CN111461835A - Travel scheme recommendation method and system, intelligent terminal and storage medium - Google Patents

Travel scheme recommendation method and system, intelligent terminal and storage medium Download PDF

Info

Publication number
CN111461835A
CN111461835A CN202010251214.1A CN202010251214A CN111461835A CN 111461835 A CN111461835 A CN 111461835A CN 202010251214 A CN202010251214 A CN 202010251214A CN 111461835 A CN111461835 A CN 111461835A
Authority
CN
China
Prior art keywords
schemes
scheme
station
module
keys
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.)
Granted
Application number
CN202010251214.1A
Other languages
Chinese (zh)
Other versions
CN111461835B (en
Inventor
厉亮
陈雪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou Chuanglv Tianxia Information Technology Co ltd
Original Assignee
Suzhou Chuanglv Tianxia Information Technology Co ltd
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
Application filed by Suzhou Chuanglv Tianxia Information Technology Co ltd filed Critical Suzhou Chuanglv Tianxia Information Technology Co ltd
Priority to CN202010251214.1A priority Critical patent/CN111461835B/en
Publication of CN111461835A publication Critical patent/CN111461835A/en
Application granted granted Critical
Publication of CN111461835B publication Critical patent/CN111461835B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Economics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • General Physics & Mathematics (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to a method and a system for recommending a travel scheme, an intelligent terminal and a storage medium, wherein the method comprises the following steps: s100, constructing schemes, namely constructing all the schemes according to the real-time surplus tickets and the initial station and the arrival station selected by the user; s200, rejecting unreasonable schemes, and rejecting unreasonable schemes in all the schemes constructed in the step S100 according to a set rule; s300, coarsely arranging schemes, and filtering the scheme with high similarity; s400, recalling schemes, namely recalling the schemes meeting set recall rules in the rejected and filtered schemes into a candidate selection set; s500, fine discharging the schemes, and screening a plurality of schemes from the schemes in the step S300 and the schemes in the candidate set according to a set fine discharging rule to serve as final schemes; and S600, exposing the scheme, and presenting the final scheme obtained in the step S500 to a user for selection. The invention can recommend a reasonable, reliable and high-acceptance scheme for the user in a plurality of schemes, and effectively improves the ticket purchasing efficiency.

Description

Travel scheme recommendation method and system, intelligent terminal and storage medium
Technical Field
The invention relates to the technical field of internet, in particular to a method and a system for recommending a travel scheme, an intelligent terminal and a storage medium.
Background
At present, when a user purchases a ticket on a network, a departure station and a destination station can be input on a ticket purchasing page of a network ticket purchasing system, and then ticket information meeting requirements is presented for the user on the basis of the departure station and the destination station. When there is no direct train number or the direct train number is insufficient between the departure station and the destination station, some transfer schemes are provided.
The patent publication with the prior publication number of CN105574597A discloses a method for implementing network ticket booking, which comprises the following steps: obtaining a travel ticket buying request of a user; analyzing the travel ticket buying request to obtain travel information, wherein the travel information comprises a departure station, a destination station and travel time; determining an optional transfer journey from the departure station to the destination station through the transfer station according to the trip journey information; screening out a preferred transit journey with transit journey total journey time length meeting preset conditions and available remaining tickets from the selectable transit journeys, wherein the transit journey total journey time length is the total time required for arriving at the destination station from the departure station through a transit station; booking is made for the preferred transit trip.
However, the above prior art solutions have the following drawbacks: in practical application, more schemes with repeated train numbers, such as multiple transfer schemes with the same train number in the first and second courses but different transfer sites, are screened out only according to the condition, and if the schemes are presented to the user, the number of the schemes is large and complicated, so that the user is greatly hindered from viewing and selecting the schemes, and ticket purchasing efficiency is affected.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a travel scheme recommendation method, a travel scheme recommendation system, an intelligent terminal and a storage medium.
The invention aims at: the method for recommending the itinerary scheme has the advantages that the reasonable, reliable and high-acceptance scheme can be recommended for the user in a plurality of schemes, and ticket purchasing efficiency is effectively improved.
The second purpose of the invention is that: provided is a travel plan recommendation system which has the advantage of being capable of improving ticket purchasing efficiency.
The third purpose of the invention is that: the utility model provides an intelligent terminal, it has can make the user look over reasonable and reliable scheme fast, very big promotion user experience.
The fourth purpose of the invention is: a computer-readable storage medium is provided, which can store a corresponding program and has a feature of facilitating efficient ticket purchasing.
The above object of the present invention is achieved by the following technical solutions:
a travel plan recommendation method includes the following steps,
s100, scheme construction: constructing all schemes according to the real-time surplus tickets and the initial station and the arrival station selected by the user;
s200, removing unreasonable schemes: rejecting unreasonable schemes in all the schemes constructed in the step S100 according to a set rule, wherein the unreasonable schemes at least comprise the scheme that the transit time is out of the range of the set time threshold;
s300, scheme coarse arrangement: a scheme with high similarity of filtering schemes;
s400, scheme recall: recalling schemes meeting set recall rules in the rejected and filtered schemes into a selected candidate set;
s500, fine arrangement of a scheme: screening a plurality of schemes from the schemes after rough filtration in the step S300 and the schemes in the candidate set according to a set fine-sorting rule to be used as final schemes;
s600, scheme exposure: and presenting the final scheme obtained in the step S500 to the user and providing the user for selection.
By adopting the technical scheme, the schemes which are unreasonable and have high train number repeatability can be filtered, a small part of schemes which are possibly accepted by the user are recalled into the candidate set by adopting the recall mechanism, and finally the final scheme is screened out from the schemes which are left after filtering and the schemes in the candidate set and presented to the user, so that the scheme which is reasonable, reliable and high in acceptance is recommended to the user in a plurality of schemes, and the ticket purchasing efficiency is effectively improved.
The present invention in a preferred example may be further configured to: step S100 specifically includes the following substeps:
s101, entering parameter conversion: converting data input by a user or a selected city into a starting station and an arrival station;
s102, selecting a transit city: selecting a proper transfer station according to the starting station and the arrival station, or setting the transfer station to be empty when the transfer station does not exist;
s103, remaining ticket query: inquiring all the remaining tickets according to the starting station, the arrival station and the transfer stations, wherein each transfer station inquires at least 2 times, namely the starting station-the transfer station and the transfer station-the arrival station inquire at least once;
s104, constructing a scheme: constructing all schemes according to the two-course surplus ticket;
s105, scheme scoring: and (5) performing basic scoring on the constructed scheme.
By adopting the technical scheme, the remaining tickets and all possible schemes can be quickly inquired according to the data input by the user, and the screening of the following schemes is paved, so that the ticket purchasing efficiency is better improved.
The present invention in a preferred example may be further configured to: step S200 specifically includes the following substeps:
s201, unreasonable removing of transfer duration: removing the scheme with the transit time length out of the range of the set time threshold;
s202, removing a detour scheme: removing the scheme with the detour index out of a specific threshold range, wherein the detour index = the linear distance between the departure station and the arrival station/(the linear distance between the departure station and the transfer station + the linear distance between the transfer station and the arrival station);
s203, no-ticket elimination: and eliminating the scheme that all the seats in the two courses have no ticket.
By adopting the technical scheme, various unreasonable schemes can be eliminated, and the scheme recommended to the user is reasonable, reliable and high in acceptance.
The present invention in a preferred example may be further configured to: step S300 specifically includes the following substeps:
s301, repeating and removing the overfrequency: eliminating other schemes except the scheme with the shortest total time length from a plurality of schemes with the same travel number;
s302, removing excess computing power: when the expected coarse row number a within the computing power is greater than or equal to the actual solution number B, all solutions remain; when the expected coarse row number A is smaller than the actual scheme number B, A schemes are screened out from the actual schemes.
By adopting the technical scheme, the scheme rejection mechanism is further enhanced, more unreasonable schemes can be rejected, and the reasonability and reliability of the recommended scheme for the user are ensured.
The present invention in a preferred example may be further configured to: the substep S302 specifically comprises the following steps:
s303, judging: when the estimated coarse line number a within the calculation capacity is greater than or equal to the actual scenario number B, all scenarios are retained and proceed to step S400; when the expected coarse row number a is smaller than the actual scenario number B, proceed to step S304;
s304, grouping: splicing all parameter types of the schemes into keys, wherein one key represents one group, and putting all the schemes into different groups according to the difference of the keys;
s305, sampling: if the number K of all keys is larger than a set threshold value F, sampling is carried out according to the following steps:
a1, calculating the score of the scheme in each key according to a set scoring formula, and taking the minimum value to represent the score of the corresponding key;
a2, sampling the importance of the keys according to the scores of the keys, selecting a certain number of keys, and then randomly selecting one scheme from the schemes corresponding to the keys;
a3, repeating the step A2 until F schemes are selected;
if the number K of all keys is less than or equal to a set threshold value F, sampling according to the following steps:
b1, randomly drawing a scheme from each key; then, calculating the score of the scheme in each key according to a set scoring formula, and taking the minimum value to represent the score of the corresponding key;
b2, according to the scores of the keys, sampling the importance of the keys, selecting a certain number of keys, and then randomly selecting one scheme from the schemes corresponding to the keys;
b3, repeating the step B2 until F-K schemes are selected.
By adopting the technical scheme, the number of the schemes can be ensured to be in the effective computing capacity, the situation that the processing is difficult due to too many schemes is avoided, and the scheme processing speed is effectively ensured.
The present invention in a preferred example may be further configured to: step S400 specifically includes the following substeps:
s401, calling a hot scheme: the scheme with the highest monthly volume is taken into a candidate set;
s402, recall of click schemes: incorporating the clicked scheme in 7 days into a candidate set;
s403, self-selection scheme recall: and (4) incorporating the scheme just finished with other users in the same line into the candidate set.
By adopting the technical scheme, the recent popular schemes can be recalled and put into candidate concentration, and the scheme with higher acceptability is presented for the user more favorably.
The present invention in a preferred example may be further configured to: step S500 specifically includes the following substeps:
s501, setting a label: labeling the schemes after rough filtering in the step S300 and the schemes in the candidate set according to the characteristics of the schemes;
s502, score correction: deducting the basic scores corresponding to the schemes according to the situation corresponding to the label of each scheme to obtain final scores or calculating each scheme and the characteristics thereof through a model to obtain final scores;
s503, scheme screening: and screening a plurality of schemes as final schemes according to the sequence of the final scores from small to large.
By adopting the technical scheme, the schemes in the candidate set can be effectively screened, and the reasonability of the scheme finally presented to the user is ensured.
The second aim of the invention is realized by the following technical scheme:
a trip plan recommendation system based on the method of the above technical solution comprises,
the scheme building module is used for building all schemes according to the real-time surplus tickets and the starting station and the arrival station selected by the user;
a scheme rejection module, which comprises,
the first removing submodule is used for removing the scheme of which the transfer time length is out of the range of the set time threshold;
the second elimination submodule is used for eliminating the scheme with the detour index out of a specific threshold range, wherein the detour index = the linear distance between the departure station and the arrival station/(the linear distance between the departure station and the transfer station + the linear distance between the transfer station and the arrival station); and the number of the first and second groups,
the third eliminating submodule is used for eliminating the scheme that all seats in the two courses have no ticket;
the scheme is characterized in that a coarse-row module comprises,
the first coarse ranking sub-module is used for eliminating other schemes except the scheme with the shortest total time length from a plurality of schemes with the same travel number; and the number of the first and second groups,
a second coarse ranking sub-module for reserving all of the solutions when the expected coarse ranking number a within the computing power is greater than or equal to the actual solution number B; the method is also used for screening A schemes from the actual schemes when the expected coarse row number A is smaller than the actual scheme number B;
the scheme recalling module is used for recalling the schemes meeting set recall rules in the schemes rejected by the scheme rejecting module and filtered by the scheme coarse arrangement module into a selected candidate set;
the scheme fine-arranging module is used for screening a plurality of schemes from the schemes filtered by the scheme coarse-arranging module and the schemes in the candidate set according to a set fine-arranging rule to serve as final schemes; and the number of the first and second groups,
and the scheme exposure module is used for presenting the final scheme obtained by the scheme fine ranking module to a user and providing the user with the selection.
By adopting the technical scheme, the reasonable, reliable and high-acceptance scheme can be recommended to the user in a plurality of schemes constructed based on the destination requirements of the user, the checking difficulty and the selecting difficulty of the user are reduced, and the ticket purchasing efficiency can be improved.
The third object of the invention is realized by the following technical scheme:
an intelligent terminal comprising a memory and a processor, the memory having stored thereon a computer program that can be loaded by the processor and carry out any of the methods described above.
Through adopting above-mentioned technical scheme, it has the characteristics that can make the user look over reasonable and reliable scheme fast, very big promotion user experience.
The fourth object of the invention is realized by the following technical scheme:
a computer readable storage medium storing a computer program capable of being loaded by a processor and performing any of the methods described above.
By adopting the technical scheme, the corresponding program can be stored, and the method has the characteristic of facilitating the realization of efficient ticket purchasing.
In summary, the invention includes at least one of the following beneficial technical effects:
1. by setting the steps of scheme construction, unreasonable scheme rejection, scheme rough arrangement, scheme recall, scheme fine arrangement and scheme exposure, a reasonable, reliable and high-acceptance scheme can be recommended to a user in a plurality of schemes, the browsing intensity and the selection difficulty of the user are reduced, and the ticket purchasing efficiency is effectively improved;
2. by setting the method for re-screening the schemes from the filtered remaining schemes and the method for screening the schemes from the schemes of the candidate set, the screened final schemes are more reasonable and reliable.
Drawings
FIG. 1 is a flow chart of a travel itinerary recommendation method shown in an embodiment one;
FIG. 2 is a flowchart illustrating steps S100-S400 according to an embodiment;
FIG. 3 is a flow chart of substeps S401-substep S403 shown in one embodiment;
FIG. 4 is a flow chart illustrating substeps 501-substep S503 according to one embodiment;
fig. 5 is a block diagram showing the configuration of the travel scenario recommendation system according to the second embodiment.
In the figure, 10, a scheme building block; 20. a scheme rejection module; 21. a first culling submodule; 22. a second culling submodule; 23. a third culling submodule; 30. a scheme coarse arrangement module; 31. a first coarse row sub-module; 32. a second coarse row sub-module; 40. a proposal recall module; 50. a scheme fine-arranging module; 60. and a scheme exposure module.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention.
Example one
Referring to fig. 1, a method for recommending a travel itinerary according to the present invention includes the following steps,
s100, scheme construction: constructing all schemes according to the real-time surplus tickets and the initial station and the arrival station selected by the user;
s200, removing unreasonable schemes: rejecting unreasonable schemes in all the schemes constructed in the step S100 according to a set rule;
s300, scheme coarse arrangement: a scheme with high similarity of filtering schemes;
s400, scheme recall: recalling schemes meeting set recall rules in the rejected and filtered schemes into a selected candidate set; the recall rule is one or more of a hot project, a real-time deal, a low-price route, the shortest time and the latest browse;
s500, fine arrangement of a scheme: screening a plurality of schemes from the schemes in the step S300 and the schemes in the candidate set according to a set fine ranking rule to serve as final schemes;
s600, scheme exposure: and presenting the final scheme obtained in the step S500 to the user and providing the user for selection.
Referring to fig. 2, step S100 specifically includes the following sub-steps:
s101, entering parameter conversion: converting data input by a user or a selected city into a starting station and an arrival station; specifically, the method comprises the steps of avoiding the same city, recording error of entering and participating conversion, and converting city-to-city input by a user into station-to-station input;
s102, selecting a transit city: selecting a proper transfer station according to the starting station and the arrival station, or setting the transfer station to be empty when the transfer station does not exist; if the transfer station does not need the user to switch the city, the transfer station is configured as a station with the maximum transfer volume in the travel route; if the user needs to switch the city, matching an optimal transfer station for the user according to the distance of the city and the commuting line condition;
s103, remaining ticket query: inquiring all the remaining tickets according to the starting station, the arrival station and the transfer stations, wherein each transfer station inquires at least 2 times; in this embodiment, the remaining ticket query for the transit city is configured to query for 2 times, that is, one query for departure station-transit station and one query for transit station-arrival station;
s104, constructing a scheme: constructing all schemes which can reach the final arrival station from the initial station according to the two-stroke surplus tickets;
s105, scheme scoring: and (5) performing basic scoring on the constructed scheme.
It should be noted that, in the scheme scoring step S105, taking the two-pass scheme as an example, the preset scheme scoring formula is: y = 0.1905X 1+ 0.0869X 2+ 0.0869X 3+ 0.2455X 4+ 0.1217X 5+ 0.0604X 6+ 0.1704X 7+ 0.0377X 8+ 0.01X 9, wherein X1-X9 are total time length, transit times, price, first-pass departure time, first-pass arrival time, second-pass arrival time, two-pass classification score and advance departure time respectively, and constants in front of X1-X9 are weights of the first-pass departure time and the second-pass arrival time respectively;
(1) the total time length X1 is the second-pass arrival time-first-pass departure time, and is configured to be 1, 6-12 hours for 6 hours, 2, 12-18 hours for 3, 18-24 hours for 4, and so on, and the score is increased by 1 every 6 hours. (2) The transit time length X2 is the second-trip departure time-first-trip arrival time, and is configured to be 1, 1-2 hours for 2, 2-3 hours for 3, 3-4 hours for 4, 4-5 hours for 5, 5-6 hours for 6 in 1 hour, and so on, and the score is increased by 1 every 1 hour. (3) The number of transitions is X3, the same station is configured as 1, and the different station is configured as 2. (4) The price X4 is configured to be 1 within 200, the 400 is configured to be 2, the 600 is configured to be 3, and so on, and the score is added by 1 for every 200. (5) The first departure time X5, points 9-22 are configured as 1, points 23-8 are configured as 2. (5) The first departure time X5, points 9-22 are configured as 1, points 23-8 are configured as 2. (6) The first-pass arrival time X6, points 6-22 are configured as 1, and points 23-5 are configured as 2. (7) The second departure time X7, points 8-22 are configured as 1, and points 23-7 are configured as 2. (8) Two-program class score X8, TF configured 4, TT configured 1, TB configured 4, FT configured 6, FF configured 6, BB configured 6, BT configured 6, AF configured 3, FA configured 3, TC configured 3. (8) The departure time X9 is advanced, namely the small time difference between the order placing and the departure time of the user is configured to be 2 within 2 hours; the time period of more than 2 hours was set to 1.
Referring to fig. 2, step S200 specifically includes the following sub-steps:
s201, unreasonable removing of transfer duration: removing the scheme with the transit time length out of the range of the set time threshold;
s202, removing a detour scheme: eliminating schemes with the detour indexes exceeding a specific threshold range, wherein the detour indexes = the straight-line distance between the departure station and the arrival station/(the straight-line distance between the departure station and the transfer station + the straight-line distance between the transfer station and the arrival station);
s203, no-ticket elimination: eliminating the scheme that the ticket-free condition reaches two passes; specifically, for example, the journey is a journey A → a → B → C, and if the trains A → B and B → C have no ticket, the scheme is rejected.
Referring to fig. 2, step S300 specifically includes the following sub-steps:
s301, repeating and removing the overfrequency: eliminating other schemes except the scheme with the shortest total time length from a plurality of schemes with the same travel number; for example, there are three schemes of G01+ suzhou transit + G02, G01+ Nanjing transit + G02, and G01+ Yangzhou transit + G02, if the total duration of G01+ suzhou transit + G02 is shortest, the scheme of G01+ suzhou transit + G02 is retained, and the other two schemes are eliminated; for another example, the four schemes are respectively G01+ G11, G01+ G12, G01+ G13 and G01+ G14, if the total duration of G01+ G11 is shortest, the scheme of G01+ G11 is reserved, and the other three schemes are removed;
s302, removing excess computing power: when the estimated number of coarse lines a in the calculation capacity is greater than or equal to the actual number of plans B after the step S301, all plans remain; when the expected coarse row number A is smaller than the actual scheme number B, A schemes are screened out from the actual schemes.
Specifically, the substep S302 comprises the following steps:
s303, judging: when the estimated coarse line number a within the calculation capacity is greater than or equal to the actual scenario number B, all scenarios are retained and proceed to step S400; when the expected coarse row number a is smaller than the actual scenario number B, proceed to step S304;
s304, splicing all parameter types of the schemes into keys, wherein one key represents one group, and putting all the schemes into different groups according to the difference of the keys;
for example, the parameter types of the scheme include city, travel tool, price, time, and number of transfers; marking as a _1 for a departure station _ arrival station under the city parameters; under the travel tool parameters, taking two travel car numbers as an example, splicing the travel tool of the first travel and the travel tool of the second travel, for example, GD, G represents a high-speed rail, D represents a motor car, and is marked as b _ 1; in the price parameters, obtaining the price of a transfer scheme, giving the price of a sleeper if the transfer scheme relates to that a train crosses night, giving the price of a hard seat (fast) or a second seat (high-speed rail car) if the train does not cross night, and dividing the price into barrels in every N yuan, wherein the barrels are marked as c _ 1; in the time dimension, dividing a starting time point into barrels every 30 minutes and recording the starting time point as d _1, dividing a transit time point into barrels every 20 minutes and recording the transit time point as d _2, dividing a transit starting time point into barrels every hour and recording the transit starting time point as d _3, and dividing a total time point into barrels every hour and recording the total transit time point as d _ 4; considering whether the same station is transferred or not under the transfer times parameter, and recording as e _ 1;
finally, splicing the a-e into keys, and putting all the schemes into different groups according to the difference of the keys;
s305, sampling: if the number K of all keys is larger than a set threshold value F, sampling is carried out according to the following steps:
a1, calculating the score of the scheme in each key according to a set scoring formula, and taking the minimum value to represent the score of the corresponding key;
a2, according to the scores of the keys, importance sampling is carried out on the keys, a certain number of keys are selected, the lower the score of the key is, the more important the score of the key is, the smaller the total duration of the scheme under the same score is, the more important the scheme is, then a plurality of keys with the front importance degrees are selected, and one scheme is randomly selected from the schemes corresponding to the keys;
a3, repeating the step A2 until F schemes are selected, wherein F is less than or equal to A;
if the number K of all keys is less than or equal to a set threshold value F, sampling according to the following steps:
b1, randomly drawing a scheme from each key; then, calculating the score of the scheme in each key according to a set scoring formula, and taking the minimum value to represent the score of the corresponding key;
b2, according to the scores of the keys, performing importance sampling on the keys, selecting a certain number of keys, then selecting a plurality of keys with the front importance and randomly selecting one scheme from the schemes corresponding to the keys;
b3, repeating the step B2 until F-K schemes are selected, wherein F-K is less than or equal to A.
Referring to fig. 3, step S400 specifically includes the following sub-steps:
s401, calling a hot scheme: the scheme with the highest monthly volume is taken into a candidate set;
s402, recall of click schemes: incorporating the clicked scheme in 7 days into a candidate set;
s403, self-selection scheme recall: and (4) incorporating the scheme just finished with other users in the same line into the candidate set.
Referring to fig. 4, step S500 is a process including the following sub-steps:
s501, setting a label: labeling the schemes in the step S300 and the schemes in the candidate set according to the characteristics of the schemes; in the embodiment, the labels comprise the same-vehicle transfer, the shortest time, the low-price line, the earliest arrival, the 100% ticketing, the hot scheme, the latest browsing, the sleeping overnight, the faster than direct, the self-organizing scheme and the historical purchase;
s502, score correction: deducting the basic scores corresponding to the schemes according to the conditions corresponding to the labels of each scheme to obtain final scores or calculating each scheme and the characteristics of the schemes through a model to obtain the final scores; specifically, when the basic score corresponding to the scheme is deducted according to the condition corresponding to the label of each scheme and the final score of each scheme is obtained, the same-vehicle transfer is deducted for 0 minute, the shortest time duration is deducted for 2 minutes, the low-price line is deducted for 1 minute, the earliest arrival is deducted for 2 minutes, 100% ticket deduction is deducted for 0.9 minute, the hot scheme is deducted for 0.9 minute, the latest browsing deduction is 0.45 minute, the sleeping overnight deduction is 0.3 minute, the direct quick deduction is 0.45 minute, the self-set scheme is deducted for 0.15 minute, and the historical purchase deduction is 0.15 minute; when each scheme and the characteristics thereof are calculated through a model to obtain a final score, each scheme and the characteristics thereof are calculated and scored through XGB 4J to obtain the final score of each scheme;
s503, scheme screening: and screening a plurality of previous schemes according to the sequence of the final scores from small to large, wherein the schemes screened in the step are used as final schemes exposed to the user.
Example two
Referring to fig. 5, the system for recommending a itinerary scheme based on the method for recommending an itinerary scheme disclosed in the first embodiment of the present invention includes a scheme constructing module 10, a scheme rejecting module 20, a scheme rough ranking module 30, a scheme recalling module 40, a scheme fine ranking module 50, and a scheme exposing module 60.
The project building module 10 is used to build all projects based on the real-time remainder tickets and the user selected starting and arrival stations. Specifically, the scheme building module 10 converts data input by a user or a selected city into an origin station and an arrival station; then selecting a proper transfer station according to the starting station and the arrival station, or setting the transfer station to be empty when the transfer station does not exist; then inquiring all the remaining tickets according to the starting station, the arrival station and the transfer stations, wherein each transfer station inquires at least 2 times; finally, all schemes which can reach the final arrival station from the initial station are constructed according to the remaining tickets.
The solution culling module 20 includes a first culling sub-module 21, a second culling sub-module 22, and a third culling sub-module 23. The first removing submodule 21 is configured to remove a scheme in which the transfer duration is outside the set time threshold range; the second eliminating submodule 22 is configured to eliminate a scenario in which a detour index is outside a specific threshold range, where the detour index = a straight distance between the departure station and the arrival station/(a straight distance between the departure station and the transfer station + a straight distance between the transfer station and the arrival station); the third culling sub-module 23 is used for culling the scheme which reaches two passes in the no-ticket condition.
The solution coarse-row module 30 comprises a first coarse-row sub-module 31 and a second coarse-row sub-module 32. The first coarse ranking sub-module 31 is used for eliminating other schemes with the shortest total time length from the schemes with the same trip number; a second coarse ranking sub-module 32 for reserving all of the solutions when the expected coarse ranking number a within the computing power is greater than or equal to the actual solution number B; it is also used to screen a solutions from the actual solutions when the expected coarse row number a is less than the actual solution number B.
The proposal recalling module 40 is used for recalling the proposals meeting the set recall rule in the proposals rejected by the proposal rejecting module 20 and filtered by the proposal coarse-ranking module 30 into the selected candidate set; the recall rule is one or more of a hit plan, a real-time deal, a low-price route, a shortest duration, and a latest browse. Specifically, in this embodiment, the plan recalling module 40 is configured to recall and incorporate the plan with the highest volume in the last month, the plan clicked in the last 7 days, and the plan just completed by other users in the same route into the candidate set.
The scheme fine-ranking module 50 is configured to screen a plurality of schemes from the schemes filtered by the scheme coarse-ranking module 30 and the schemes in the candidate set according to a set fine-ranking rule, which is the same as the scheme fine-ranking rule of steps S501 to S503 in the first embodiment.
The scenario exposure module 60 is configured to present the final scenario obtained by the scenario refinement module 50 to the user and provide the user with a selection, that is, present to the user both the scenario screened from the scenarios filtered by the scenario refinement module 30 and the scenario screened from the scenarios in the candidate set.
EXAMPLE III
An intelligent terminal comprises a memory and a processor, wherein the memory stores a computer program which can be loaded by the processor and executes a travel scheme recommendation method in the first embodiment.
Example four
A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing a trip plan recommendation method according to a first embodiment, the computer-readable storage medium includes, for example: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
It is to be understood that the above-described embodiments are only some of the embodiments of the present invention, and not all of them. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In addition, the term "and/or" herein is only one kind of association relationship describing an associated object, and means that there may be three kinds of relationships, for example, a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship, unless otherwise specified.

Claims (10)

1. A travel plan recommendation method is characterized by comprising the following steps,
s100, scheme construction: constructing all schemes according to the real-time surplus tickets and the initial station and the arrival station selected by the user;
s200, removing unreasonable schemes: rejecting unreasonable schemes in all the schemes constructed in the step S100 according to a set rule, wherein the unreasonable schemes at least comprise the scheme that the transit time is out of the range of the set time threshold;
s300, scheme coarse arrangement: a scheme with high similarity of filtering schemes;
s400, scheme recall: recalling schemes meeting set recall rules in the rejected and filtered schemes into a selected candidate set;
s500, fine arrangement of a scheme: screening a plurality of schemes from the schemes after rough filtration in the step S300 and the schemes in the candidate set according to a set fine-sorting rule to be used as final schemes;
s600, scheme exposure: and presenting the final scheme obtained in the step S500 to the user and providing the user for selection.
2. The method according to claim 1, characterized in that step S100 comprises in particular the following sub-steps:
s101, entering parameter conversion: converting data input by a user or a selected city into a starting station and an arrival station;
s102, selecting a transit city: selecting a proper transfer station according to the starting station and the arrival station, or setting the transfer station to be empty when the transfer station does not exist;
s103, remaining ticket query: inquiring all the remaining tickets according to the starting station, the arrival station and the transfer stations, wherein each transfer station inquires at least 2 times, namely the starting station-the transfer station and the transfer station-the arrival station inquire at least once;
s104, constructing a scheme: constructing all schemes according to the two-course surplus ticket;
s105, scheme scoring: and (5) performing basic scoring on the constructed scheme.
3. The method according to claim 1, wherein step S200 comprises the following sub-steps:
s201, unreasonable removing of transfer duration: removing the scheme with the transit time length out of the range of the set time threshold;
s202, removing a detour scheme: removing the scheme with the detour index out of a specific threshold range, wherein the detour index = the linear distance between the departure station and the arrival station/(the linear distance between the departure station and the transfer station + the linear distance between the transfer station and the arrival station);
s203, no-ticket elimination: and eliminating the scheme that all the seats in the two courses have no ticket.
4. The method according to claim 1, wherein step S300 comprises the following sub-steps:
s301, repeating and removing the overfrequency: eliminating other schemes except the scheme with the shortest total time length from a plurality of schemes with the same travel number;
s302, removing excess computing power: when the expected coarse row number a within the computing power is greater than or equal to the actual solution number B, all solutions remain; when the expected coarse row number A is smaller than the actual scheme number B, A schemes are screened out from the actual schemes.
5. The method according to claim 4, characterized in that sub-step S302 comprises in particular the following sub-steps:
s303, judging: when the estimated coarse line number a within the calculation capacity is greater than or equal to the actual scenario number B, all scenarios are retained and proceed to step S400; when the expected coarse row number a is smaller than the actual scenario number B, proceed to step S304;
s304, grouping: splicing all parameter types of the schemes into keys, wherein one key represents one group, and putting all the schemes into different groups according to the difference of the keys;
s305, sampling: if the number K of all keys is larger than a set threshold value F, sampling is carried out according to the following steps:
a1, calculating the score of the scheme in each key according to a set scoring formula, and taking the minimum value to represent the score of the corresponding key;
a2, sampling the importance of the keys according to the scores of the keys, selecting a certain number of keys, and then randomly selecting one scheme from the schemes corresponding to the keys;
a3, repeating the step A2 until F schemes are selected;
if the number K of all keys is less than or equal to a set threshold value F, sampling according to the following steps:
b1, randomly drawing a scheme from each key; then, calculating the score of the scheme in each key according to a set scoring formula, and taking the minimum value to represent the score of the corresponding key;
b2, according to the scores of the keys, sampling the importance of the keys, selecting a certain number of keys, and then randomly selecting one scheme from the schemes corresponding to the keys;
b3, repeating the step B2 until F-K schemes are selected.
6. The method according to claim 1, wherein step S400 comprises the following sub-steps:
s401, calling a hot scheme: the scheme with the highest monthly volume is taken into a candidate set;
s402, recall of click schemes: incorporating the clicked scheme in 7 days into a candidate set;
s403, self-selection scheme recall: and (4) incorporating the scheme just finished with other users in the same line into the candidate set.
7. The method according to claim 2, wherein step S500 comprises the following sub-steps:
s501, setting a label: labeling the schemes after rough filtering in the step S300 and the schemes in the candidate set according to the characteristics of the schemes;
s502, score correction: deducting the basic scores corresponding to the schemes according to the situation corresponding to the label of each scheme to obtain final scores or calculating each scheme and the characteristics thereof through a model to obtain final scores;
s503, scheme screening: and screening a plurality of schemes as final schemes according to the sequence of the final scores from small to large.
8. A trip scenario recommendation system based on the method of claim 1, comprising,
the scheme building module (10) is used for building all schemes according to the real-time surplus tickets and the starting station and the arrival station selected by the user;
a protocol culling module (20) comprising,
the first removing submodule (21) is used for removing the scheme of which the transfer time length is out of the range of the set time threshold;
a second eliminating submodule (22) for eliminating the scheme with the detour index out of the specific threshold range, wherein the detour index = the straight distance between the departure station and the arrival station/(the straight distance between the departure station and the transfer station + the straight distance between the transfer station and the arrival station); and the number of the first and second groups,
a third eliminating submodule (23) for eliminating the scheme that all the seats in the two courses have no ticket;
a scheme rough-arranging module (30) comprising,
a first coarse ranking sub-module (31) for eliminating the other schemes except the scheme with the shortest total time length from the plurality of schemes with the same travel number; and the number of the first and second groups,
a second coarse ranking sub-module (32) for reserving all solutions when the expected coarse ranking number a within the computing power is greater than or equal to the actual solution number B; the method is also used for screening A schemes from the actual schemes when the expected coarse row number A is smaller than the actual scheme number B;
a proposal recalling module (40) for recalling the proposal which meets the set recall rule and is rejected by the proposal rejecting module (20) and filtered by the proposal coarse ranking module (30) into a selection candidate set;
the scheme fine-arranging module (50) is used for screening a plurality of schemes from the schemes filtered by the scheme coarse-arranging module (30) and the schemes in the candidate set according to a set fine-arranging rule to serve as a final scheme; and the number of the first and second groups,
and the scheme exposure module (60) is used for presenting the final scheme obtained by the scheme fine ranking module (50) to a user and providing the user with selection.
9. An intelligent terminal, comprising a memory and a processor, the memory having stored thereon a computer program that can be loaded by the processor and that executes the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, in which a computer program is stored which can be loaded by a processor and which executes the method of any one of claims 1 to 7.
CN202010251214.1A 2020-04-01 2020-04-01 Stroke scheme recommendation method, system, intelligent terminal and storage medium Active CN111461835B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010251214.1A CN111461835B (en) 2020-04-01 2020-04-01 Stroke scheme recommendation method, system, intelligent terminal and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010251214.1A CN111461835B (en) 2020-04-01 2020-04-01 Stroke scheme recommendation method, system, intelligent terminal and storage medium

Publications (2)

Publication Number Publication Date
CN111461835A true CN111461835A (en) 2020-07-28
CN111461835B CN111461835B (en) 2023-05-23

Family

ID=71685770

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010251214.1A Active CN111461835B (en) 2020-04-01 2020-04-01 Stroke scheme recommendation method, system, intelligent terminal and storage medium

Country Status (1)

Country Link
CN (1) CN111461835B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112232596A (en) * 2020-11-07 2021-01-15 苏州创旅天下信息技术有限公司 Multi-mode transport transit city optimization method, system, terminal and storage medium
CN113610533A (en) * 2021-07-27 2021-11-05 北京淇瑀信息科技有限公司 Resource allocation method and device for transaction refusing user and electronic equipment

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104063831A (en) * 2014-04-18 2014-09-24 北京交大思源科技有限公司 Method and device for providing railway route schemes
CN105574597A (en) * 2015-12-25 2016-05-11 北京奇虎科技有限公司 Realization method and system of online ticket booking

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104063831A (en) * 2014-04-18 2014-09-24 北京交大思源科技有限公司 Method and device for providing railway route schemes
CN105574597A (en) * 2015-12-25 2016-05-11 北京奇虎科技有限公司 Realization method and system of online ticket booking

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112232596A (en) * 2020-11-07 2021-01-15 苏州创旅天下信息技术有限公司 Multi-mode transport transit city optimization method, system, terminal and storage medium
CN112232596B (en) * 2020-11-07 2023-11-24 苏州创旅天下信息技术有限公司 Multi-mode intermodal transit city optimization method, system, terminal and storage medium
CN113610533A (en) * 2021-07-27 2021-11-05 北京淇瑀信息科技有限公司 Resource allocation method and device for transaction refusing user and electronic equipment
CN113610533B (en) * 2021-07-27 2024-08-16 北京淇瑀信息科技有限公司 Resource allocation method and device for transaction refusing user and electronic equipment

Also Published As

Publication number Publication date
CN111461835B (en) 2023-05-23

Similar Documents

Publication Publication Date Title
CN102099803A (en) Method and computer system for automatically answering natural language questions
EP1477921A1 (en) Airline revenue planning and forecasting system and method
Li et al. Optimizing flexible one-to-two matching in ride-hailing systems with boundedly rational users
CN111461835A (en) Travel scheme recommendation method and system, intelligent terminal and storage medium
CN112232596B (en) Multi-mode intermodal transit city optimization method, system, terminal and storage medium
WO2018180302A1 (en) Shared-use fee calculation system
Shafahi et al. SpeedRoute: Fast, efficient solutions for school bus routing problems
CN113919772B (en) Time-varying vehicle path planning method and system with time window
Ata et al. Spatial pricing: An empirical analysis of taxi rides in New York City
CN106997396A (en) Supply and demand information interconnected method and system
Lee et al. Path-oriented synchronized transit scheduling using time-dependent data
CN112529660A (en) Block chain-based electronic ticketing method and system, electronic equipment and storage medium
CN112711711A (en) Knowledge base-based client marketing cue recommendation method and device
CN116579587A (en) Target object distribution method and device
CN110751519B (en) Automatic change method and device for team passenger tickets
CN111612493B (en) User grouping method and device
Gao et al. Optimal parameters of service in a public transportation market with pricing
JP5823332B2 (en) Personal behavior preference analysis system and personal behavior preference analysis method
CN111753386B (en) Data processing method and device
CN115222297B (en) Bus route optimization adjustment scheme evaluation method, electronic device and storage medium
CN112070275A (en) Method, device and equipment for cooperatively limiting current of rail transit road network
CN113053156B (en) Intelligent bus station addressing method based on bus radius method
JP4364801B2 (en) Area division system
Triki et al. A combinatorial auction‐based approach for ridesharing in a student transportation system
Jin et al. Negotiating concession period for highway public-private partnerships with imperfect information

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
GR01 Patent grant
GR01 Patent grant