CN107251022A - Personalized ranking for the search result of the data base querying for correlation of travelling - Google Patents
Personalized ranking for the search result of the data base querying for correlation of travelling Download PDFInfo
- Publication number
- CN107251022A CN107251022A CN201680010605.6A CN201680010605A CN107251022A CN 107251022 A CN107251022 A CN 107251022A CN 201680010605 A CN201680010605 A CN 201680010605A CN 107251022 A CN107251022 A CN 107251022A
- Authority
- CN
- China
- Prior art keywords
- itinerary
- profile
- traveller
- value
- subset
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/14—Travel agencies
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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 OR CALCULATING; 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 OR CALCULATING; 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]
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)
- Accounting & Taxation (AREA)
- Databases & Information Systems (AREA)
- Development Economics (AREA)
- Human Resources & Organizations (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Method, system and computer program product for processing data library inquiry result.Based on the search condition provided by traveller from itinerary database retrieval Query Result, and the subset of itinerary is selected based on composite profile from search result.Traveller can recognize that the result of which of the subset of their preferences itinerary and this selection is used to update the probability profile of traveller.Search result can use the ranking again of the composite profile after updating, and based on the new subset of the selection itinerary of ranking again.Traveller can indicate which of new subset itinerary is preferred, and based on this update probability profile.Repeating the process of placement search result and update probability profile can repeat, until successive ranking result set matches each other.
Description
Technical field
Present invention relates generally to computer and computer software, more particularly to related for travelling based on user profiles
Database search result assigns method, system and the computer program product of ranking.
Background technology
Reservation (reservation) system for the service (such as flight) for correlation of travelling generally includes one or more
The database of the storage data related to the travel service provided.These databases can include comprising for determining service valency
The fare data storehouse of the data of lattice, and comprising for determine service whether the availability databases of available data.Subscribe system
System, which can also provide permission end user (such as traveller and travel agency) and search for and make a reservation for (book) satisfaction, expects route condition
Travel service instrument.Therefore, the instrument provided can include enabling other systems from database access data
One or more machine interfaces.Global distribution system (GDS) can be included by providing the system accessed by these machine interfaces, with
And the system runed by travel agency or other resellers.
For reserving travel service, traveller can access related by the travelling of the indirect dealer offer of travel service
Website.As a part for prior defined procedure, traveller can provide search condition (such as starting point, destination, travel dates, pre-
Determine grade etc.), and use the site promoter search inquiry.In response to receiving search inquiry, by one of system above-mentioned
The search engine of offer can be from database retrieval data, and generate the search knot for the itinerary for including meeting search condition
Really.The quality of the search result provided by the related website of travelling can be by an online dealer of Related product of travelling
The deciding factor that (such as online travel agency or agency) is distinguished with another online dealer.Therefore, search online
The traveller of the related products ＆ services (flight, hotel or the automobile leasing such as fixed a price) of rope travelling can be based primarily upon search
As a result quality and like an online dealer more than another.
The travelling search result provided by the website of online dealer generally display is matched with the search inquiry of traveller
A large amount of travelling suggestions.Therefore, such substantial amounts of travelling suggestion that traveller can be searched embodying in result is flooded.This can be hindered
Traveller's selection travelling suggestion is made a reservation for, and traveller is likely difficult to identify that they like from a large amount of similar search results
A search result.This consequence that can not select search result can be traveller's delay purchase.Traveller can also
Seek another dealer to obtain desired travel service.Thus, providing excessive search result to traveller can reduce often
The secondary purchase number of times checked or " conversion (conversion) rate " of website.Excessive travelling suggestion Ye Huishi trips are provided to traveller
Passerby is difficult to find that they think maximally related search result.This can negatively affect satisfaction of the client to online dealer,
This may cause traveller using a different online dealer to buy travel products.
Accordingly, it would be desirable to for handle traveller inquire about and show the improvement system of the related search result of travelling, method and
Computer program product, it helps traveller to select search result from a large amount of search results for matching its inquiry.
The content of the invention
There is provided a kind of method that ranking is carried out to search result in an embodiment of the present invention.This method includes receiving
Including the search result of the multiple itineraries matched with search inquiry.Can the probability profile based on traveller from multiple travellings
The first itinerary and the second itinerary are selected in stroke.This method can like the first itinerary to be more than based on traveller
The instruction of second itinerary updates the probability profile of traveller.This method may then based on probability profile to determine travelling
The composite profile of person, and based on composite profile come to search result progress ranking.
There is provided a kind of device for being used to carry out search result ranking in another embodiment of the present invention.The dress
Put including processor and the memory for being coupled to processor.Memory includes program code, when program code is by computing device
When, device is received the search result for the multiple itineraries for including matching with search inquiry, and the probability based on traveller
Profile selects the first itinerary and the second itinerary from multiple itineraries.The device can receive traveller and like
One itinerary is more than the instruction of the second itinerary, and updates the probability profile of traveller based on the instruction.The device
The composite profile of traveller can be determined based on probability profile, and based on composite profile come to search result progress ranking.
Include the calculating of non-transient computer readable storage medium storing program for executing there is provided a kind of in another embodiment of the present invention
Machine program product, medium includes program code.Program code can be configured as receiving processor when by computing device
Including the search result of the multiple itineraries matched with search inquiry, and the probability profile based on traveller is from multiple travellings
The first itinerary and the second itinerary are selected in stroke.Processor can receive traveller and like the first itinerary to be more than
The instruction of second itinerary, and the probability profile of traveller is updated based on the instruction.Processor can be based on probability letter
Shelves determine the composite profile of traveller, and based on composite profile come to search result progress ranking.
Brief description of the drawings
Be attached in this specification and constitute part thereof of accompanying drawing illustrate various embodiments of the present invention also, together with
The detailed description of general description of the invention and example given below given above, for explaining the implementation of the present invention
Example.
Fig. 1 is the schematic diagram of the Illustrative Operating Environments of multiple computing systems for including being communicated via network.
Fig. 2 is the schematic diagram of Fig. 1 exemplary computer system.
Fig. 3 is showing for the computing system of the website, search engine, run-length data storehouse and the profiles database that include travelling correlation
It is intended to.
Fig. 4 is the schematic diagram of the probability profile used by the search engine in Fig. 3.
Fig. 5 is that diagram can be performed the process for being shown to traveller to select search result to be used for by Fig. 3 computing system
Flow chart.
Fig. 6 is to show to be performed the personal preference so that probability profile is suitable to reflection traveller by Fig. 3 computing system
Process flow chart.
Embodiment
Embodiments of the invention can be realized by processing and Database Systems (reservation system of such as computerization).Place
Reason and Database Systems can be configured as by recognizing the trip that meets the itinerary of travelling search inquiry and will identify that
Every trade journey responds travelling search inquiry as search result return.In the case where travelling by air, processing and Database Systems
In the global distribution system (GDS) that itinerary of the identification including the flight from Duo Jia airlines can be embodied in.GDS is also
It can provide including service (such as rail travel, hotel room, automobile leasing, sightseeing and other activities related to travelling)
Itinerary.
Processing and Database Systems can include the search engine for receiving search inquiry.Search inquiry can be derived from and visit
The traveller of the website of the indirect dealer (such as travel agency) of travel service is asked, and the number of definition search condition can be included
According to (starting point, destination, travel dates, predetermined grade etc.).In response to receiving search inquiry, search engine can be used
Data in one or more databases are come the itinerary retrieved or otherwise generation meets search condition.Database can
To include the database of the cache with the itinerary precalculated.Database can also include providing from it generating trip
The service availability of the real time data of every trade journey and fare data storehouse.
Search engine can carry out ranking according to the distinctive composite profile of traveller to search result.Then, search engine
The set of one or more itineraries can be selected from search result based on this ranking, and selected result is shown
To traveller.Thus, search engine can provide the itinerary of recommendation according to the personal preference of traveller for traveller.With lacking
The system of this few feature is compared, by selecting the set of itinerary to be used to be shown to traveller, this hair from search result
Bright embodiment can reduce and be sent to traveller, be stored in the cache of search result or database or must be by
The quantity for the search result that reason and Database Systems are handled.Thus, embodiments of the invention can reduce processing and data base set
The processing load of system simultaneously improves its performance.
Composite profile can be determined based on the probability profile of traveller.In order to train probability profile to match traveller's
Personal preference, search engine can select the subset of two or more itineraries based on probability profile from search result,
And ask traveller to indicate which itinerary is preferred.Preference is indicated between itinerary in response to traveller, is searched
Index is held up can be with update probability profile, to optimize the identification of selected itinerary by composite profile.Then, search is drawn
The new subset and repetition training process of itinerary can be selected based on the probability profile of renewal by holding up.The process can repeat many
It is secondary, until search engine can reliably predict user preference.
By capturing the personal preference of traveller and carrying out ranking to search result using these preferences, search engine can
There is higher value to traveller to accurately determine which search result.Then these search results can be shown to trip
Passerby so that provide the itinerary matched with its people's preference for traveller, without being divided substantial amounts of search result
Class.Thus, embodiments of the invention can enable traveller avoid assessing excessive amounts of itinerary preferred to find
Selection.
Referring now to Figure 1, operating environment 10 can include GDS 12, one or more trips according to an embodiment of the invention
Row service provider system (such as business system 14), one or more indirect dealer system (such as travel agency's systems
, and traveller's system 18 16).Business system 14 can with one or more databases (such as availability databases 20 and
Fare data storehouse 22) communication.Each in GDS 12, business system 14, travel agency's system 16 and traveller's system 18 can
To be communicated by network 24.Network 24 can include allowing for the one or more special or public of data exchange
Network (for example, internet).
GDS 12 can be configured as by enabling travel agency, checking operator or other indirect dealers via GDS 12
Make a reservation for (registration) (book) on business system 14 and subscribe logical between business system 14 and travel agency's system 16 to promote
Letter.GDS 12 can safeguard the link that multiple business systems are arrived via network 24, and this enables GDS self-validations in 12 future
The reservation request of operator or travel agency is routed to the operator of corresponding operation.Thus, business system 14 and system of travel agency
System 16 can via the singular association to GDS 12 in multiple airlines book flight.
Business system 14 can include computer reservation system (CRS), and it enables GDS 12 or travel agency's system 16
Enough subscribe and pay air ticket.Therefore, business system 14 can be interacted with availability databases 20 and fare data storehouse 22,
So that travel service is fixed a price and subscribed in response to the predetermined request from GDS 12 or other inquiries.Business system 14 can be with
Interacted with other business system (not shown) directly or through GDS 12 so that checking operator can sell by
The ticket at the seat that the operator of operation provides.Then, the operator of operation can open for the service provided to checking operator
Bill.
Travel agency's system 16 can provide the interface for accessing GDS 12 to travel agency, and the interface enables travel agency
Search and reserving travel stroke.Travel agency's system 16 can also include the addressable application of traveller's system 18, and the application causes
Traveller can search for and reserving travel stroke with the help of no travel agency.This application can include for example using by trip
The web browser that passerby's system 18 is provided is through the related website of the addressable travelling of network 24.
Traveller's system 18 can include desktop computer, laptop computer, tablet PC, smart phone or any
Other suitable computing devices.Traveller can use traveller's system 18 by accessing GDS 12, operation quotient system through network 24
System 14, travel agency's system 16 or any other suitable system are searched for and reserving travel service.For example, traveller can start
Browser application, and using browser application in the website provided by travel agency's system 16 or by GDS 12, business system
Travel service is searched on the website that 14 or any other suitable systems are provided.Traveller can be by inputting net by payment information
Stand to make a reservation for selected travel service.
Referring now to Figure 2, can be in one or more computer equipments or system (such as exemplary computer system 30)
On realize the GDS 12 of operating environment 10, business system 14, travel agency's system 16, traveller's system 18, availability databases
20 and fare data storehouse 22.Computer system 30 can include processor 32, memory 34, mass storage memory equipment
36th, input/output (I/O) interface 38 and man-machine interface (HMI) 40.Computer system 30 can also be via network 24 or I/O
Interface 38 is operatively coupled to one or more external resources 42.External resource can include but is not limited to server, data
Storehouse, mass-memory unit, ancillary equipment, the network service based on cloud or can be used by computer system 30 it is any other
Suitable computer resource.
Processor 32 can include being selected from microprocessor, microcontroller, digital signal processor, microcomputer, centre
Manage unit, field programmable gate array, programmable logic device, state machine, logic circuit, analog circuit, digital circuit or
One or many of any other equipment of (analog or digital) signal is manipulated based on the operational order in memory 34 is stored in
Individual equipment.Memory 34 can include single memory equipment or multiple memory devices, including but not limited to read-only storage
(ROM), random access memory (RAM), volatile memory, nonvolatile memory, static RAM
(SRAM), dynamic random access memory (DRAM), flash memory, cache memory or times for being capable of storage information
What miscellaneous equipment.Mass storage memory equipment 36 can include data storage device, such as hard disk drive, optical drive
Device, tape drive, volatibility or nonvolatile solid state equipment or any other equipment for being capable of storage information.
Processor 32 can be operated under the control of the operating system 46 in residing in memory 34.Operating system 46 can be with
Manage computer resource so that be presented as (the application such as resided in memory 34 of one or more computer software applications
48) computer program code can have the instruction performed by processor 32.In alternative embodiments, processor 32 can be straight
Connect execution and apply 48, operating system 46 can be omitted in this case.One or more data structures 50 can also be resided in
In memory 34, and it can be used for storing or manipulating data by processor 32, operating system 46 or using 48.
I/O interfaces 38 can provide by processor 32 be operatively coupled to miscellaneous equipment and system (such as network 24 or
External resource 42) machine interface.Thus, can be by being communicated via I/O interfaces 38 with network 24 or outside using 48
Portion's resource 42 cooperates, to provide the various features for being constituted the embodiment of the present invention, function, using, process or module.Using 48
There can also be the program code performed by one or more external resources 42, or otherwise rely on by computer system 30
Function or signal that outside other systems or networking component are provided.In fact, giving almost endless hardware and software configuration
Be it is possible, those skilled in the art will appreciate that embodiments of the invention can include be located at computer system 30 outside,
It is distributed between multiple computers or other external resources 42 or by as the service (such as cloud computing service) through network 24
The application that the computing resource (hardware and software) of offer is provided.
HMI 40 can be operatively coupled to the processor 32 of computer system 30 in known manner, to allow user
Directly interacted with computer system 30.HMI 40 can include video or aplhanumeric visual display, touch-screen, loudspeaker and energy
Enough provide a user any other suitable audio and visual detector of data.HMI 40 can also include receiving to come from
The order or input of user and the input equipment and control (such as alphanumeric key that the input of input is sent to processor 32
Disk, sensing equipment, keypad, button, control handle, microphone etc.).
Database 44 may reside within mass storage memory equipment 36, and can be used to collect and organize by
The data that various systems described herein and module are used.Database 44 can include data and storage and the branch for organizing data
Hold data structure.Especially, database 44 can be arranged to any data base organization or structure, including but not limited to close
It is database, hierarchical data base, network data base or its combination.It can be performed using type of service as instruction on processor 32
The data base management system of computer software application access the letter being stored in the record of database 44 in response to inquiry
Breath or data, wherein inquiry can be dynamically determined and performed by operating system 46, other application 48 or one or more modules.
Referring now to Figure 3, in an exemplary embodiment of the present invention, computer system 60 can provide search engine 62,
Run-length data storehouse 64 and profiles database 66.Computer system 60 can be related with travelling to business system 14 website 68 lead to
Letter.In an embodiment of the present invention, computer system 60 can be provided by GDS 12, and the website 68 for correlation of travelling can be by
Travel agency's system 16 is provided.But, it will be appreciated by the skilled addressee that the website 68 related to travelling of computer system 60
It can be provided by any suitable computer system.
Traveller can use traveller's system 18 to access the one or more webpages provided by the related website 68 of travelling.
Traveller can interact with webpage, to search for and reserving travel stroke.These itineraries can include one or more
Travel service, such as flight, hotel room, automobile leasing or any other travel service.Therefore, traveller's system 18 can be with
Search inquiry 70 is sent to the related website 68 of travelling.Search inquiry 70 can include the data for defining search condition, such as
Starting point and destination, travel dates and/or predetermined grade.In response to receiving search inquiry 70, the related website 68 of travelling can
So that the search inquiry 72 of the data including defining search condition and traveller's identity is sent into search engine 62.
In response to receiving search inquiry 72, search engine 62 can generate database search queries 74 and send it to
Run-length data storehouse 64.In response to receiving database search queries 74, run-length data storehouse 64, which can be recognized, meets search condition
One or more itineraries.The itinerary identified can include storage reserving travel stroke in the caches or
The itinerary of the new determination retrieved from business system 14.The itinerary newly determined can be according to from availability databases
20 service useful datas (for example, available seat) received and the fare data next life received from fare data storehouse 22
Into.The itinerary newly determined can also be stored in the caches and accessed by run-length data storehouse 64 so that they can be used as
For the reserving travel stroke of the subsequent search queries received by search engine 62.
In response to identification itinerary, run-length data storehouse 64 can send the response for including search result to search engine 62
76, wherein search result includes the itinerary identified.When run-length data storehouse 64 compiles search result, or response
In receiving response 76, search engine 62 can send the inquiry for the data for including defining traveller's identity to profiles database 66
78.Inquiry 78 can ask profiles database 66 to provide user profiles for traveller.If there are user profiles for traveller,
So user profiles can be sent to search engine 62 by profiles database 66 in response 80.If traveller does not have user's letter
Shelves, then default profile can be sent to search engine 62 by profiles database 66, or response 80 can be indicated for identification
User profiles are not present in the traveller gone out.In the latter case, search engine 62 can generate default profile for traveller.
Each user profiles can be associated with the composite profile CP including coefficient sets, wherein each coefficient is gone with travelling
The characteristics of journey, is corresponding.The value of each coefficient can indicate that the corresponding feature of itinerary has more important for traveller.For example, being
One of number can be corresponding with the overall travel time of stroke.In order to provide this feature on monetary cost or including itinerary
Travel products price CMThere is more important instruction for traveller, monetary value can be had by being assigned to the value of the coefficient
(for example, hour of 20 dollars/hourage).
Composite profile CP can be used to determine " totle drilling cost " of itinerary, and it captures the price C of travel productsMAnd
The perception value added or subtracted for traveller by the special characteristic of itinerary.Use the exemplary system of 20 dollars/hour above
Count to characterize the cost of hourage, can have with the overall travel time itinerary of 4 hours and be added to itinerary
Price CMComposite profile CP=$ 80, to reflect totle drilling cost C to the traveller that is discussedTOT=CM+CP.By contrast, phase is connected
Can have with the itinerary that starting point and destination but overall travel time are 8 hours and be added to itinerary price CMCP
=$ 160.Therefore, in the above example, it is assumed that all other feature of itinerary is all identical, with 4 hour hourage
Itinerary price CMBy must than the itinerary with 8 hour hourage price CMIt is big more than 80 dollars, so as to right
There is higher totle drilling cost C for travellerTOT.Thus, composite profile CP can provide the trip for determining itinerary to being discussed
The means of the overall property expected of passerby, it not only captures the price for each travel products for constituting stroke but also capture convenience factor is to travelling
The influence of person.
Once composite profile CP, it is known that can determine each to search in response 80 using composite profile CP once search engine 62
The totle drilling cost C of hitch fruitTOT.Search engine 62 may then based on their totle drilling cost CTOTRanking is carried out to search result, always
Cost CTOTLower search result is given higher ranking.Then, search engine 62 can be sent out to the related website 68 of travelling
Response 82 is sent, the response includes at least a portion of the placement search result selected based on ranking.For example, selected search knot
Fruit can include the search result for representing the predetermined quantity of top ranked search result.In response to receiving response 82, phase of travelling
The website 68 of pass can format result and send the response 84 for being directed to search inquiry 70, and the response is included by search engine 62
Select to be shown to the search result of traveller.Response 84 can include causing that for example runs in traveller's system 18 to browse
The data of window of the device application display filled with selected search result.
Fig. 4 is depicted including N number of value profile VP1-VPNArray probability profile 90 schematic diagram.Probability profile 90 it is every
Individual value profile VP1-VPNProbability or weight w can be included1-wN(by each value profile VP1-VPNRepresented along the height of vertical axis).
That is, probability profile 90 can be by array [w1×VP1, w2×VP2... wN×VPN] represent.Each value profile of probability profile 90
VP1-VPNSet (the v of M itinerary feature can also be included1, v2... vM).Weight w1-wNIt can provide traveller's
Corresponding value profile VP is provided in composite profile CP mixed function1-VPNRelative efficacy.Therefore, being each worth profile VP1-VPN
Weight w1-wNCan be based on corresponding value profile VP1-VPNIdentified how well from the subset of itinerary for being discussed
Traveller for have minimum totle drilling cost CTOTItinerary determine.The subset of itinerary can for example including selected from
Meet two or more itineraries of the more grand tour stroke set of searching request.
It is worth each coefficient v of profile1-vMThe value of corresponding itinerary feature can be described.As an example, coefficient v1-vM
It can include and following corresponding coefficient：Cost/hour for hourage is (for example, increase by 20 per hour of hourage
Dollar), the cost/stand (for example, each terminal increases by 30 dollars) for each required station or connection, for overnight
It is the cost at each station of transfer demand/stand (for example, transfer overnight increase by 100 dollars), related to the operator of undesirable type
Connection cost (if for example, operator is Acme Air, then increase by 40 dollars), it is associated with the operator of undesirable type
Cost (if for example, operator is low cost operation business, then increase by 50 dollars), it is related to specific airline operator's policy
The cost of connection is (if for example, operator only allows a personal luggage or with the record excessively postponed, then increase by 40 is beautiful
Member), or if the fringe cost outside preferred hourage before 8 a.m. (for example, for setting out the departure time
Flight, increase by 30 dollars), this is only to enumerate several examples.
Consider that all possible wish level that all conditions are paid can be caused unpractical value letter for management
The quantity N of shelves.In an embodiment of the present invention, in order to which the quantity of value profile is maintained at into manageable level, each coefficient
Value can be quantified as centrifugal pump.Therefore, the cost associated with each coefficient, which can be quantified as having, exceedes preset range
The value of (for example, $ 0 to $ 100) fixed increment (for example, $ 10).As an example, by the value profile constituted with following corresponding coefficient
N=6 × 11 will be caused4=87846 value profiles：There is the value set of the increments of $ 10 for being mapped in $ 0 to $ 50 scope
Hourage cost/hour, and cost/stand for transfer overnight, be classified as undesirable operation for selection
The cost penalty of business, for do not expect type operator cost and the cost for the undesirable departure time, wherein
It is each to be mapped to the value set in 0 to 100 scope with 10 increments.
Probability profile can be used to drive the search to itinerary, and selected from the search result by search output
Top ranked search result.Probability profile can also be used to select from the output of the impersonal theory search for itinerary
Select the maximally related suggestion of matching traveller's search terms.In any case, the composite profile of traveller can be from value profile
VP1-VPNOne or more of construction.For example, can be by taking weighted value profile VP as follows1-VPNSummation come determine travelling
The composite profile CP of person：
Wherein weight w1-wNIt is normalized such that：
As a result the composite profile CP of gained can provide the set (v of M itinerary feature for traveller1, v2,
...vM), wherein composite profile CP each itinerary feature (v1, v2... vM) include each value profile VP1-VPNCorrespondence
Stroke feature (v1, v2... vM) weighted sum.
Referring now to Figure 5, flow chart depict can be provided by search engine 62 or by computing system 60 some its
The process 100 that its proper modules is performed, to determine each value profile VP for being constituted the probability profile 90 for specific traveller1-
VPNWeight w1-wN。
In square frame 102, process 100 can obtain search result, and search result includes the search with being provided by traveller
Multiple itineraries of set of circumstances matching.These search results can be for example in response to database search queries 74 be sent to
Run-length data storehouse 64 and searched engine 62 is received, as described above with reference to Figure 3.Then, the side of may be advanced to of process 100
Frame 104 and determine traveller whether have pre-existing probability profile 90.If such as traveller is under former occasion
The related website of used travelling, then traveller can have pre-existing probability profile 90.Return traveller can be based on
The data structure (for example, HTTP cookie) for the identification traveller being stored in traveller's system 18, it is supplied to travelling related
The user identity or account number of website 68 or any other suitable method and be identified.
If traveller is associated with pre-existing probability profile 90 (the "Yes" branch of decision block 104), then mistake
Journey 100 may be advanced to square frame 106 and retrieve probability profile 90 from data storage location (such as profiles database 66).If
Traveller is not associated with pre-existing probability profile 90, for example, traveller is first time user's (decision block of system
104 "No" branch), then process 100 may be advanced to square frame 108 and retrieve default probability profile 90.The present invention's
In embodiment, default probability profile 90 can be wherein weight w1-wNProbability profile 90 with equal value.This can cause in value
The unified probable value of all conditions used in profile.
As an example, in an embodiment of the present invention, coefficient v150 it can be carried from $ 0 to $ so that $ 10 is increment for travelling
For cost/hour, coefficient v2100 the cost often stood, coefficient v can be provided for increment with $ 10 from $ 0 to $3Can be 100 from $ 0 to $
Using $ 10 cost of the operator as low cost operation business, coefficient v are provided for increment4100 it can be provided from $ 0 to $ with $ 10 for increment
Using the cost of non-preferential operator, and coefficient v5It can provide and be included in the trip segment set out between not preferred working days time
Cost.These costs can repeatedly be added for specific stroke.For example, can be the itinerary with the undesirable departure time
The cost that starts early of each leg addition, for example, each addition in the inbound of stroke and outbound leg is once.
Then, process 100 may be advanced to square frame 110, and to traveller provide the travelling preference on them the problem of.
Exemplary travelling preference can include preferred airline, hourage, non-preferential operator or can be used to side
Help the preferred itinerary of identification or determine totle drilling cost CTOTAny other travelling preference.Then process 100 can be generated including referring to
Show the user profiles of the preference of traveller and the data of probability profile, and this user profiles is stored in profiles database 66
In case using in the future.In response to obtaining pre-existing or acquiescence probability profile, process 100 may be advanced to square frame 112.
In square frame 112, process 100 can select search result set.Therefore, process 100 can give birth to from probability profile
Into composite profile.Then, process 100 can use composite profile to carry out ranking to search result, and based on search result
Ranking selects search result set.For example, search result set can include representing top ranked search in search result
As a result a part.
In response to selection search result set, process 100 may be advanced to square frame 114, and from search result set
The subset of itinerary is selected (for example, a pair of itinerary TI1、TI2).Then, process 100 can ask traveller from subset
One itinerary of middle selection.Itinerary can be selected based on the composite profile defined by probability profile 90 so that subset
In each itinerary maximum information on composite profile is provided.In alternative embodiments of the present invention, it can be subset
(for example, four itinerary TI1、TI2、TI3、TI4) the more than two itineraries of selection., can in this alternate embodiment
To require traveller by the itinerary in subset from most preferably to worst carry out ranking, rather than simply choose one of work
For most preferred itinerary.The instruction of which itinerary of traveller's preference is received in response to process 100, process can be former
Enter square frame 116.
In square frame 116, process 100 can update the probability letter of traveller based on the selected itinerary of traveller
Shelves 90.As an example, the subset for wherein itinerary includes two itinerary TI1To TI2Embodiment, traveller choosing
Select itinerary TI1Rather than TI2Itinerary TI can be indicated1Totle drilling cost CToTLess than itinerary TI2Totle drilling cost
CToT.I.e.：
CTOT(TI1) ＜ CTOT(TI2)
Use above-mentioned example property coefficient VP (v1, v2, v3, v4, v5) this might mean that：
(CM2-CM1)+v1×(t2-t1)+v2×(S2-S1)+v3×(NLCC2-NLCC1)
+v4×(NUDC2-NUDC1)+v5×(NUTs2-NUTS1) ＞ 0
Wherein tnRepresent hourage, SnRepresent transfer number of times, NLCCnRepresent the leg time flown in low cost operation business
Number, NUDCnRepresent the leg number of times flown in desired operator, and NUTSnRepresent to be used for corresponding itinerary TI1、TI2
It is undesirable when bay for example, early start time) quantity.
Process 100 can use this information to update traveller's composite profile CP.Therefore, process 100 can be based on trip
The entirely sensible hypothesis of passerby or traveller not exclusively reasonably assume to carry out update probability profile 90.In complete rational model
Under, process 100 can be only using generation and traveller's selection itinerary TI when generating the value profile of traveller1Rather than TI2
The value profile VP for the result being consistent1-VPN.That is, the value profile VP incompatible with the selection of travelleriCan be excluded it is determined that
Outside traveller's composite profile CP.Therefore, each value letter for the result that the itinerary selection with traveller is matched can not be produced
Shelves VPiWeight wiIt can be configured to 0.Produce the value profile VP for the result that the itinerary selection with traveller is matchediPower
Weight wiThen can be normalized such that remaining non-zero weight and be one (unity).
In some cases, traveller can be selected with selecting inconsistent itinerary before traveller.That is, travel
The selection of person can seem unreasonable.This be probably because traveller made mistakes, or due to route some aspect
(for example, traveller is predetermined personal rather than travel for commercial purpose) different from normal parameter.In order to adapt to may be with traveller reality
Border or the inconsistent selection of usual preference, process 100 can use " unreasonable model " to update the value profile of traveller.Not
Under Reasonable Mode, with the selection of traveller the weight of consistent each value profile can from the value increase before it, and with travelling
The weight for selecting inconsistent each value profile of person can be reduced from value before.The amount of weight changes can be by for determining
" reasonable property coefficient " r for being multiplied by the multiplier of previous weight is controlled.Under this model, it can not will can produce with traveller's
Each value profile VP of the result of itinerary selection matchingiRenewal weight wiIt is set to (1-r) × wi.So that：
wi(updated)=(1-r) × wi
On the contrary, each value profile VP for the result that the itinerary selection produced with traveller can be matchedjRenewal
Weight wjIt is set to：
Thus, the weight w of renewali、wjIt may rely on previous weight wi、wj.For reasonable property coefficient r=100%,
Unreasonable model can be provided and rational model identical result.This model may need the selection for considering to recommend one by one, with
Update weight.Under any model, each coefficient of value profile can be provided by following formula：
Once probability profile 90 and composite profile CP have been updated, process 100 just may be advanced to square frame 118 and using more
New composite profile CP carries out ranking again to search result.Then, process 100 may be advanced to square frame 120 and determine weight
Whether the search result before new ranking matches the search result after ranking.If with search afterwards before ranking again
As a result there is no difference between, or have differences but be considered as inappreciable difference, then search result may be considered that
It is matching.If for example, the rank order or one part of search result are (for example, the ranking result in search result set
Order) change due to ranking again, then can have differences.Whether difference is considered as that how many knot significantly may rely on
Fruit is changed, or the ranking dependent on the result changed.For example, the order change of relatively low ranking result can be
Inappreciable difference.
If search result mismatches (the "No" branch of decision block 120), then process 100 may be advanced to square frame
112, and the composite profile CP based on renewal selects new search result set.If search result matches (decision block 118
"Yes" branch), then process 100 may be advanced to square frame 122 and show the search result of multiple top rankeds.Process 100
The user profiles of traveller can also be updated according to the probability profile of renewal so that receive search from traveller next time and look into
The probability profile that the retrieval of process 100 updates during inquiry.
Referring now to Figure 6, flow chart depict can be provided by search engine 62 or by computing system 60 some its
Its proper modules is come the process 130 that performs, so which itinerary determined for itinerary Resource selection.Although for letter
Exemplary embodiment uses the set of two itineraries for the sake of list, but it will be appreciated by the skilled addressee that can also
Using including the set of two or more itinerary.In square frame 132, process 130 can select a pair of trips from search result
Every trade journey.Can be based on itinerary the characteristics of, selects initial right.For example, process can be based on into original selection travelling row
Journey, since the minimum search result of cost.But, the invention is not restricted to any other tool based on cost or itinerary
The selection of body feature or combination of features.
Once having selected for a pair of itineraries, process 130 just may be advanced to square frame 134 and determine selected
Whether one of itinerary of centering " dominates " another itinerary.If each condition of an itinerary with it is another
The respective conditions of one itinerary are compared to all with preferred value, then can be considered as this itinerary leading described another
One itinerary.For example, compared with another itinerary, one of itinerary is less expensive, faster, with less
Transfer, in preferred operator, and with the preferred departure time.If one of itinerary is occupied an leading position
(the "Yes" branch of decision block 134), then process may be advanced to square frame 136.If any one itinerary of a centering
Another ("No" branch of decision block 134) is not dominated, then process 130 may be advanced to square frame 138.
In square frame 138, process 130 can from the probability profile of traveller selective value profile VPi, and using selected
The value profile VP selectediTo determine the totle drilling cost C of each itinerary in this pair of itineraryTOT.In embodiments of the invention
In, process 130 can be according to the value profile VP with the coefficient value in the centre for allowing coefficient value scopeiGive time of preference
Sequence carrys out selective value profile VPi.Using close to middle value profile VPiProcess 130 can be allowed than by random by starting selection course
Selective value profile VPiQuickly select suitable itinerary test pair.It is worth profile VPiSelection be based on it is assumed hereinafter that：Each with
The value of the characteristics of itinerary (that is, preference of the traveller to the actual conditions of itinerary) corresponding coefficient, which is typically distributed on, is
In the allowed band of numerical value.
When having selected for value profile VPiWhen, process 130 may be advanced to square frame 140 and based on which itinerary
It is identified as by being worth profile with minimum totle drilling cost CTOTWill value profile VPiIt is added to value profile VPiOne of two subsets.Cause
This, value profile VPiA subset can use and produce minimum totle drilling cost C for one of itinerary of itinerary centeringTOT
Value profile VPiTo fill, and it is worth profile VPiAnother subset can use for the itinerary centering another travel
Stroke produces minimum totle drilling cost CTOTValue profile VPiFilling.
In an embodiment of the present invention, the additional value profile VP from probability profileiCan be based on the relative of its coefficient
Value is added to value profile VPiCorresponding subset.Therefore, and as an example, for itinerary TI1And TI2Pair, it is foregoing
Equation：
(CM2-CM1)+v1×(t2-t1)+v2×(S2-S1)+v3×(NLCC2-NLCC1)
+v4×(NUDC2-NUDC1)+v5×(NUTS2-NUTS1) ＞ 0
Indicated value profile VPiBy itinerary TI1It is identified as with relatively low totle drilling cost CTOT.In this exemplary cases
Under, if (t2-t1) for just, then, if other coefficient v2、v3、v4、v5Value and currency profile VPiValue it is equal, then mistake
Journey 130, which can determine to have, is more than currency profile VPiCoefficient viValue coefficient viAll other value profile VPiAlso will
Itinerary TI1It is identified as preferred itinerary.If on the contrary, (t2-t1) it is negative, then, if other coefficient v2、v3、v4、
v5Value and currency profile VPiValue it is equal, then process 130 can determine have be less than currency profile VPiCoefficient vi's
The coefficient v of valueiAll other value profile VPiAlso by itinerary TI1It is identified as preferred itinerary.In other words, value is worked as
Profile VPiDue to a feature (the characteristics of being more precisely due to and there is the itinerary of occurrence pair of itinerary
The value profile VP answerediCoefficient) and when being added to a subset of value profile, if with it is examined the characteristics of it is corresponding additional
Value profile corresponding coefficient have indicate be used for be examined feature traveller's preference more notable rank set-point so that
It is clear that the additional value profile of all that, which will also be added to same subsets, then with except an examined coefficient/
Outside feature and VPiThe identical coefficient/feature of coefficient/feature all additional value profile VPjIt is added to identical
Subset, without calculating them for itinerary to TI1And TI2Corresponding CTOT.Add accordingly, in response to the first subset
Plus with the first coefficient and at least one have corresponding set-point the second coefficient the first value profile, from the attached of probability profile
Plus second value profile can be added to the first subset of value profile, the additional second value profile have it is corresponding at least one
The value of the identical value of second coefficient and corresponding first coefficient --- dependent on the every of the itinerary pair weighted by coefficient
The characteristics of individual itinerary --- the value of the first coefficient with more than or less than the first value profile.This principle can be expanded to
Other coefficients, to reduce the iterations that process 130 has to carry out.Thus, identification itinerary TI1Or TI2Value profile VPi
If can be than by assessing each value profile VP respectivelyiIt is possible to be quickly determined.
In response to the value profile VP that will be addediIt is added to one of subset, process 130 may be advanced to square frame 142, and
And whether each subset of determination value profile has the value profile VP of at least minimum numberi.Can be based on the letter of each subset intermediate value
Shelves VPiAccumulation weight determine value profile VPiEach subset intermediate value profile VPiQuantity.Accumulation weight can for example pass through
To each value profile VP in evaluated subsetiWeight wiSum to determine.
Process 130 may then advance to square frame 144 and determination value profile VPiEach subset accumulation weight whether
More than minimum or threshold value the 40% of the accumulation weight of all values profile (for example, in probability profile).If each subset has
Have more than the accumulation weight (the "Yes" branch of decision block 144) of threshold value, then process 130 may be advanced to square frame 146 and
A pair of itineraries are selected as test pair.If the accumulation weight of each subset is no more than the threshold value ("No" of decision block 144
Branch), then process 130 may be advanced to square frame 148.
In square frame 148, process 130 may determine whether all values profile VP of probability profile1-VPNIt is evaluated
One of and be assigned to the subset of value profile.If not assessing all value profile VP also1-VPN(the "No" of decision block 148 point
Branch), then process may be advanced to square frame 150, and select from probability profile before square frame 140 is proceeded to it is next not
Assessed value profile VPi, with repeat assessment process.If have evaluated all value profile VP1-VPN(decision block 148
"Yes" branch), then process 130 may be advanced to square frame 136.
In square frame 136, process 130 may determine whether all possible itinerary to being evaluated.That is, search
With the presence or absence of any unique combination for the itinerary not being evaluated also in hitch fruit.If not all possible travelling row
Journey is to being evaluated (the "No" branch of decision block 136), then process 130 may be advanced to square frame 152 and advance
Lower a pair of itineraries are selected before beginning the evaluation at process to square frame 134.
If have evaluated all possible itinerary to (the "Yes" branch of decision block 136), then process 130
It may be advanced to square frame 154.In square frame 154, which process 130 can determine to itinerary in value profile VPiIt is each
Most uniform segmentation is produced between the accumulation weight of subset.That is, process 130 can determine which is tired at it to itinerary generation
There is the value profile VP of minimum difference between product weightiSubset.This can recognize wherein accumulation weight close in probability profile
All values profile VP1-VPNTotal accumulation weight 50% subset.Then, process 130 can select this to itinerary conduct
The subset of itinerary.
In general, being performed so that no matter the routine for realizing embodiments of the invention (is implemented as one of operating system
Point also it is implemented as specific application, component, program, object, module or command sequence, or even its subset) herein all
It can be referred to as " computer program code ", or referred to as " program code ".Program code generally includes computer-readable instruction,
Computer-readable instruction is in the resident various memories and storage device in a computer of different time also, when by computer
One or more of processor read and perform when, make that computer carry out perform embody the embodiment of the present invention each side
Operated necessary to the operation in face and/or element.For the computer-readable program instructions for the operation for performing the embodiment of the present invention
Can be such as assembler language or the source code or object code write with any combination of one or more programming languages.
Various program codes described herein can be recognized based on the application realized in a particular embodiment of the present invention.
It will be understood, however, that, following any specific program nomenclature uses just for the sake of convenient, therefore the present invention
It is not limited to only use in any concrete application for being recognized and/or being implied by this nomenclature.In addition, giving in general
Computer program can be organized into the endless mode of routine, process, method, module, object etc., and can reside in
Distribution program between various software layers (for example, operating system, storehouse, API, using, applet etc.) in normatron
The various modes of function, it is to be appreciated that embodiments of the invention are not limited to specific group of procedure described herein function
Knit and distribute.
The program code embodied in any application/module described herein can in various different formats individually or collectively
It is distributed as program product.Especially, the computer-readable storage medium thereon with computer-readable program instructions can be used
Matter carrys out distribution program code, so that each side of computing device embodiments of the invention.
It is inherently that non-transient computer-readable recording medium can be including for storage information, (such as computer can
Reading instruction, data structure, program module or other data) any method or technique realize volatibility and it is non-volatile and
Removable and immovable tangible medium.Computer-readable recording medium can also include RAM, ROM, erasable programmable only
Read memory (EPROM), Electrically Erasable Read Only Memory (EEPROM), flash memory or other solid-state memory skills
Art, portable optic disk read-only storage (CD-ROM) or other optical memory, cassette, tape, magnetic disk storage or other
Magnetic storage apparatus, or any other medium for storing desired information and being read by computer can be used to.Calculate
Machine readable storage medium storing program for executing be not construed as transient signal in itself (for example, radio wave or other propagation electromagnetic waves, by transmission
The electromagnetic wave of medium (such as waveguide) propagation or the electric signal sent by wire).Computer-readable program instructions can be from
Computer-readable recording medium downloads to computer, another type of programmable data processing unit or another equipment, or
Person downloads to outer computer or External memory equipment via network.
The computer-readable program instructions being stored in computer-readable medium can be used to instruct computer, other classes
The programmable data processing unit or miscellaneous equipment of type work in a specific way so that be stored in the finger in computer-readable medium
Order produces product, and the product includes the finger for function, action and/or the operation specified in implementation process figure, sequence chart and/or block diagram
Order.Computer program instructions can be provided to all-purpose computer, special-purpose computer or other programmable data processing units
One or more processors, to produce machine so that the instruction performed via one or more processors make that one must be performed be
Column count, with the function of being specified in implementation process figure, sequence chart and/or block diagram, action and/or operation.
In some alternative embodiments, function, action and/or the operation specified in flow chart, sequence chart and/or block diagram can
To be as one man reordered with the embodiment of the present invention, serial process and/or concurrent processing.Moreover, flow chart, sequence chart and/
Or any one in block diagram can include the square frame more or less square frames as one man more shown than with the embodiment of the present invention.
Terms used herein is just for the sake of description specific embodiment, without being intended to limitation embodiments of the invention.
As it is used herein, singulative " one " and "the" are also intended to including plural form, unless the context clearly indicates otherwise.
It will be further understood that ought be in this manual in use, term " comprising " specifies the feature, integer, step, operation, element
And/or the presence of part, but it is not excluded for one or more of the other feature, integer, step, operation, element, component and/or its group
The presence or addition of conjunction.In addition, with regard to term "comprising", " having ", " carrying ", " Consists of " or its variant in " specific implementation
For the degree used in mode " or " claims ", these terms are intended in the way of similar to term " comprising " be bag
Capacitive.
Although although illustrating the whole and in considerable detail of the present invention by the description of various embodiments
Describe these embodiments, but applicant being not intended to scope of the following claims is limited or limited in any way
To such details.Those skilled in the art will be readily seen from additional advantage and modification.Therefore, wider side of the invention
Face is not limited to detail, representational apparatus and method and shown or described illustrated examples.Thus, without departing substantially from
In the case of the spirit or scope of the present general inventive concept of the applicant, deviation can be made from these details.
Claims (20)
1. a kind of method that ranking is carried out to search result, methods described includes：
Search inquiry is sent from computer to database；
The first search result is received from database by computer, the first search result includes more than first matched with search inquiry
Itinerary；
The first itinerary and 2nd Brigade are selected from more than first itinerary by probability profile of the computer based on traveller
Every trade journey；
Traveller is received at computer likes the first itinerary to be more than that the first of the second itinerary indicates；
Indicated to update the probability profile of traveller based on first by computer；
The composite profile of traveller is determined based on probability profile by computer；And
Composite profile is based on by computer ranking is carried out to the first search result.
2. the method as described in claim 1, wherein probability profile include multiple value profiles, each value profile has weight, and
And the probability profile based on traveller selects the first itinerary and the second itinerary to include from more than first itinerary：
A pair of itineraries are selected from more than first itinerary；
It is determined that an itinerary in the pair of itinerary to be identified as to the first son of the value profile of preferred itinerary
Collection, and another itinerary in the pair of itinerary is identified as to the second son of the value profile of preferred itinerary
Collection；
Determine the second quantity of the value profile in the first quantity and yield in the second subset of the value profile in the first subset；
The pair of itinerary is appointed as the first itinerary and the second itinerary.
3. method as claimed in claim 2, wherein, exceed minimum in response to the first quantity and the second quantity, perform institute
State a pair of itineraries and be appointed as the first itinerary and the second itinerary.
4. method as claimed in claim 2, wherein, minimum is no more than in response to the first quantity or the second quantity, passes through weight
Final election select the pair of itinerary and determine the first subset and yield in the second subset until both the first quantity and the second quantity all
Higher than minimum or all possible itinerary to having already been selected, the pair of itinerary is appointed as to perform
One itinerary and the second itinerary.
5. method as claimed in claim 4, wherein, in response to all possible itinerary to be chosen and the first quantity and
Second quantity is not above minimum, causes the pair of itinerary of the minimum difference between the first quantity and the second quantity
It is designated as the first itinerary and the second itinerary.
6. the method as any one of claim 2 to 5, its intermediate value profile includes multiple coefficients, these coefficients have quilt
It is quantized into the value of the centrifugal pump with fixed increment on preset range.
7. method as claimed in claim 6, wherein the value profile of the coefficient value of the centre with the allowed band in coefficient value
It is selected first, the subset to determine value profile.
8. the method as any one of claim 2 to 7, wherein selecting the pair of trip from more than first itinerary
Every trade journey includes：
Determine whether in the pair of itinerary itinerary dominates another travelling of the pair of itinerary
Stroke；And
Another itinerary is dominated in response to an itinerary, another pair travelling row is selected from more than first itinerary
Journey.
9. the method as any one of claim 6 to 8, in addition to：
In response to the first subset add with the first coefficient and at least one have corresponding set-point the second coefficient first
It is worth profile, the additional second value profile from probability profile is added to the first subset of value profile, described additional second
Value profile has the identical value of at least one corresponding the second coefficient and dependent on the pair of itinerary weighted by coefficient
Each itinerary the characteristics of with more than or less than first value profile the first coefficient value corresponding first coefficient value.
10. the method as any one of claim 2 to 9, wherein determining that composite profile includes：
Each value profile is multiplied by its weight, to generate the value profile of weighting；And
The value profile of weighting is summed, to generate composite profile.
11. the method as any one of claim 2 to 10, wherein update probability profile includes：
It is determined that the first itinerary to be identified as to the first subset of the value profile of preferred itinerary；
It is determined that the second itinerary to be identified as to the yield in the second subset of the value profile of preferred itinerary；And
It is to update the weight of each value profile in the first subset or in yield in the second subset based on value profile.
12. method as claimed in claim 11, wherein updating the weight of each value profile includes：
If being worth profile in the first subset, then the weight of described value profile is multiplied by into the first multiplier；And
If being worth profile in yield in the second subset, then the weight of described value profile is multiplied by into the second multiplier.
13. the method as described in claim 11 or claim 12, wherein updating the weight of each value profile includes：
If being worth profile in the first subset, then the weight of described value profile is set into the first weight；And
If being worth profile in yield in the second subset, then the weight of described value profile is set into the second weight.
14. the method as any one of claim 11 to 13, in addition to：
Weight to the renewal across the multiple value profile is normalized so that the weight sum of renewal is one.
15. the method as any one of claim 11 to 14, in addition to：
Determine the first accumulation weight of the value profile in the first subset；
Determine the second accumulation weight of the value profile in yield in the second subset；And
The first itinerary and the second itinerary are selected, to cause each of the first accumulation weight and the second accumulation weight to exceed
Threshold value.
16. the method as any one of claim 1 to 15, in addition to：
Based on the probability profile after renewal from the search result of database retrieval second, the second search result includes more than second travellings
Stroke；And
In response to determining to have differences between more than first itinerary and more than second itinerary：
The 3rd itinerary and the 4th itinerary are selected from more than second itinerary；
Receiving traveller likes the 3rd itinerary to be more than that the second of the 4th itinerary indicates；And
Indicate to update the probability profile of traveller based on second.
17. method as claimed in claim 16, in addition to：
More than first itinerary is replaced with more than second itinerary；And
Repeat to retrieve the second search result from database based on the probability profile after renewal, selected from more than second itinerary
Select the 3rd itinerary and the 4th itinerary, reception traveller like the 3rd itinerary to be more than the second of the 4th itinerary
Indicate and indicate to update the probability profile of traveller based on second, until more than first itinerary and more than second travellings
Run length code matching.
18. a kind of device for being used to carry out search result ranking, described device includes：
Processor；And
The memory of processor is coupled to, memory stores data structure, data structure includes itinerary database and program
Code, program code makes described device when being executed by a processor：
Search inquiry is sent to database；
The first search result is received from database, the first search result includes more than the first travelling rows matched with search inquiry
Journey；
Probability profile based on traveller, selects the first itinerary and the second itinerary from more than first itinerary；
Receiving traveller likes the first itinerary to be more than that the first of the second itinerary indicates；
Indicate to update the probability profile of traveller based on first；
The composite profile of traveller is determined based on probability profile；And
Ranking is carried out to the first search result based on composite profile.
19. device as claimed in claim 18, wherein, program code makes described device perform as weighed when being executed by a processor
Profit requires the method any one of 2 to 17.
20. a kind of computer program, makes processor when being executed by a processor：
Search inquiry is sent to itinerary database；
Search result is received from database, search result includes the multiple itineraries matched with search inquiry；
Probability profile based on traveller, selects the first itinerary and the second itinerary from the multiple itinerary；
Receive the instruction that traveller likes the first itinerary to be more than the second itinerary；
The probability profile of traveller is updated based on indicating；
The composite profile of traveller is determined based on probability profile；And
Ranking is carried out to search result based on composite profile.
Applications Claiming Priority (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US14/623,699 | 2015-02-17 | ||
| EP15290036.1A EP3059707A1 (en) | 2015-02-17 | 2015-02-17 | Personalized ranking for search results of a travel-related database query |
| EP15290036.1 | 2015-02-17 | ||
| US14/623,699 US20160239765A1 (en) | 2015-02-17 | 2015-02-17 | Personalized ranking for search results of a travel-related database query |
| PCT/EP2016/000007 WO2016131519A1 (en) | 2015-02-17 | 2016-01-05 | Personalized ranking for search results of a travel-related database query |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107251022A true CN107251022A (en) | 2017-10-13 |
Family
ID=55085639
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201680010605.6A Pending CN107251022A (en) | 2015-02-17 | 2016-01-05 | Personalized ranking for the search result of the data base querying for correlation of travelling |
Country Status (4)
| Country | Link |
|---|---|
| CN (1) | CN107251022A (en) |
| AU (1) | AU2016222066A1 (en) |
| CA (1) | CA2975794A1 (en) |
| WO (1) | WO2016131519A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112925974A (en) * | 2019-12-06 | 2021-06-08 | 艾玛迪斯简易股份公司 | Optimized product determination system |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110322330A (en) * | 2019-07-12 | 2019-10-11 | 四川亨通网智科技有限公司 | Smart travel scenic spot order processing method based on big data |
Citations (10)
| 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 |
| US20050033616A1 (en) * | 2003-08-05 | 2005-02-10 | Ezrez Software, Inc. | Travel management system providing customized travel plan |
| CN101876979A (en) * | 2009-04-28 | 2010-11-03 | 株式会社理光 | Query expansion method and equipment |
| US7836057B1 (en) * | 2001-09-24 | 2010-11-16 | Auguri Corporation | Weighted preference inference system and method |
| US20100305984A1 (en) * | 2009-06-01 | 2010-12-02 | Scopia, LLC | Intermodal trip planner |
| CN103309864A (en) * | 2012-03-07 | 2013-09-18 | 腾讯科技（深圳）有限公司 | Method, device and system for displaying search result |
| US20140108070A1 (en) * | 2012-10-11 | 2014-04-17 | Getgoing, Inc. | Using multi-destination searches to facilitate the purchase of travel itineraries |
| US20140114705A1 (en) * | 2012-10-23 | 2014-04-24 | Olset, Inc. | Methods and systems for making travel arrangements |
| CN104239458A (en) * | 2014-09-02 | 2014-12-24 | 百度在线网络技术（北京）有限公司 | Method and device for representing search results |
-
2016
- 2016-01-05 CA CA2975794A patent/CA2975794A1/en not_active Abandoned
- 2016-01-05 AU AU2016222066A patent/AU2016222066A1/en not_active Abandoned
- 2016-01-05 CN CN201680010605.6A patent/CN107251022A/en active Pending
- 2016-01-05 WO PCT/EP2016/000007 patent/WO2016131519A1/en active Application Filing
Patent Citations (10)
| 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 |
| CN101876979A (en) * | 2009-04-28 | 2010-11-03 | 株式会社理光 | Query expansion method and equipment |
| US20100305984A1 (en) * | 2009-06-01 | 2010-12-02 | Scopia, LLC | Intermodal trip planner |
| CN103309864A (en) * | 2012-03-07 | 2013-09-18 | 腾讯科技（深圳）有限公司 | Method, device and system for displaying search result |
| US20140108070A1 (en) * | 2012-10-11 | 2014-04-17 | Getgoing, Inc. | Using multi-destination searches to facilitate the purchase of travel itineraries |
| US20140114705A1 (en) * | 2012-10-23 | 2014-04-24 | Olset, Inc. | Methods and systems for making travel arrangements |
| CN104239458A (en) * | 2014-09-02 | 2014-12-24 | 百度在线网络技术（北京）有限公司 | Method and device for representing search results |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112925974A (en) * | 2019-12-06 | 2021-06-08 | 艾玛迪斯简易股份公司 | Optimized product determination system |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2016131519A1 (en) | 2016-08-25 |
| CA2975794A1 (en) | 2016-08-25 |
| AU2016222066A1 (en) | 2017-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Antonio et al. | Big data in hotel revenue management: Exploring cancellation drivers to gain insights into booking cancellation behavior | |
| US20210224260A1 (en) | Executing and processing corporate travel search results | |
| AU2019202867B2 (en) | Smart advisory for distributed and composite testing teams based on production data and analytics | |
| US20160239765A1 (en) | Personalized ranking for search results of a travel-related database query | |
| JP6129953B2 (en) | Classification and ranking of travel-related search results | |
| US10007957B2 (en) | Selecting search results for responding to search query | |
| AU2019202454A1 (en) | Smart advisory for distributed and composite testing teams based on production data and analytics | |
| US20150142482A1 (en) | Search engine for identifying business travel proposals | |
| KR20230104082A (en) | Electronic device for recommending travel plan and method for operating the same | |
| JP6328135B2 (en) | Recommended engine for interactive search forms | |
| US20180075494A1 (en) | Sales promotion processing system and sales promotion processing program | |
| CN107251022A (en) | Personalized ranking for the search result of the data base querying for correlation of travelling | |
| CN102855553A (en) | Travel information providing method, system and non-transitory computer-readable storage medium | |
| US20220092483A1 (en) | Customer experience generator with shareable profile and autopay | |
| US20170140458A1 (en) | Method of estimating tenancy duration and mobility in rental properties | |
| JP7324440B2 (en) | Information provision device and information provision method | |
| EP2874109A1 (en) | Search engine for identifying business travel proposals | |
| KR102679809B1 (en) | Methtd and apparatus for recommending items | |
| EP3059707A1 (en) | Personalized ranking for search results of a travel-related database query | |
| EP3082077A1 (en) | Selecting search results for responding to search query | |
| US11681503B2 (en) | Machine learning visual code and action generation | |
| KR20150057983A (en) | Search engine for identifying business travel proposals | |
| KR102653483B1 (en) | Method of predicting price of artwork based on artificial intelligence | |
| AU2016202297B2 (en) | Selecting search results for responding to search query | |
| Jadhav | House Price Prediction And Recommendation System |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20171013 |
|
| WD01 | Invention patent application deemed withdrawn after publication |