Subscription based travel service
Systems and methods are disclosed for providing a subscription travel service. The systems and methods include operations for receiving travel information with a travel date for a user; computing a subscription value as a function of a booking date and the travel date; determining a minimum travel value and a maximum purchase amount based on the computed subscription value; searching a list of travel services that are available on the travel date to identify candidate travel services each having a first cost that exceeds the minimum travel value; selecting a subset of the candidate travel services that each have a second cost that is less than the maximum purchase amount; and generating for display to a user, in a graphical user interface, one or more interactive visual representations of the selected subset of the candidate travel services.
1. A computer-implemented method comprising:
receiving, by one or more processors, travel information with a travel date for a user;
computing, by the one or more processors, a subscription value as a function of a booking date and the travel date, the travel date being later than the booking date, and the subscription value comprising an accumulated value portion and an amortized value portion representing a total amount of resources that will be available to be used for consuming a travel service on the travel date;
determining, by the one or more processors, a minimum travel value and a maximum purchase amount based on the computed subscription value;
searching, by the one or more processors, a plurality of travel services that are available on the travel date to identify candidate travel services each having a first cost that exceeds the minimum travel value;
selecting, by the one or more processors, a subset of the candidate travel services that each have a second cost that is less than the maximum purchase amount;
training a machine learning technique to establish a relationship between historical travel activities and classifications of users by processing training data comprising the historical travel activities derived from a plurality of users, the training being performed by:
retrieving a portion of the training data from a storage device, the portion of the training data comprising historical travel activities of a given user of the plurality of users and a known classification of the given user;
extracting historical travel activities features from the training data for the given user;
utilizing the machine learning technique to estimate a classification for the given user based on the extracted historical travel activities features; and
updating parameters of the machine learning technique to map the estimated classification for the given user to the known classification of the given user;
applying the machine learning technique to travel activity information associated with the user to generate a classification for the user;
selecting a portion of the subset of the candidate travel services based on the generated classification of the user; and
generating, by the one or more processors, for display in a graphical user interface to the user, one or more interactive visual representations of the selected portion of the subset of the candidate travel services, the booking date being a current date on which the searching step, the selecting step, and the generating step are performed.
2. The computer-implemented method of claim 1 , wherein the machine learning technique comprises a neural network, and wherein the travel information includes a geographical destination and a length of stay; and wherein the travel date comprises a date of arrival at the geographical destination.
3. The computer-implemented method of claim 1 , wherein:
the training data comprises a number of reservations made by each of the plurality of users, a subscription duration of each of the plurality of users, a distance to travel destination of each of the plurality of users, a margin amount of each of the plurality of users, a reservation frequency of each of the plurality of users, a cancelation frequency of each of the plurality of users, and the user classification of each of the plurality of users.
4. The computer-implemented method of claim 3 , wherein:
a first set of the subset of candidate travel services are associated with a first type of travel service and a second set of the subset of candidate travel services are associated with a second type of travel service; and
the travel services comprise at least one of hotels, rental cars, airfares, homes/residences, experiential travel, guided tours, cruises, train fares, private aviation, bespoke travel, event-based travel, or space travel, wherein selecting a portion of the subset of the candidate travel services based on the generated classification of the user comprises:
selecting one or more travel services from the first set of the subset of candidate travel services in response to determining that the generated classification of the user represents a first measure of user activeness; and
selecting one or more travel services from the second set of the subset of candidate travel services in response to determining that the generated classification of the user represents a second measure of user activeness.
5. The computer-implemented method of claim 4 , wherein the first measure of user activeness comprises an active user and the second measure of user activeness comprises a non-active user.
6. The computer-implemented method of claim 1 , further comprising computing the accumulated value portion of the subscription value by:
determining a time interval between the booking date and the travel date; and
accumulating the subscription value over the determined time interval.
7. The computer-implemented method of claim 6 , wherein the time interval is monthly, further comprising:
determining a number of months between the booking date and the travel date, wherein the subscription value is computed as a function of a monthly subscription cost to the user and the number of months.
8. The computer-implemented method of claim 1 , further comprising computing the amortized value portion of the subscription value by:
determining an annual cost of a subscription of the user;
dividing the annual cost by a specified repeated time interval in a year;
determining a number of times the time interval repeats between the booking date and the travel date; and
computing the amortized value as a function of the divided annual cost and the determined number of times.
9. The computer-implemented method of claim 8 , wherein the specified repeated time interval is a week.
10. The computer-implemented method of claim 1 , wherein the minimum travel value is determined based on a percentage of the accumulated value portion.
11. The computer-implemented method of claim 1 , wherein the maximum purchase amount is determined as a function of the amortized value portion and the accumulated value portion.
12. The computer-implemented method of claim 11 , wherein the maximum purchase amount is determined based on an adjusted average of the amortized value portion and the accumulated value portion.
13. The computer-implemented method of claim 12 , wherein the adjusted average is adjusted based on an expected margin value, wherein the expected margin value is positive or negative, and wherein the expected margin value is computed based on at least one of a length of time between the booking date and the travel date or a type of travel service.
14. The computer-implemented method of claim 1 , further comprising filtering the identified candidate travel services based on a cancelation policy of the identified candidate travel services.
15. The computer-implemented method of claim 1 , wherein the graphical user interface is presented via a subscription service associated with the subscription value, further comprising aggregating the plurality of travel services by accessing one or more third-party databases that include the respective first costs of the travel services, wherein the travel services are available to non-subscribers of the subscription service for purchase at the respective first costs, further comprising:
determining that the user has reserved a plurality of travel services;
determining that a number of pending reservations included in the plurality of travel services exceeds an allowable number of pending reservations; and
preventing the user from reserving additional travel services until one or more of the reserved plurality of travel services expires or is consumed.
16. The computer-implemented method of claim 15 , further comprising accessing one or more databases of the subscription service to obtain the second cost for each of the plurality of travel services, wherein the travel services are available to subscribers of the subscription service for selection to be reserved.
17. The computer-implemented method of claim 1 , further comprising:
receiving a user selection of a visual representation of the one or more visual representations via the graphical user interface;
reserving the travel service for the user associated with the selected visual representation; and
preventing the user from searching for additional travel services until the reserved travel service expires or is consumed.
18. The computer-implemented method of claim 1 , wherein a first user classification indicates user activity with a subscription service that exceeds a threshold amount and a second user classification indicates user activity with the subscription service that is less than the threshold amount, further comprising
generating individualized travel service lists for the user based on travel behaviors, geographical location, demographics, or a margin target for the user; and
enabling the user to interact with the graphical user interface to sort the travel services.
19. A system comprising:
a memory that stores instructions; and
one or more processors on a server configured by the instructions to perform operations comprising:
receiving travel information with a travel date for a user;
computing a subscription value as a function of a booking date and the travel date, the travel date being later than the booking date, and the subscription value comprising an accumulated value portion and an amortized value portion representing a total amount of resources that will be available to be used for consuming a travel service on the travel date;
determining a minimum travel value and a maximum purchase amount based on the computed subscription value;
searching a plurality of travel services that are available on the travel date to identify candidate travel services each having a first cost that exceeds the minimum travel value;
selecting a subset of the candidate travel services that each have a second cost that is less than the maximum purchase amount;
training a machine learning technique to establish a relationship between historical travel activities and classifications of users by processing training data comprising the historical travel activities derived from a plurality of users, the training being performed by:
retrieving a portion of the training data from a storage device, the portion of the training data comprising historical travel activities of a given user of the plurality of users and a known classification of the given user;
extracting historical travel activities features from the training data for the given user;
utilizing the machine learning technique to estimate a classification for the given user based on the extracted historical travel activities features; and
updating parameters of the machine learning technique to map the estimated classification for the given user to the known classification of the given user;
applying the machine learning technique to travel activity information associated with the user to generate a classification for the user;
selecting a portion of the subset of the candidate travel services based on the generated classification of the user; and
generating for display to a user, in a graphical user interface, one or more interactive visual representations of the selected portion of the subset of the candidate travel services, the booking date being a current date on which the searching step, the selecting step, and the generating step are performed.
20. A non-transitory computer-readable medium comprising instructions stored thereon that are executable by at least one processor to cause a computing device to perform operations comprising:
receiving travel information with a travel date for a user;
computing a subscription value as a function of a booking date and the travel date, the travel date being later than the booking date, and the subscription value comprising an accumulated value portion and an amortized value portion representing a total amount of resources that will be available to be used for consuming a travel service on the travel date;
determining a minimum travel value and a maximum purchase amount based on the computed subscription value;
searching a plurality of travel services that are available on the travel date to identify candidate travel services each having a first cost that exceeds the minimum travel value;
selecting a subset of the candidate travel services that each have a second cost that is less than the maximum purchase amount;
training a machine learning technique to establish a relationship between historical travel activities and classifications of users by processing training data comprising the historical travel activities derived from a plurality of users, the training being performed by:
retrieving a portion of the training data from a storage device, the portion of the training data comprising historical travel activities of a given user of the plurality of users and a known classification of the given user;
extracting historical travel activities features from the training data for the given user;
utilizing the machine learning technique to estimate a classification for the given user based on the extracted historical travel activities features; and
updating parameters of the machine learning technique to map the estimated classification for the given user to the known classification of the given user;
applying the machine learning technique to travel activity information associated with the user to generate a classification for the user;
selecting a portion of the subset of the candidate travel services based on the generated classification of the user; and
generating for display to a user, in a graphical user interface, one or more interactive visual representations of the selected portion of the subset of the candidate travel services, the booking date being a current date on which the searching step, the selecting step, and the generating step are performed.