Database systems for similar accommodation determination

1. A method for replacing a canceled listing for a guest user, the method comprising:

receiving an indication that a booking by the guest user for a booking accommodation at a home has been canceled by a host user, the booking associated with the canceled listing and a time period;

retrieving, from an attribute database, attributes of the canceled listing, at least one of the attributes reflecting a physical object of the home;

mapping each attribute of the canceled listing to a set of similar attributes in an equivalence table, the physical object mapped to another physical object;

retrieving, from a listing database, a set of comparable listings, each comparable listing including one or more of the mapped similar attributes and being available for booking during the time period; and

booking one of the set of comparable listings for the guest user during the time period by:

inputting, to a machine learning model, the canceled listing and a set of candidate listings, each candidate listing available for booking during the time period, wherein the machine learning model is trained by:

retrieving, from a database, historical user bookings, each historical user booking associated with a set of attributes;

labelling similar historical user bookings as corresponding to one or more booking types based on the set of attributes associated with the historical user booking; and

training the machine learning model using the labelled historical user bookings;

selecting a candidate listing to be the one of the set of comparable listings based on output of the machine learning model; and

populating to a booking data structure with a connection between the one of the set of comparable listings and the guest user, the booked comparable listing being for another home having the another physical object.

2. The method of claim 1 , wherein mapping each attribute of the canceled listing to a set of similar attributes in the equivalence table comprises:

generating a plurality of sets of attributes by, for each attribute of the canceled listing:

mapping the attribute to a category in the equivalence table;

retrieving a set of attributes associated with the category; and

determining similar attributes based on attributes that occur in a threshold number of the sets of attributes of the plurality of sets of attributes.

3. The method of claim 1 , further comprising:

receiving, from the machine learning model, a percentage match between the canceled listing and each of the set of candidate listings; and

sending, for display to the guest user, a one or more of the set of candidate listings with a percentage match greater than a threshold value.

4. The method of claim 1 , wherein retrieving, from the listing database, the set of comparable listings, comprises:

retrieving user profile data for the guest user;

determining a ranked set of prioritized attributes of the guest user based on the user profile data; and

determining the set of comparable listings based on the prioritized attributes of the guest user and the mapped similar attributes.

5. The method of claim 1 , wherein retrieving, from the listing database, the set of comparable listings, comprises:

determining a set of upgrade attributes based on the mapped similar attributes, the upgrade attributes describing listings of higher quality than the canceled listing; and

determining the set of comparable listings based on the mapped similar attribute, each comparable listing including one or more of the upgrade attributes.

6. The method of claim 1 , wherein booking one of the set of comparable listings for the guest user during the time period comprises:

in response to determining that the host user canceled the booking within a threshold amount of time from a start of the time period, booking one of the set of comparable listings for the guest user by populating the data structure with a connection indicating the booking; and

in response to determining that the host user canceled the booking more than the threshold amount of time from the start of the time period:

reserving one of the set of comparable listings for the guest user by populating the data structure with a connection indicating the reservation; and

sending, for display to the guest user, an indication that the booking was canceled.

7. The method of claim 1 , wherein booking one of the set of comparable listings for the guest user during the time period by populating to the booking data structure with the connection between the one of the set of comparable listings and the guest user comprises:

removing a connection in the data structure between the canceled listing and the guest user;

adding a connection to the data structure between the one of the set of comparable listings and the guest user; and

sending, for display to the guest user, an indication that the canceled listing was replaced with the one of the set of comparable listings.

8. The method of claim 1 , further comprising:

ranking each of the set of comparable listings based on similar attributes and availability for booking during the time period;

selecting a highest ranked comparable listing;

booking the highest ranked comparable listing for the guest user during the time period by populating the data structure with a connection between the highest ranked comparable listing and the guest user; and

sending, for display to the guest user, an indication that the canceled listing was replaced with the highest ranked comparable listing.

9. The method of claim 1 , further comprising:

sending, for display to the guest user, two or more of the set of comparable listings;

receiving a selection of one of the two or more of the set of comparable listings as input from the guest user; and

booking the selected listing for the guest user during the time period by populating the data structure with a connection between the selected listing and the guest user.

10. The method of claim 1 , further comprising:

sending, for display to the guest user, two or more of the set of comparable listings;

receiving an indication that the guest user rejected each of the two or more of the set of comparable listings; and

sending, for display to the guest user, two or more other comparable listings from the set of comparable listings.

11. A system for replacing a canceled listing for a guest user, system comprising one or more processors configured execute instructions that cause the processor to perform operations of:

receiving an indication that a booking by the guest user for a booking accommodation at a home has been canceled by a host user, the booking associated with the canceled listing and a time period;

retrieving, from an attribute database, attributes of the canceled listing, at least one of the attributes reflecting a physical object of the home;

mapping each attribute of the canceled listing to a set of similar attributes in an equivalence table, the physical object mapped to another physical object;

retrieving, from a listing database, a set of comparable listings, each comparable listing including one or more of the mapped similar attributes and being available for booking during the time period; and

booking one of the set of comparable listings for the guest user during the time period by:

inputting, to a machine learning model, the canceled listing and a set of candidate listings, each candidate listing available for booking during the time period, wherein the machine learning model is trained by:

retrieving, from a database, historical user bookings, each historical user booking associated with a set of attributes;

labelling similar historical user bookings as corresponding to one or more booking types based on the set of attributes associated with the historical user booking; and

training the machine learning model using the labelled historical user bookings;

selecting a candidate listing to be the one of the set of comparable listings based on output of the machine learning model; and

populating to a booking data structure with a connection between the one of the set of comparable listings and the guest user, the booked comparable listing being for another home having the another physical object.

12. The system of claim 11 , wherein the operation of mapping each attribute of the canceled listing to a set of similar attributes in the equivalence table comprises:

generating a plurality of sets of attributes by, for each attribute of the canceled listing:

mapping the attribute to a category in the equivalence table;

retrieving a set of attributes associated with the category; and

determining similar attributes based on attributes that occur in a threshold number of the sets of attributes of the plurality of sets of attributes.

13. The system of claim 11 , the operations further comprising:

receiving, from the machine learning model, a percentage match between the canceled listing and each of the set of candidate listings; and

sending, for display to the guest user, a one or more of the set of candidate listings with a percentage match greater than a threshold value.

14. The system of claim 11 , wherein the operation of retrieving, from the listing database, the set of comparable listings, comprises:

retrieving user profile data for the guest user;

determining a ranked set of prioritized attributes of the guest user based on the user profile data; and

determining the set of comparable listings based on the prioritized attributes of the guest user and the mapped similar attributes.

15. The system of claim 11 , wherein the operation of retrieving, from the listing database, the set of comparable listings, comprises:

determining a set of upgrade attributes based on the mapped similar attributes, the upgrade attributes describing listings of higher quality than the canceled listing; and

determining the set of comparable listings based on the mapped similar attribute, each comparable listing including one or more of the upgrade attributes.

16. A non-transitory computer-readable medium configured to store instructions, the instructions when executed by a processor cause the processor to:

receive an indication that a booking by the guest user for a booking accommodation at a home has been canceled by a host user, the booking associated with the canceled listing and a time period;

retrieve, from an attribute database, attributes of the canceled listing, at least one of the attributes reflecting a physical object of the home;

map each attribute of the canceled listing to a set of similar attributes in an equivalence table, the physical object mapped to another physical object;

retrieve, from a listing database, a set of comparable listings, each comparable listing including one or more of the mapped similar attributes and being available for booking during the time period; and

book one of the set of comparable listings for the guest user during the time period by;

inputting, to a machine learning model, the canceled listing and a set of candidate listings, each candidate listing available for booking during the time period, wherein the machine learning model is trained by:

retrieving, from a database, historical user bookings, each historical user booking associated with a set of attributes;

labelling similar historical user bookings as corresponding to one or more booking types based on the set of attributes associated with the historical user booking; and

training the machine learning model using the labelled historical user bookings;

selecting a candidate listing to be the one of the set of comparable listings based on output of the machine learning model; and

populating to a booking data structure with a connection between the one of the set of comparable listings and the guest user, the booked comparable listing being for another home having the another physical object.

17. The non-transitory computer-readable medium of claim 16 , wherein the instructions to map each attribute of the canceled listing to a set of similar attributes in the equivalence table comprise instructions to:

generate a plurality of sets of attributes by, for each attribute of the canceled listing:

mapping the attribute to a category in the equivalence table;

retrieving a set of attributes associated with the category; and

determine similar attributes based on attributes that occur in a threshold number of the sets of attributes of the plurality of sets of attributes.

18. The non-transitory computer-readable medium of claim 17 , the instructions further comprising instructions to:

receive, from the machine learning model, a percentage match between the canceled listing and each of the set of candidate listings; and

send, for display to the guest user, a one or more of the set of candidate listings with a percentage match greater than a threshold value.

19. The non-transitory computer-readable medium of claim 17 , wherein the instructions to retrieve, from the listing database, the set of comparable listings, comprise instructions to:

retrieve user profile data for the guest user;

determine a ranked set of prioritized attributes of the guest user based on the user profile data; and

determine the set of comparable listings based on the prioritized attributes of the guest user and the mapped similar attributes.