method of claim 5, further comprising:
retrieving attributes for each available listing in the region;
categorizing each available listing of the region based on the attributes of each available listing, each category associated with a set of similar attributes;
determining a percentage of historical host user cancelations for each category;
determining, for each category, a subset of the available listings to reserve based at least in part on the amount of listings to reserve and the percentage for the category;
moving the subsets of available listings to reserve from the available listing database to the reserved listing database.

### Claim 7 (depends on 1)

The method of claim 1, wherein booking one of the listings of the reserved listing database for the guest user for the time period comprises:
retrieving user profile data for the guest user;
ranking reserved listings from the reserved listing database for the guest user based on attributes of the reserved listings; and
booking the highest ranked listing for the guest user by populating the booking data structure with a connection between the guest user and the highest ranked listing.

### Claim 8 (depends on 7)

The method of claim 7, wherein ranking reserved listings from the reserved listing database for the guest user based on attributes of the reserved listings comprises:
determining, based on the user profile data, a set of prioritized attributes for the guest user;
retrieving attributes of the reserved listings; and
ranking the reserved listings based on inclusion of the prioritized attributes in the attributes for each reserved listing.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
sending, for display to the guest user, two or more of the listings in the reserved listing database;
receiving a selection of one of the two or more of the listings in the reserved listing database as input from the guest user; and
sending a command to book the selection for the guest user during the time period.

### Claim 10 (depends on 1)

The method of claim 1, further comprising:
sending, for display to the guest user, two or more of the listings in the reserved listing database;
receiving an indication that the guest user rejected each of the two or more of the listings in the reserved listing database; and
sending, for display to the guest user, two or more other listings in the reserved listing database.

### Claim 11 (depends on 1)

The method of claim 1, further comprising:
sending, for display to the guest user, two or more of the reserved accommodations;
receiving a selection of one of the reserved accommodations as input from the guest user; and
booking the selection for the guest user.

### Claim 12

A system for replacing a canceled accommodation for a guest user using reserved accommodations, the system comprising one or more processors configured execute instructions that cause the processor to perform operations of:
retrieving, by a homestay application, information describing historical bookings and listings for accommodations in a region;
determining one or more common attributes of the listings in the region;
determining, based on the historical bookings, a historical cancelation rate of bookings for the region;
determining an amount of listings to reserve for a time period based at least in part on the historical cancelation rate;
moving, from an available listing database to a reserved listing database, the amount of listings for the region, each moved listing including one or more of the common attributes;
sending, to each host user of the moved listings, an indication that the listing has become unavailable for the time period;
receiving an indication that a booking by a guest user has been canceled by a host user, the canceled booking specifying the region and a subset of the time period; and
booking one of the listings in the reserved listing database for the guest user for the time period by populating a booking data structure with a connection between the guest user and the one of the listings in the reserved listing database.

### Claim 13 (depends on 12)

The system of claim 12, wherein the operation of determining the amount of listings to reserve for the time period based at least in part on the historical cancelation rate comprises:
comparing the historical cancelation rate to a threshold value;
responsive to the historical cancelation rate exceeding or equaling the threshold value, determining the amount of listings to reserve based at least in part on the historical cancelation rate and an amount of available listings in the region during the time period; and
responsive to the historical cancelation rate being less than the threshold value, determining the amount of listings to reserve to be a baseline value.

### Claim 14 (depends on 12)

The system of claim 12, wherein the operations further comprise:
inputting, to a machine learning model, available listings for a region over a time period, each available listing associated with a set of attributes;
receiving, from the machine learning model, a percentage match for each of the available listings indicating a likelihood of matching an ideal listing for the region;
ranking each of the available listings based on the percentage match; and
moving, from the available listing database to the reserved listing database, a top percentage of the ranked available listings equaling the amount of listings for the region.

### Claim 15 (depends on 12)

The system of claim 12, wherein the operation of booking one of the listings of the reserved listing database for the guest user for the time period comprises:
retrieving user profile data for the guest user;
ranking reserved listings from the reserved listing database for the guest user based on attributes of the reserved listings; and
booking the highest ranked listing for the guest user by populating the booking data structure with a connection between the guest user and the highest ranked listing.

### Claim 16 (depends on 12)

The system of claim 12, wherein the operations further comprise:
sending, for display to the guest user, two or more of the listings in the reserved listing database;
receiving a selection of one of the two or more of the listings in the reserved listing database as input from the guest user; and
sending a command to book the selection for the guest user during the time period.

### Claim 17

A computer readable medium configured to store instructions, the instructions when executed by a processor cause the processor to:
retrieve, by a homestay application, information describing historical bookings and listings for accommodations in a region;
determine one or more common attributes of the listings in the region;
determine, based on the historical bookings, a historical cancelation rate of bookings for the region;
determine an amount of listings to reserve for a time period based at least in part on the historical cancelation rate;
move, from an available listing database to a reserved listing database, the amount of listings for the region, each moved listing including one or more of the common attributes;
send, to each host user of the moved listings, an indication that the listing has become unavailable for the time period;
receive an indication that a booking by a guest user has been canceled by a host user, the canceled booking specifying the region and a subset of the time period; and
book one of the listings in the reserved listing database for the guest user for the time period by populating a booking data structure with a connection between the guest user and the one of the listings in the reserved listing database.

### Claim 18 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the instructions to determine the amount of listings to reserve for the time period based at least in part on the historical cancelation rate comprise instructions to:
compare the historical cancelation rate to a threshold value;
responsive to the historical cancelation rate exceeding or equaling the threshold value, determine the amount of listings to reserve based at least in part on the historical cancelation rate and an amount of available listings in the region during the time period; and
responsive to the historical cancelation rate being less than the threshold value, determine the amount of listings to reserve to be a baseline value.

### Claim 19 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the instructions further comprise instructions to:
input, to a machine learning model, available listings for a region over a time period, each available listing associated with a set of attributes;
receive, from the machine learning model, a percentage match for each of the available listings indicating a likelihood of matching an ideal listing for the region;
rank each of the available listings based on the percentage match; and
move, from the available listing database to the reserved listing database, a top percentage of the ranked available listings equaling the amount of listings for the region.

### Claim 20 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the instructions to book one of the listings of the reserved listing database for the guest user for the time period comprise instructions to:
retrieve user profile data for the guest user;
rank reserved listings from the reserved listing database for the guest user based on attributes of the reserved listings; and
book the highest ranked listing for the guest user by populating the booking data structure with a connection between the guest user and the highest ranked listing.
