WO2016131519A1 - Classement personnalisé pour des résultats de recherche d'une interrogation de base de données basée sur un voyage - Google Patents

Classement personnalisé pour des résultats de recherche d'une interrogation de base de données basée sur un voyage Download PDF

Info

Publication number
WO2016131519A1
WO2016131519A1 PCT/EP2016/000007 EP2016000007W WO2016131519A1 WO 2016131519 A1 WO2016131519 A1 WO 2016131519A1 EP 2016000007 W EP2016000007 W EP 2016000007W WO 2016131519 A1 WO2016131519 A1 WO 2016131519A1
Authority
WO
WIPO (PCT)
Prior art keywords
travel
profile
traveler
value
itineraries
Prior art date
Application number
PCT/EP2016/000007
Other languages
English (en)
Inventor
David Pasero
Haixiang Zhao
Laure Canis
Pierre Dor
Feten Zmerli
Original Assignee
Amadeus S.A.S.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from EP15290036.1A external-priority patent/EP3059707A1/fr
Priority claimed from US14/623,699 external-priority patent/US20160239765A1/en
Application filed by Amadeus S.A.S. filed Critical Amadeus S.A.S.
Priority to CA2975794A priority Critical patent/CA2975794A1/fr
Priority to CN201680010605.6A priority patent/CN107251022A/zh
Priority to AU2016222066A priority patent/AU2016222066A1/en
Publication of WO2016131519A1 publication Critical patent/WO2016131519A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • 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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/14—Travel agencies
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2457—Query processing with adaptation to user needs
    • G06F16/24578—Query processing with adaptation to user needs using ranking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/02—Reservations, e.g. for tickets, services or events
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • 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
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]

Definitions

  • the invention generally relates to computers and computer software, and in particular to methods, systems, and computer program products that assign a rank to a travel-related database search result based on a user profile.
  • Reservation systems for travel related services typically include one or more databases that store data relating to the travel services being offered. These databases may include a fare database containing data used in determining a price of a service, and an availability database containing data used to determine if the service is available. Reservation systems may also provide tools that allow end-users, such as travelers and travel agents, to search for and book travel services that satisfy desired trip criteria. To this end, the tools provided may include one or more machine interfaces that enable other systems to access data from the databases. Systems provided access through these machine interfaces may include a Global Distribution System (GDS), as well as systems operated by travel agencies or other resellers.
  • GDS Global Distribution System
  • a traveler may access a travel-related website provided by an indirect seller of travel services.
  • the traveler may provide search criteria, such as an origin, destination, travel dates, booking class, etc., and launch a search query using the website.
  • a search engine provided by one of the aforementioned systems may retrieve data from the databases and generate search results comprising travel itineraries that satisfy the search criteria.
  • the quality of search results provided by the travel-related website can be a deciding factor that distinguishes one online seller of travel-related products, such as online travel agents or agencies, from another.
  • travelers searching online for travel-related products and services such as priced flights, hotels, or car rentals, may prefer one online seller over another based largely on the quality of search results.
  • Travel search results provided by websites of online sellers typically display a large number of travel recommendations that match the traveler's search query. As a result, the traveler may be overwhelmed by the sheer number of travel recommendations embodied in the search results. This may hinder selection of a travel recommendation for booking by the traveler, who may have difficulty identifying one search result that they prefer from a large number of similar search results.
  • One consequence of this inability to select a search result may be that the traveler delays their purchase.
  • the traveler may also seek out another seller to obtain the desired travel services. Presenting an excessive number of search results to the traveler may thereby lower the number of purchases per view, or the "conversion rate" for the website.
  • Presenting an excessive number of travel recommendations to the traveler may also make it difficult for the traveler to find what they consider to be the most relevant search results. This may adversely impact customer satisfaction with the online seller, which could lead to the traveler using a different online seller to purchase their travel products.
  • a method of ranking search results includes receiving search results comprising a plurality of travel itineraries matching a search query.
  • a first travel itinerary and a second travel itinerary may be selected from the plurality of travel itineraries based on a probabilistic profile of a traveler.
  • the method may update the probabilistic profile of the traveler based on an indication the traveler prefers the first travel itinerary to the second travel itinerary.
  • the method may then determine a composite profile of the traveler based on the probabilistic profile, and rank the search results based on the composite profile.
  • an apparatus for ranking search results includes a processor and a memory coupled to the processor.
  • the memory includes program code that, when executed by the processor, causes the apparatus to receive the search results comprising the plurality of travel itineraries matching the search query, and select the first travel itinerary and the second travel itinerary from the plurality of travel itineraries based on the probabilistic profile of the traveler.
  • the apparatus may receive the indication the traveler prefers the first travel itinerary to the second travel itinerary, and update the probabilistic profile of the traveler based on the indication.
  • the apparatus may further determine the composite profile of the traveler based on the probabilistic profile, and rank the search results based on the composite profile.
  • a computer program product includes a non-transitory computer-readable storage medium including program code.
  • the program code may be configured, when executed by the processor, to cause the processor to receive the search results comprising the plurality of travel itineraries matching the search query, and select the first travel itinerary and the second travel itinerary from the plurality of travel itineraries based on the probabilistic profile of the traveler.
  • the processor may receive the indication the traveler prefers the first travel itinerary to the second travel itinerary, and update the probabilistic profile of the traveler based on the indication.
  • the processor may further determine the composite profile of the traveler based on the probabilistic profile, and rank the search results based on the composite profile.
  • FIG. 1 is a diagrammatic view of an exemplary operating environment including a plurality of computing systems in communication via a network.
  • FIG. 2 is a diagrammatic view of an exemplary computing system of FIG. 1.
  • FIG. 3 is a diagrammatic view of a computing system including a travel-related website, a search engine, an itinerary database, and a profile database.
  • FIG. 4 is a diagrammatic view of a probabilistic profile used by the search engine in FIG. 3.
  • FIG. 5 is a flow-chart illustrating a process that may be executed by the computing system of FIG. 3 to select search results for display to a traveler.
  • FIG. 6 is a flow-chart illustrating a process that may be executed by the computing system of FIG. 3 to adapt the probabilistic profile to reflect the personal preferences of the traveler.
  • Embodiments of the invention may be implemented by a processing and database system, such as a computerized reservation system.
  • the processing and database system may be configured to respond to a travel search query by identifying travel itineraries that satisfy the travel search query, and returning the identified travel itineraries as search results.
  • the processing and database system may be embodied in a Global Distribution System (GDS) that identifies travel itineraries including flights from multiple airlines.
  • GDS Global Distribution System
  • the GDS may also provide travel itineraries that include services such as train travel, hotel rooms, car rentals, sightseeing, and other travel-related activities.
  • the processing and database system may include a search engine that receives a search query.
  • the search query may originate from a traveler who is accessing a website of an indirect seller of travel services, such as a travel agency, and may include data defining search criteria, such as an origin, destination, travel dates, booking class, etc.
  • the search engine may retrieve or otherwise generate travel itineraries that satisfy the search criteria using data in one or more databases.
  • the databases may include databases having caches of pre-computed travel itineraries.
  • the databases may also include service availability and fare databases that provide real-time data from which travel itineraries may be generated.
  • the search engine may rank the search results according to a composite profile specific to the traveler.
  • the search engine may then select a set of one or more travel itineraries from the search results based on this ranking and display the selected results to the traveler.
  • the search engine may thereby provide the traveler with recommended travel itineraries in accordance with the personal preferences of the traveler.
  • embodiments of the invention may reduce the number of search results that are transmitted to the traveler, stored in a cache or database of search results, or that must be processed by the processing and database system, as compared to systems that lack this feature.
  • Embodiments of the invention may thereby reduce the processing load on, and improve the performance of, the processing and database system.
  • the composite profile may be determined based on a probabilistic profile of the traveler.
  • the search engine may select a subset of two or more travel itineraries from the search results based on the probabilistic profile, and request the traveler indicate which travel itinerary is preferred.
  • the search engine may update the probabilistic profile to optimize identification of the selected travel itinerary by the composite profile.
  • the search engine may then select a new subset of travel itineraries based on the updated probabilistic profile and repeat the training process. The process may be repeated multiple times until the search engine can reliably predict user preferences.
  • the search engine may accurately determine which search results have a higher value to the traveler. These search results may then be displayed to the traveler so that the traveler is provided with travel itineraries which match their personal preferences without having to sort through a large number of search results. Embodiments of the invention may thereby enable travelers to avoid evaluating an excessive number of travel itineraries to find a preferred choice.
  • an operating environment 10 in accordance with an embodiment of the invention may include the GDS 12, one or more travel service provider systems, such as carrier system 14, one or more indirect seller systems, such as travel agency system 16, and a traveler system 18.
  • the carrier system 14 may be in communication with one or more databases, such as an availability database 20 and a fare database 22.
  • Each of the GDS 12, carrier system 14, travel agency system 16, and traveler system 18 may communicate through a network 24.
  • the network 24 may include one or more private or public networks (e.g., the Internet) that enable the exchange of data.
  • the GDS 12 may be configured to facilitate communication between the carrier system 14 and travel agency system 16 by enabling travel agents, validating carriers, or other indirect sellers to book reservations on the carrier system 14 via the GDS 12.
  • the GDS 12 may maintain links to a plurality of carrier systems via the network 24 that enable the GDS 12 to route reservation requests from the validating carrier or travel agency to a corresponding operating carrier.
  • the carrier system 14 and travel agency system 16 may thereby book flights on multiple airlines via a single connection to the GDS 12.
  • the carrier system 14 may include a Computer Reservation System (CRS) that enables the GDS 12 or travel agency system 16 to reserve and pay for airline tickets.
  • CRS Computer Reservation System
  • the carrier system 14 may interact with the availability database 20 and fare database 22 to price and reserve travel services in response to booking requests or other queries from the GDS 12.
  • the carrier system 14 may also interact with other carrier systems (not shown), either directly or through the GDS 12, to enable a validating carrier to sell tickets for seats provided by the operating carrier. The operating carrier may then bill the validating carrier for the services provided.
  • the travel agency system 16 may provide travel agents with an interface for accessing the GDS 12 that enables agents to search for and book travel itineraries.
  • the travel agency system 16 may also include an application accessible by the traveler system 18 that enables the traveler to search for and book travel itineraries without the help of a travel agent.
  • This application may comprise, for example, a travel-related website that is accessible over the network 24 using a web-browser provided by the traveler system 18.
  • the traveler system 18 may comprise a desktop computer, laptop computer, tablet computer, smart phone, or any other suitable computing device.
  • the traveler may use the traveler system 18 to search for and book travel services by accessing the GDS 12, carrier system 14, travel agency system 16, or any other suitable system though the network 24.
  • the traveler may launch a browser application, and use the browser application to search for travel services on the website provided by the travel agency system 16, or a website provided by the GDS 12, carrier system 14, or any other suitable system.
  • the traveler may book a selected travel service by entering payment information into the website.
  • the GDS 12, carrier system 14, travel agency system 16, traveler system 18, availability database 20, and fare database 22 of operating environment 10 may be implemented on one or more computer devices or systems, such as exemplary computer system 30.
  • the computer system 30 may include a processor 32, a memory 34, a mass storage memory device 36, an input/output (I/O) interface 38, and a Human Machine Interface (HMI) 40.
  • the computer system 30 may also be operatively coupled to one or more external resources 42 via the network 24 or I/O interface 38.
  • External resources may include, but are not limited to, servers, databases, mass storage devices, peripheral devices, cloud-based network services, or any other suitable computer resource that may be used by the computer system 30.
  • the processor 32 may include one or more devices selected from microprocessors, micro-controllers, digital signal processors, microcomputers, central processing units, field programmable gate arrays, programmable logic devices, state machines, logic circuits, analog circuits, digital circuits, or any other devices that manipulate signals (analog or digital) based on operational instructions that are stored in the memory 34.
  • Memory 34 may include a single memory device or a plurality of memory devices including, but not limited to, read-only memory (ROM), random access memory (RAM), volatile memory, non-volatile memory, static random access memory (SRAM), dynamic random access memory (DRAM), flash memory, cache memory, or any other device capable of storing information.
  • the mass storage memory device 36 may include data storage devices such as a hard drive, optical drive, tape drive, volatile or non-volatile solid state device, or any other device capable of storing information.
  • the processor 32 may operate under the control of an operating system 46 that resides in memory 34.
  • the operating system 46 may manage computer resources so that computer program code embodied as one or more computer software applications, such as an application 48 residing in memory 34, may have instructions executed by the processor 32.
  • the processor 32 may execute the application 48 directly, in which case the operating system 46 may be omitted.
  • One or more data structures 50 may also reside in memory 34, and may be used by the processor 32, operating system 46, or application 48 to store or manipulate data.
  • the I/O interface 38 may provide a machine interface that operatively couples the processor 32 to other devices and systems, such as the network 24 or external resource 42.
  • the application 48 may thereby work cooperatively with the network 24 or external resource 42 by communicating via the I/O interface 38 to provide the various features, functions, applications, processes, or modules comprising embodiments of the invention.
  • the application 48 may also have program code that is executed by one or more external resources 42, or otherwise rely on functions or signals provided by other system or network components external to the computer system 30. Indeed, given the nearly endless hardware and software configurations possible, persons having ordinary skill in the art will understand that embodiments of the invention may include applications that are located externally to the computer system 30, distributed among multiple computers or other external resources 42, or provided by computing resources
  • the HMI 40 may be operatively coupled to the processor 32 of computer system 30 in a known manner to allow a user to interact directly with the computer system 30.
  • the HMI 40 may include video or alphanumeric displays, a touch screen, a speaker, and any other suitable aud io and visual indicators capable of providing data to the user.
  • the HMI 40 may also include input devices and controls such as an alphanumeric keyboard, a pointing device, keypads, pushbuttons, control knobs, microphones, etc., capable of accepting commands or input from the user and transmitting the entered input to the processor 32.
  • a database 44 may reside on the mass storage memory device 36, and may be used to collect and organize data used by the various systems and modules described herein.
  • the database 44 may include data and supporting data structures that store and organize the data.
  • the database 44 may be arranged with any database organization or structure including, but not limited to, a relational database, a hierarchical database, a network database, or combinations thereof.
  • a database management system in the form of a computer software application executing as instructions on the processor 32 may be used to access the information or data stored in records of the database 44 in response to a query, where a query may be dynamically determined and executed by the operating system 46, other applications 48, or one or more modules.
  • a computer system 60 may provide a search engine 62, an itinerary database 64, and a profile database 66.
  • the computer system 60 may be in communication with the carrier system 14 and a travel- related website 68.
  • the computer system 60 may be provided by the GDS 12, and the travel-related website 68 may be provided by the travel agency system 16.
  • the travel-related website 68 could be provided by any suitable computer system.
  • the traveler may access one or more web-pages provided by the travel -related website 68 using the traveler system.18.
  • the traveler may interact with the web-pages to search for and book travel itineraries. These travel itineraries may comprise one or more travel services, such as flights, hotel rooms, car rentals, or any other travel service.
  • the traveler system 18 may transmit a search query 70 to the travel-related website 68.
  • the search query 70 may include data defining search criteria, such as an origin and destination, travel dates, and/or booking class.
  • the travel-related website 68 may transmit a search query 72 including data defining the search criteria and traveler identity to the search engine 62.
  • the search engine 62 may generate and transmit a database search query 74 to the itinerary database 64.
  • the itinerary database 64 may identify one or more travel itineraries that satisfy the search criteria.
  • the identified travel itineraries may include predetermined travel itineraries stored in a cache or newly determined travel itineraries retrieved from the carrier system 14.
  • the newly determined travel itineraries may be generated in accordance with service availability data (e.g., available seats) received from the availability database 20, and fare data received from the fare database 22.
  • the newly determined travel itineraries may also be stored in the cache and accessed by the itinerary database 64 so that they are available as predetermined travel itineraries for subsequent search queries received by the search engine 62.
  • the itinerary database 64 may transmit a response 76 to the search engine 62 containing search results comprising the identified travel itineraries. While the itinerary database 64 is compiling the search results, or in response to receiving the response 76, the search engine 62 may transmit a query 78 to the profile database 66 that includes data defining the traveler identity. The query 78 may request the profile database 66 provide a user profile for the traveler. If a user profile exists for the traveler, the profile database 66 may transmit the user profile to the search engine 62 in a response 80.
  • the profile database 66 may transmit a default profile to the search engine 62, or the response 80 may indicate that no user profile exists for the identified traveler. In the latter case, the search engine 62 may generate a default profile for the traveler.
  • Each user profile may be associated with a composite profile CP comprising a set of coefficients, with each coefficient corresponding to a characteristic of the travel itinerary.
  • the value of each coefficient may indicate how important the corresponding characteristic of the travel itinerary is to the traveler.
  • one of the coefficients may correspond to the total travel time of the itinerary.
  • the value assigned to the coefficient may have a monetary value (e.g., $20/hoUr of travel time).
  • the composite profile CP may be used to determine a "total cost" of the travel itinerary that captures both the price CM of the travel products and the perceived value added or subtracted for the traveler by specific features of the travel itinerary.
  • the price M of travel itinerary having the 4 hour travel time would have to be more than $80 greater than the price CM of the travel itinerary having the 8 hour travel time in order to have a higher total cost CTOT to the traveler, assuming all other features of the travel itineraries are the same.
  • the composite profile CP may thereby provide a means of determining an overall desirability of the travel itinerary to the traveler in question that captures both the price of the travel products comprising the itinerary and the impact of convenience factors on the traveler.
  • the search engine 62 may apply the composite profile CP to determine the total cost CTOT for each of the search results in response 80.
  • the search engine 62 may then rank the search results based on their total cost CTOT, with search results having a lower total cost CTOT being given a higher rank.
  • the search engine 62 may then transmit a response 82 to the travel-related website 68 that includes at least a portion of the ranked search results selected based on rank.
  • the selected search results may comprise a predetermined number of search results representing the highest ranking search results.
  • the travel-related website 68 may format the results and transmit a response 84 to the search query 70 that includes the search results selected by the search engine 62 for display to the traveler.
  • the response 84 may comprise data that causes the browser application running on the traveler system 18 to display a window populated with the selected search results, for example.
  • FIG. 4 depicts a diagrammatic view of a probabilistic profile 90 comprising an array of N value profiles VP/ - VPN.
  • Each value profile VPi - VPN of the probabilistic profile 90 may comprise a probability, or weight w ⁇ - WN, (represented by the height of each value profile
  • the probabilistic profile 90 may be represented by an array [w ⁇ VP], w 2 VPi, . . . WN* VPN].
  • Each value profile VPi - VPN of the probabilistic profile 90 may further comprise a set of travel itinerary characteristics (v ⁇ , v 2 , ... VM).
  • the weight wi - WN may provide the relative effect of the corresponding value profile VPi - VPN in a blending function that provides the composite profile CP of the traveler.
  • each value profile VPi - VPN may be determined based to how well the corresponding value profile VPi - VPN identifies travel itineraries having the lowest total cost CTOT for the traveler in question from a subset of travel itineraries.
  • the subset of travel itineraries may, for example, comprise two or more travel itineraries selected from a larger set of travel itineraries satisfying a search request.
  • Each coefficient vi - VM of the value profile may describe a value of a corresponding travel itinerary characteristic.
  • the coefficients vi - VA/ may include coefficients corresponding to the cost/hour for travel time (e.g., add $20 for each hour of travel time), a cost/stop for each stop or connection required (e.g., add $30 for each stopover), a cost/stop for each stop having an overnight stopover requirement (e.g., add $100 for an overnight stopover), a cost associated with an undesirable carrier (e.g., add $40 if carrier is Acme Air), a cost associated with an undesirable type of carrier (e.g., add $50 if carrier is a low-cost carrier), a cost associated with a particular air carrier policy (e.g., add $40 if carrier only allows one carry- on bag or has a record of excessive delays), or an additional cost if departure is outside a preferred travel time (e.g., add $30 for flights departing prior to 8AM), to name just a few.
  • each coefficient may be quantized into discrete values.
  • the costs associated with each coefficient may be quantized to values having fixed increments (e.g., $10) over a pre-determined range (e.g., $0 to $ 100).
  • the probabilistic profile may be used to drive the search for travel itineraries, and the highest ranking search results selected from the search results output by the search.
  • the probabilistic profile may also be used to select the most relevant recommendations from the output of a non-personalized search for travel itineraries matching the travelers search terms.
  • the composite profile for the traveler may be constructed from one or more of the value profiles VPi - VPN.
  • the composite profile CP for the traveler may be determined by taking a sum of the weighted value profiles VPi - VPN as follows: N
  • the resulting composite profile CP may provide the set of M travel itinerary characteristics (v ⁇ , v 2 , ... VM) for the traveler in which each itinerary characteristic (v ⁇ , v 2 , ... VM) of the composite profile CP comprises the weighted sum of the corresponding itinerary characteristics (v ⁇ , v 2 , . . . VM) of each of the value profiles VPi - VPN.
  • a flow-chart depicts a process 100 that may be performed by the search engine 62, or some other suitable module provided by the computing system 60, to determine the weights w ⁇ - M>N of each value profile VPi - VPN comprising the probabilistic profile 90 for a specific traveler.
  • the process 100 may obtain search results comprising a plurality of travel itineraries that match a set of search criteria provided by the traveler. These search results may be received, for example, by the search engine 62 in response to transmitting the database search query 74 to the itinerary database 64 as described above with respect to FIG. 3. The process 100 may then proceed to block 104 and determine if the traveler has a pre-existing probabilistic profile 90.
  • the traveler may have a pre-existing probabilistic profile 90 if, for example, the traveler has used the travel-related website on a previous occasion.
  • a returning traveler may be identified based on a data structure stored on the traveler system 18 that identifies the traveler (e.g., an HTTP cookie), a user identity or account number provided to the travel-related website 68, or any other suitable method.
  • the process 100 may proceed to block 106 and retrieve the probabilistic profile 90 from a data storage location, such as the profile database 66. If the traveler is not associated with a pre-existing probabilistic profile 90, e.g., the traveler is a first time user of the system ("NO" branch of decision block 104), the process 100 may proceed to block 108 and retrieve a default probabilistic profile 90.
  • the default probabilistic profile 90 may be a probabilistic profile 90 in which the weights W I - WN have equal values. This may result in a uniform probabilistic value across all criteria used in the value profiles.
  • coefficient vi may provide the cost/hour for travel in $10 increments from $0 to $50
  • coefficient v 2 may provide the cost per stop in $10 increments from $0 to $100
  • coefficient v 3 may provide the cost of the carrier being a low-cost carrier in $10 increments from $0 to $100
  • coefficient v 4 may provide the cost of using a non-favored carrier in $10 increments from $0-$100
  • coefficient v 5 may provide the cost of including a travel segment departing during a non-preferred time slot.
  • These costs may be added multiple times for a particular itinerary. For example, the cost of an early departure could be added for each leg of the travel itinerary that has an undesirable departure time, e.g., one each for the inbound and outbound legs of the journey.
  • the process 100 may then proceed to block 110 and present questions to the traveler regarding their travel preferences.
  • Exemplary travel preferences may include preferred airlines, travel times, non-favored carriers, or any other travel preference that could be used to help identify preferred travel itineraries or determine the total cost CTOT-
  • the process 100 may then generate a user profile that includes data indicative of the traveler's preferences and probabilistic profile, and save this user profile in the profile database 66 for future use. In response to obtaining either the pre-existing or default probabilistic profile, the process 100 may proceed to block 1 12.
  • the process 100 may select a set of the search results. To this end, the process 100 may generate the composite profile from the probabilistic profile. The process 100 may then rank the search results using the composite profile, and select the set of search results based on the rank of the search results. For example, the set of search results may comprise a portion of the search results representing the highest ranking search results.
  • the process 100 may proceed to block 1 14, and select a subset of travel itineraries (e.g., a pair of travel itineraries TI ⁇ , Th) from the set of search results.
  • the process 100 may then request that the traveler select one travel itinerary from the subset.
  • the travel itineraries may be selected based on the composite profile defined by the probabilistic profile 90 so that each travel itinerary in the subset provides the maximum information on the composite profile.
  • more than two travel itineraries may be selected for the subset (e.g., four travel itineraries Th, Th, TI3, TI4).
  • the traveler may be asked to rank the travel itineraries in the subset from best to worst rather than simply selecting one as the most preferred travel itinerary.
  • the process may proceed to block 1 16.
  • the process 100 may update the probabilistic profile 90 of the traveler based on which travel itinerary the traveler selected.
  • the traveler selecting travel itinerary TI ⁇ over Th may indicate that the total cost CTOT of travel itinerary TI ⁇ is less than the total cost CTOT of travel of travel itinerary Th. That is:
  • tgue represents the travel time
  • S n represents the number of stopovers
  • N cn represents the number of legs flown on a low-cost carrier
  • NuDCn represents the number of legs flown on an undesirable carrier
  • Nursn represents the number of undesirable time slots (e.g., early departure times) for the respective travel itinerary TI ⁇ , Th.
  • the process 100 may use this information to update the travelers composite profile CP.
  • the process 100 may update the probabilistic profile 90 based on an assumption that the traveler is purely rational, or on an assumption that the traveler is not purely rational.
  • the process 100 may only use value profiles VPi - FiV that produce a result in agreement with the traveler's selection of travel itinerary TI ⁇ over Th in generating the value profile for the traveler. That is, a value profile VPi that is not compatible with the traveler's choice may be excluded from determining the travelers composite profile CP.
  • the weight w, of each value profile VPi that fails to produce a result matching the travel itinerary selection of the traveler may be set to 0.
  • the weights w, of the value profiles VPi that produce a result matching the travel itinerary selection of the traveler may then be normalized so that the sum of the remaining non-zero weights is unity.
  • the traveler may select a travel itinerary that is inconsistent with the. traveler's previous selections. That is, the traveler's selection may appear irrational. This may be due to the traveler making a mistake, or due to some aspect of the trip varying from the normal parameters, e.g., the traveler is booking personal rather than business travel.
  • the process 100 may update the value profile of the traveler using an "irrational model". Under the irrational model, the weight of each value profile that is consistent with the traveler's choice may be increased from its previous value, and the weight of each value profile that is inconsistent with the traveler's selection may be reduced from its previous value.
  • the amount that the weights are changed may be controlled by a "rationality coefficient" r that is used to determine a multiplier by which the previous weight is multiplied.
  • the updated weight w, of each value profile VPi that fails to produce a result matching the travel itinerary selection of the traveler may be set to (1 - r)xw,. so that:
  • the updated weight wj of each value profile VPj that produces a result matching the travel itinerary selection of the traveler may be set to:
  • each coefficient of the value profile may be provided by:
  • the process 100 may proceed to block 1 18 and re-rank the search results using the updated composite profile CP.
  • the process 100 may then proceed to block 120 and determine if the search results before the re-ranking match the search results after the ranking.
  • the search results may be considered as matching if there is no difference between the search results before and after the re-ranking, or a difference exists but is considered to be an insignificant difference.
  • a difference may exist, for example, if the ranked order of the search results, or a portion thereof (e.g., the order of ranked results in the set of search results) has changed due to the re-ranking. Whether the difference is considered significant may depend on how many of the results have changed, or the rank of the results that have changed. For example, a change in the order of lower ranking results may be an insignificant difference.
  • the process 100 may proceed to block 1 12 and select a new set of search results based on the updated composite profile CP. If the search results match ("YES" branch of decision block 1 18), the process 100 may proceed to block 122 and display a number of the highest ranked search results. The process 100 may also update the user profile of the traveler in accordance with the updated probabilistic profile so that the updated probabilistic profile is retrieved by the process 100 the next time a search query is received from the traveler.
  • a flow-chart depicts a process 130 that may be performed by the search engine 62, or some other suitable module provided by the computing system 60, to determine which travel itineraries are selected for the set of travel itineraries.
  • the process 130 may select a pair of travel itineraries from the search results. The initial pair may be selected based on characteristics of the travel itineraries. For example, the process may select travel itineraries based on cost, starting with search results that have the lowest cost. However, the invention is not limited to selections based on cost, or any other specific characteristic or combinations of characteristics of the travel itinerary.
  • the process 130 may proceed to block 134 and determine if one of the travel itineraries in the selected pair "dominates" the other travel itinerary.
  • One travel itinerary may be considered to dominate the other travel itinerary if each criteria of the one travel itinerary has a preferred value as compared to the corresponding criteria of the other travel itinerary. For example, the one travel itinerary is cheaper, faster, has fewer stopovers, is on a preferred carrier, and has preferred departure times as compared to the other travel itinerary. If one of the travel itineraries dominates ("YES" branch of decision block 134), the process may proceed to block 136. If neither travel itinerary of the pair dominates the other ("NO" branch of decision block 134), the process 130 may proceed to block 138.
  • the process 130 may select a value profile VP t from the probabilistic profile of the traveler, and determine the total cost CTOT of each travel itinerary of the pair of travel itineraries using the selected value profile VP
  • the process 130 may select value profiles VPi in an order that gives a preference to value profiles VPi having coefficient values in the middle of the allowable range of coefficient values. Beginning the selection process using value profiles VPi near the middle may allow the process 130 to select a suitable test pair of travel itineraries more quickly than by randomly selecting the value profile VPi.
  • the selection of value profiles VPi is based on the assumption that the values of coefficients each corresponding to a characteristic of the travel itinerary, i.e. a preference of a traveler to a specific criteria of the travel itinerary, are normally distributed within the allowable range of coefficient values.
  • the process 130 may proceed to block 140 and add the value profile Vn to one of two subsets of value profiles VPi based on which travel itinerary is identified as having the lowest total cost CTOT by the value profile.
  • one subset of value profiles VPi may be populated with value profiles VPi that produce the lowest total cost CTOT for one of the travel itineraries of the pair, and the other subset of value profiles VPi may be populated with value profiles VPi that produce the lowest total CTOT for the other of the travel itineraries of the pair.
  • probabilistic profile may be added to a corresponding subset of value profiles VP, based on the relative values of their coefficients.
  • the process 130 may determine that all other value profiles VPi having a coefficient vi with a value greater than the coefficient vi of the present value profile VPi would also identify travel itinerary TI ⁇ as the preferred travel itinerary if the other coefficients v 2 , v 3 , v , v 5 are equal in value to those of the present value profile VPi.
  • the process 130 may determine that all other value profiles VPt having a coefficient vi with a value less than the coefficient vi of the present value profile VPi would identify travel itinerary Th as the preferred travel itinerary if the other coefficients v 2 , v 3 , v 4 , v 5 are equal in value to those in the present value profile VPi.
  • the value profile VPi is added to one of the subsets of value profiles due to one characteristic of the travel itinerary, more precisely due to a coefficient of the value profile VPi corresponding to the characteristic of the travel itinerary having a specific value, then all additional value profiles VPj with coefficients/characteristics identical to those of VPi other than the one
  • coefficient/characteristic being examined are added to the same subset without calculating their respective CTOT for the pair of travel itineraries TI ⁇ and Th, if the respective coefficient of the additional value profiles corresponding to the examined characteristic has a given value indicating a more significant level of traveler preference for the examined characteristic so that it is clear that all those additional value profiles will also be added to that same subset.
  • additional second value profiles from the probabilistic profile may be added to the first subset of value profiles with identical values of the respective at least one second coefficients and with values of the respective first coefficients being - depending on characteristics of each travel itineraries of the pair of travel itineraries being weighted by the coefficients - greater or less than the value of the first coefficient of the first value profile.
  • This principle may be extended to the other coefficients to reduce the number of iterations the process 130 must execute.
  • the value profiles VPi that identify travel itineraries TI ⁇ or Th may thereby be determined more quickly than would be possible if by evaluating each value profile VPi individually.
  • the process 130 may proceed to block 142 and determine if each subset of value profiles has at least a minimum amount of value profiles VPi.
  • the amount of the value profiles VPi in each subset of value profiles VP may be determined based on a cumulative weight of the value profiles VPi in each subset.
  • the cumulative weight may be determined, for example, by summing the weight w, of each value profile VPi in the subset being evaluated.
  • the process 130 may then proceed to block 144 and determine if the cumulative weight of each subset of value profiles VPi exceeds a minimum amount, or threshold (e.g., 40% of the cumulative weight of all the value profiles in the probabilistic profile). If each of the subsets has a cumulative weight exceeding the threshold ("YES" branch of decision block 144), the process 130 may proceed to block 146 and select the pair of travel itineraries as the test pair. If the cumulative weight of each subset does not exceed the threshold ("NO" branch of decision block 144), the process 130 may proceed to block 148.
  • threshold e.g. 40% of the cumulative weight of all the value profiles in the probabilistic profile.
  • the process 130 may determine if all the value profiles VPi - VPN of the probabilistic profile have been evaluated and assigned to one of the subsets of value profiles If all the value profiles VP ⁇ - VPN have not been evaluated ("NO" branch of decision block 148), the process may proceed to block 150 and select the next unevaluated value profile VPi from the probabilistic profile before proceeding to block 140 to repeat the evaluation process. If all the value profiles VPi - VPN have been evaluated (“YES" branch of decision block 148), the process 130 may proceed to block 136.
  • the process 130 may determine if all possible pairs of travel itineraries have been evaluated. That is, are there any unique combinations of travel itineraries from the search results that have not been evaluated. If not all possible pairs of travel itineraries have been evaluated ("NO" branch of decision block 136), the process 130 may proceed to block 152 and select the next pair of travel itineraries before proceeding to block 134 to begin the evaluation process.
  • the process 130 may proceed to block 154.
  • the process 130 may determine which pair of travel itineraries produced the most even split between the cumulative weight of each subset of value profiles VP T . That is, the process 130 may determine which pair of travel itineraries produced subsets of value profiles VPi with the smallest difference between their cumulative weights. This may identify subsets in which the cumulative weight is as close to 50% of the total cumulative weight of all the value profiles VP; - VPN in the probabilistic profile. The process 130 may then select this pair of travel itineraries as the subset of travel itineraries.
  • routines executed to implement the embodiments of the invention may be referred to herein as "computer program code,” or simply "program code.”
  • Program code typically comprises computer-readable instructions that are resident at various times in various memory and storage devices in a computer and that, when read and executed by one or more processors in a computer, cause that computer to perform the operations necessary to execute operations and/or elements embodying the various aspects of the embodiments of the invention.
  • Computer- readable program instructions for carrying out operations of the embodiments of the invention may be, for example, assembly language or either source code or object code written in any combination of one or more programming languages.
  • the program code embodied in any of the applications/modules described herein is capable of being individually or collectively distributed as a program product in a variety of different forms.
  • the program code may be distributed using a computer-readable storage medium having computer-readable program instructions thereon for causing a processor to carry out aspects of the embodiments of the invention.
  • Computer-readable storage media which is inherently non-transitory, may include volatile and non-volatile, and removable and non-removable tangible media implemented in any method or technology for storage of information, such as computer-readable instructions, data structures, program modules, or other data.
  • Computer-readable storage media may further include RAM, ROM, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other solid state memory technology, portable compact disc read-only memory (CD-ROM), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and which can be read by a computer.
  • a computer-readable storage medium should not be construed as transitory signals per se (e.g., radio waves or other propagating electromagnetic waves, electromagnetic waves propagating through a transmission media such as a waveguide, or electrical signals transmitted through a wire).
  • Computer-readable program instructions may be downloaded to a computer, another type of programmable data processing apparatus, or another device from a computer- readable storage medium or to an external computer or external storage device via a network.
  • Computer-readable program instructions stored in a computer-readable medium may be used to direct a computer, other types of programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the computer- readable medium produce an article of manufacture including instructions that implement the functions, acts, and/or operations specified in the flow-charts, sequence diagrams, and/or block diagrams.
  • the computer program instructions may be provided to one or more processors of a general purpose computer, a special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the one or more processors, cause a series of computations to be performed to implement the functions, acts, and/or operations specified in the flow-charts, sequence diagrams, and/or block diagrams.
  • the functions, acts, and/or operations specified in the flow-charts, sequence diagrams, and/or block diagrams may be re-ordered, processed serially, and/or processed concurrently consistent with embodiments of the invention.
  • any of the flow-charts, sequence diagrams, and/or block diagrams may include more or fewer blocks than those illustrated consistent with embodiments of the invention.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Human Resources & Organizations (AREA)
  • Databases & Information Systems (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Data Mining & Analysis (AREA)
  • Primary Health Care (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

L'invention concerne des procédés, des systèmes et des produits-programmes informatiques pour traiter des résultats d'interrogation de base de données. Des résultats d'interrogation sont extraits d'une base de données d'itinéraires de voyage sur la base de critères de recherche fournis par un voyageur, et un sous-ensemble d'itinéraires de voyage est sélectionné à partir des résultats de recherche sur la base d'un profil composite. Le voyageur peut identifier quels itinéraires de voyage du sous-ensemble il préfère, et le résultat de cette sélection est utilisé pour mettre à jour un profil probabiliste du voyageur. Les résultats de recherche peuvent être reclassés à l'aide d'un profil composite mis à jour, et un nouveau sous-ensemble d'itinéraires de voyage est sélectionné sur la base du reclassement. Le voyageur peur indiquer quel itinéraire de voyage est préféré à partir du nouveau sous-ensemble, et le profil probabiliste est mis à jour sur la base de ceci. Le processus de reclassement de résultats de recherche et de mise à jour du profil probabiliste peut être répété jusqu'à ce que des ensembles successifs de résultats classés correspondent les uns aux autres.
PCT/EP2016/000007 2015-02-17 2016-01-05 Classement personnalisé pour des résultats de recherche d'une interrogation de base de données basée sur un voyage WO2016131519A1 (fr)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CA2975794A CA2975794A1 (fr) 2015-02-17 2016-01-05 Classement personnalise pour des resultats de recherche d'une interrogation de base de donnees basee sur un voyage
CN201680010605.6A CN107251022A (zh) 2015-02-17 2016-01-05 用于旅行相关的数据库查询的搜索结果的个性化排名
AU2016222066A AU2016222066A1 (en) 2015-02-17 2016-01-05 Personalized ranking for search results of a travel-related database query

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US14/623,699 2015-02-17
EP15290036.1 2015-02-17
EP15290036.1A EP3059707A1 (fr) 2015-02-17 2015-02-17 Classement personnalisé de résultats de recherche d'une requête de base de données liée aux voyages
US14/623,699 US20160239765A1 (en) 2015-02-17 2015-02-17 Personalized ranking for search results of a travel-related database query

Publications (1)

Publication Number Publication Date
WO2016131519A1 true WO2016131519A1 (fr) 2016-08-25

Family

ID=55085639

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2016/000007 WO2016131519A1 (fr) 2015-02-17 2016-01-05 Classement personnalisé pour des résultats de recherche d'une interrogation de base de données basée sur un voyage

Country Status (4)

Country Link
CN (1) CN107251022A (fr)
AU (1) AU2016222066A1 (fr)
CA (1) CA2975794A1 (fr)
WO (1) WO2016131519A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110322330A (zh) * 2019-07-12 2019-10-11 四川亨通网智科技有限公司 基于大数据的智慧旅游景区订单处理方法

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100305984A1 (en) * 2009-06-01 2010-12-02 Scopia, LLC Intermodal trip planner
US20140108070A1 (en) * 2012-10-11 2014-04-17 Getgoing, Inc. Using multi-destination searches to facilitate the purchase of travel itineraries

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020147619A1 (en) * 2001-04-05 2002-10-10 Peter Floss Method and system for providing personal travel advice to a user
US20020173978A1 (en) * 2001-05-17 2002-11-21 International Business Machines Corporation Method and apparatus for scoring travel itineraries in a data processing system
US7836057B1 (en) * 2001-09-24 2010-11-16 Auguri Corporation Weighted preference inference system and method
US20050033616A1 (en) * 2003-08-05 2005-02-10 Ezrez Software, Inc. Travel management system providing customized travel plan
CN101876979B (zh) * 2009-04-28 2012-08-29 株式会社理光 查询扩展方法及查询扩展系统
CN103309864B (zh) * 2012-03-07 2018-10-19 深圳市世纪光速信息技术有限公司 一种搜索结果显示方法、装置及系统
EP2912624A4 (fr) * 2012-10-23 2016-04-27 Olset Inc Procédés et systèmes d'organisation de voyage
CN104239458A (zh) * 2014-09-02 2014-12-24 百度在线网络技术（北京）有限公司 搜索结果的展现方法和装置

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100305984A1 (en) * 2009-06-01 2010-12-02 Scopia, LLC Intermodal trip planner
US20140108070A1 (en) * 2012-10-11 2014-04-17 Getgoing, Inc. Using multi-destination searches to facilitate the purchase of travel itineraries

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110322330A (zh) * 2019-07-12 2019-10-11 四川亨通网智科技有限公司 基于大数据的智慧旅游景区订单处理方法

Also Published As

Publication number Publication date
CA2975794A1 (fr) 2016-08-25
AU2016222066A1 (en) 2017-08-24
CN107251022A (zh) 2017-10-13

Similar Documents

Publication Publication Date Title
US20160239765A1 (en) Personalized ranking for search results of a travel-related database query
US10007957B2 (en) Selecting search results for responding to search query
JP6129953B2 (ja) 旅行関連検索結果の分類およびランク付け
US20130290324A1 (en) Categorizing and ranking travel-related database query results
US20190228346A1 (en) Computerized Travel Itinerary Recommendation Tool and Method
US20190147468A1 (en) Location evaluation
WO2019023361A1 (fr) Systèmes et procédés de mise en correspondance par machine de stock d'hébergement à partir de systèmes disparates de prestataires de réservation
CA2851574A1 (fr) Systeme et procede pour classement de candidats
US10678800B2 (en) Recommendation prediction based on preference elicitation
US10740824B2 (en) Product delivery system and method
US20170365014A1 (en) Systems, methods and non-transitory computer readable storage media for tracking and evaluating predictions regarding relationships
WO2016131519A1 (fr) Classement personnalisé pour des résultats de recherche d'une interrogation de base de données basée sur un voyage
KR20170139188A (ko) 상품 정보 검색 공유 서비스 제공 방법
WO2019200296A1 (fr) Systèmes informatiques et procédés pour améliorer la qualité de contenu de pages web
CN115391666A (zh) 医院在线推荐方法、装置、计算机设备和存储介质
EP3059707A1 (fr) Classement personnalisé de résultats de recherche d'une requête de base de données liée aux voyages
US20170140458A1 (en) Method of estimating tenancy duration and mobility in rental properties
Catalano et al. Book allocations in a university library: An evaluation of multiple formulas
EP3082077A1 (fr) Sélection des résultats de recherche permettant de répondre à une interrogation de recherche
EP3540606A1 (fr) Système et procédé de fourniture de produits
AU2016202297B2 (en) Selecting search results for responding to search query
EP2874109A1 (fr) Moteur de recherche permettant d'identifier des propositions de voyage d'affaire
KR20150057983A (ko) 비지니스 여행 제안을 식별하는 검색 엔진
US20220309469A1 (en) Comparing job seekers and jobs by parameterizing both job descriptions and job seeker descriptions to a common set of parameters
KR102150634B1 (ko) 상품정보통합관리시스템 및 그 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16700317

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2975794

Country of ref document: CA

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2016222066

Country of ref document: AU

Date of ref document: 20160105

Kind code of ref document: A

122 Ep: pct application non-entry in european phase

Ref document number: 16700317

Country of ref document: EP

Kind code of ref document: A1