CN107679674A - The Forecasting Methodology and system of the overseas hotel's house type service deficiency of OTA platforms - Google Patents
The Forecasting Methodology and system of the overseas hotel's house type service deficiency of OTA platforms Download PDFInfo
- Publication number
- CN107679674A CN107679674A CN201710994934.5A CN201710994934A CN107679674A CN 107679674 A CN107679674 A CN 107679674A CN 201710994934 A CN201710994934 A CN 201710994934A CN 107679674 A CN107679674 A CN 107679674A
- Authority
- CN
- China
- Prior art keywords
- house type
- hotel
- information
- order
- overseas
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/12—Hotels or restaurants
Abstract
The invention discloses the Forecasting Methodology and system of the overseas hotel's house type service deficiency of OTA platforms, the Forecasting Methodology includes:Before user places an order, user profile is predicted;Order history information corresponding with user profile is obtained, with the base attribute information for moving in a day historical information, supplier's channel historical information and house type according to user profile;The information of acquisition is respectively processed by XGBoost algorithms, the service deficiency Probabilistic Prediction Model of overseas hotel's house type is established to predict the service deficiency probability of overseas hotel's house type of user's selection, and then corresponding Intervention Strategy is taken to overseas hotel's house type.Present invention aims at provide a kind of service deficiency probability by establishing overseas hotel's house type of overseas hotel's house type service deficiency Probabilistic Prediction Model prediction user's selection, so as to take corresponding Intervention Strategy, realize the intervention to overseas hotel's house type, reduce service deficiency, Consumer's Experience is improved, lifts brand image.
Description
Technical field
The present invention relates to grass roots marketing techniques field, the prediction of more particularly to a kind of overseas hotel's house type service deficiency of OTA platforms
Method and system.
Background technology
At present, it is less that the interference method of service deficiency overseas hotel's house type occurs, mainly with manually for sending out in history
The high hotel of raw house type service deficiency frequency carries out artificial outgoing call and patrolled based on room and artificial investigation method.But due to OTA
(Online Travel Agent, online travel agency) order volume is larger, and manually investigate it is regular relatively simple, and
The house type of be likely to occur service deficiency can not be covered well.Further, since the number of days one that overseas order is subscribed in advance
As it is longer, day is shorter for moving in of being covered by way of room is patrolled in artificial outgoing call, therefore to the house type of service deficiency may occur
Intervene limited.
The content of the invention
The technical problem to be solved in the present invention is to patrol room or people by artificial outgoing call present in prior art to overcome
The defects of work investigation method can not cover overseas hotel's house type of be likely to occur service deficiency well, it is therefore intended that carry
For the Forecasting Methodology and system of a kind of overseas hotel's house type service deficiency of OTA platforms.
The present invention is that solve above-mentioned technical problem by following technical proposals:
The present invention provides a kind of Forecasting Methodology of the overseas hotel's house type service deficiency of OTA platforms, and the Forecasting Methodology includes:
S1, before user places an order, predict user profile;The user profile includes overseas hotel's house type letter of user's selection
Cease and move in day information;
S2, corresponding with user profile order history information obtained according to the user profile, moves in day history together
The base attribute information of information, supplier's channel historical information and house type;
S3, by XGBoost algorithms to the order history information, described with moving in a day historical information, the supplier
The base attribute information of channel historical information and the house type is respectively processed, and the service deficiency for establishing overseas hotel's house type is general
Rate forecast model;
S4, the overseas hotel's house type selected according to the service deficiency Probabilistic Prediction Model prediction user service lack
Fall into probability;
S5, judge whether the service deficiency probability is more than given threshold, when being judged as YES, to the overseas hotel room
Type takes corresponding Intervention Strategy.
It is preferred that the expression formula of the service deficiency Probabilistic Prediction Model is:
Wherein,
Score represents the service deficiency probability, and T represents a number for decision tree, and t represents the t decision tree, ft(xi) table
Show the function of decision tree, w represents leaf weight, yiModel actual value is represented,The model predication value of t-1 tree, θ before expression
Partial derivative is represented, l represents residual error function, and γ and β are adjustable parameter,For by the model actual value and institute
State the residual error function that model predication value defines, giRepresent residual error function pairFirst derivative, hiRepresent residual error function pairSecond dervative, n represent training sample quantity, i represent i-th of training sample, i span is between 1~n
Integer.
It is preferred that the Intervention Strategy includes directly closing house type, house type is postponed in the sequence of OTA platforms, sends postal
Part carries out at least one of secondary-confirmation house type by outgoing call intervention notifications user with overseas hotel is automatically reminded to.
It is preferred that the order history information includes setting day of the overseas hotel's house type of user's selection before the lower odd-numbered day
The order numbers information of service deficiency occurs in number;
The same day history packet of moving in includes the overseas hotel belonging to hotel's house type abroad before the lower odd-numbered day
Same move in setting number of days occurs the order numbers information of service deficiency day and/or moves in whether day is weekend information;
The same order numbers information for moving in day generation service deficiency is used to be characterized in the setting time before the lower odd-numbered day
In the order inside to place an order, the identical quantity on order for moving in day generation service deficiency after the lower odd-numbered day;
Supplier's channel historical information corresponds to the attribute information of supplier including house type and/or the history of supplier is ordered
The situation of service deficiency occurs in list;Wherein, the attribute information of supplier includes whether supplier belongs to what is directly signed with OTA platforms
The class information that information and/or OTA platforms are set to supplier;
The base attribute information of the house type includes whether house type is the information for retaining room;Wherein, the reservation room represents
Hotel is supplied to the room that OTA platforms are freely sold.
It is preferred that the service deficiency full room before including validating that, confirm after full room, to shop without room, to shop without it is predetermined, confirm
At least one of appreciated after preceding rise in price and confirmation;
Full room is used to characterizing after user places an order before the confirmation, and OTA platforms are apprised of when checking house type information with hotel
The situation in the full room of the house type;
Full room is used to characterizing after user places an order after the confirmation, and when checking house type information with hotel, hotel accuses OTA platforms
Know that house type described in OTA platforms has room, after OTA platforms are completed to acknowledgement of orders, the full room of house type described in OTA platforms is informed in hotel
Situation.
Described to be used to characterizing after user places an order without room to shop, preparation does not have the overseas hotel when moving in the overseas hotel
The situation of house type;
It is described to shop without after being intended for characterizing and being used for predetermined order, preparation does not have user pre- when moving in the overseas hotel
Determine the situation of sequence information;
Appreciated before the confirmation for characterizing after user places an order, OTA platforms are apprised of when checking house type information with hotel
The price of the house type needs to go up;
Appreciated after the confirmation for characterizing after user places an order, when checking house type information with hotel, hotel does not have OTA platforms
Inform that house type described in OTA platforms needs to appreciate, after OTA platforms are completed to acknowledgement of orders, hotel is informed described in OTA platforms
House type needs situation about appreciating.
The present invention also provides a kind of forecasting system of the overseas hotel's house type service deficiency of OTA platforms, the forecasting system bag
Include first acquisition unit, second acquisition unit, model and establish unit, predicting unit and processing unit;
Before the first acquisition unit places an order for user, user profile is predicted;The user profile selects including user
Overseas hotel's house type information and move in day information;
The second acquisition unit is used to obtain order history corresponding with the user profile according to the user profile
Information, the same base attribute information for moving in a day historical information, supplier's channel historical information and house type;
The model establish unit be used for by XGBoost algorithms to the order history information, it is described together move in calendar
The base attribute information of history information, supplier's channel historical information and the house type is respectively processed, and establishes overseas wine
The service deficiency Probabilistic Prediction Model of shop house type;
The predicting unit is used for the overseas wine that user's selection is predicted according to the service deficiency Probabilistic Prediction Model
The service deficiency probability of shop house type;
The processing unit is used to judge whether the overseas hotel house type service deficiency probability is more than given threshold, is sentencing
When breaking to be, corresponding Intervention Strategy is taken to the overseas hotel house type.
It is preferred that the expression formula of the service deficiency Probabilistic Prediction Model is:
Wherein,
Score represents the service deficiency probability, and T represents a number for decision tree, and t represents the t decision tree, ft(xi) table
Show the function of decision tree, w represents leaf weight, yiModel actual value is represented,The model predication value of t-1 tree, θ before expression
Partial derivative is represented, l represents residual error function, and γ and β are adjustable parameter, and i and n are natural number.
It is preferred that the Intervention Strategy includes directly closing house type, house type is postponed in the sequence of OTA platforms, sends postal
Part carries out at least one of secondary-confirmation house type by outgoing call intervention notifications user with overseas hotel is automatically reminded to.
It is preferred that the order history information includes setting day of the overseas hotel's house type of user's selection before the lower odd-numbered day
The order numbers information of service deficiency occurs in number;
It is described to take day with moving in overseas wine same move in that day history packet included belonging to overseas hotel's house type
Be engaged in defect order numbers information and/or move in whether day is weekend information;
The same order numbers information for moving in day generation service deficiency is used to be characterized in the setting time before the lower odd-numbered day
In the order inside to place an order, the identical quantity on order for moving in day generation service deficiency after the lower odd-numbered day;
Supplier's channel historical information corresponds to the attribute information of supplier including house type and/or the history of supplier is ordered
The situation of service deficiency occurs in list;Wherein, the attribute information of supplier includes whether supplier belongs to what is directly signed with OTA platforms
The class information that information and/or OTA platforms are set to supplier;
The base attribute information of the house type includes whether house type is the information for retaining room;Wherein, the reservation room represents
Hotel is supplied to the room that OTA platforms are freely sold.
It is preferred that the service deficiency full room before including validating that, confirm after full room, to shop without room, to shop without it is predetermined, confirm
At least one of appreciated after preceding rise in price and confirmation;
Full room is used to characterizing after user places an order before the confirmation, and OTA platforms are apprised of when checking house type information with hotel
The situation in the full room of the house type;
Full room is used to characterizing after user places an order after the confirmation, and when checking house type information with hotel, hotel accuses OTA platforms
Know that house type described in OTA platforms has room, after OTA platforms are completed to acknowledgement of orders, the full room of house type described in OTA platforms is informed in hotel
Situation.
Described to be used to characterizing after user places an order without room to shop, preparation does not have the overseas hotel when moving in the overseas hotel
The situation of house type;
It is described to shop without after being intended for characterizing and being used for predetermined order, preparation does not have user pre- when moving in the overseas hotel
Determine the situation of sequence information;
Appreciated before the confirmation for characterizing after user places an order, OTA platforms are apprised of when checking house type information with hotel
The price of the house type needs to go up;
Appreciated after the confirmation for characterizing after user places an order, when checking house type information with hotel, hotel does not have OTA platforms
Inform that house type described in OTA platforms needs to appreciate, after OTA platforms are completed to acknowledgement of orders, hotel is informed described in OTA platforms
House type needs situation about appreciating.
The positive effect of the present invention is:
The present invention moves in a day historical information, supplier by the corresponding order history information of user's acquisition of information, together
The base attribute information of channel historical information and house type, and the data of acquisition are respectively processed by XGBoost algorithms, build
Overseas hotel's house type service deficiency Probabilistic Prediction Model is found, is predicted and used according to overseas hotel's house type service deficiency Probabilistic Prediction Model
The service deficiency probability of overseas hotel's house type of family selection, so as to take corresponding Intervention Strategy, realize efficiently to overseas wine
Shop house type is intervened, and reduces service deficiency, improves Consumer's Experience, lifts the brand image of overseas hotel and OTA platforms.
Brief description of the drawings
Fig. 1 is the flow chart of the Forecasting Methodology of the overseas hotel's house type service deficiency of OTA platforms of embodiments of the invention 1;
Fig. 2 is the module signal of the forecasting system of the overseas hotel's house type service deficiency of OTA platforms of embodiments of the invention 2
Figure.
Embodiment
The present invention is further illustrated below by the mode of embodiment, but does not therefore limit the present invention to described reality
Apply among a scope.The experimental method of unreceipted actual conditions in the following example, conventionally and condition, or according to business
Product specification selects.
Embodiment 1
As shown in figure 1, the flow chart of the Forecasting Methodology for the overseas hotel's house type service deficiency of OTA platforms of the present embodiment.
The Forecasting Methodology of the overseas hotel's house type service deficiency of OTA platforms of the present embodiment includes:
S101, before user places an order, predict user profile;
Wherein, the user profile includes overseas hotel's house type information of user's selection and moves in day information;
S102, corresponding with user profile order history information obtained according to the user profile, moves in calendar together
The base attribute information of history information, supplier's channel historical information and house type;
Wherein, the order history information includes setting number of days of the overseas hotel's house type of user's selection before the lower odd-numbered day
The interior order numbers information that service deficiency occurs;
It is described to take day with moving in overseas wine same move in that day history packet included belonging to overseas hotel's house type
Be engaged in defect order numbers information and/or move in whether day is weekend information;
The same order numbers information for moving in day generation service deficiency is used to be characterized in the setting time before the lower odd-numbered day
In the order inside to place an order, the identical quantity on order for moving in day generation service deficiency after the lower odd-numbered day;
For example, nearly 30 days of lower odd-numbered day same moved in day that (with female hotel, female hotel represents the actual entity wine moved in of user
Shop) occur service deficiency order numbers refer to:Such as current lower odd-numbered day is on October 1st, 2017, and order moves in the time as 2017
On October 5, in.Now, be 1 day-September of September 30 days by single time under statistics, and with the order (October 1 was placed an order)
Corresponding house type (house type corresponds to same female hotel), and move in all orders that day is October 5 and take
The order numbers for defect of being engaged in.
Supplier's channel historical information corresponds to the attribute information of supplier including house type and/or the history of supplier is ordered
The situation of service deficiency occurs in list;Wherein, the attribute information of supplier includes whether supplier belongs to what is directly signed with OTA platforms
The class information that information and/or OTA platforms are set to supplier;
Wherein, a source of houses part for OTA platforms derives from and the direct cooperation in hotel, another part derive from acquisition hotel
House type sells the third party of power, i.e. supplier.The attribute information of supplier includes the different channels of supplier;Different channels are main
Including straight label, direct-connected supplier of group, cooperation supplier and other OTA suppliers etc..Carried for the supplier of different channels
Type of buying a house in installments is different with the mode of management room availability.
The base attribute information of the house type includes whether house type is the information for retaining room;
Wherein, the reservation room represents that hotel is supplied to the room that OTA platforms are freely sold, and OTA platforms will can be reserved
Room sold with different house type, and need not be confirmed with hotel.
S103, by XGBoost algorithms to the order history information, described with moving in a day historical information, the supply
The base attribute information of business's channel historical information and the house type is respectively processed, and establishes the service deficiency of overseas hotel's house type
Probabilistic Prediction Model;
It is to be not involved in the service deficiency probabilistic forecasting mould of the data to foundation of model training by using test data
Type carries out model checking, and the prediction result of the service deficiency Probabilistic Prediction Model is commented by recall rate and accuracy rate
Valency.The interpretational criteria of model is in the case where recall rate is certain, and accuracy rate is higher, the service deficiency Probabilistic Prediction Model
Prediction result is more accurate.
Wherein, the service deficiency full room before including validating that, confirm after full room, to shop without room, to shop without it is predetermined, confirm before
At least one of rise in price after appreciating and confirming;
For OTA platforms, do not know that hotel is actual usually and whether also have room, therefore after user places an order, OTA platforms
Need to be checked with hotel and (retain some special circumstances that need not be checked with hotel such as room except room belongs to), then
User's order is confirmed, confirmation i.e. herein represent the confirmations that are carried out to user's order of OTA.
Full room is used to characterizing after user places an order before the confirmation, and OTA platforms are apprised of when checking house type information with hotel
The situation in the full room of the house type;
Wherein, the result in the full room of the house type is informed user by OTA platforms after the full room of the house type is apprised of, convenient
User carries out cancelling confirmation or carries out other operations to order.
Full room is used to characterizing after user places an order after the confirmation, and when checking house type information with hotel, hotel accuses OTA platforms
Know that house type described in OTA platforms has room, after OTA platforms are completed to acknowledgement of orders, the full room of house type described in OTA platforms is informed in hotel
Situation.
Described to be used to characterizing after user places an order without room to shop, preparation does not have the overseas hotel when moving in the overseas hotel
The situation of house type;
It is described to shop without after being intended for characterizing and being used for predetermined order, preparation does not have user pre- when moving in the overseas hotel
Determine the situation of sequence information;
Appreciated before the confirmation for characterizing after user places an order, OTA platforms are apprised of when checking house type information with hotel
The price of the house type needs to go up;
Wherein, the price of the house type is needed to go up by OTA platforms after the price for being apprised of the house type needs to go up
Result feed back to user, facilitate user to carry out cancelling confirmation to order or carry out other operations.
Appreciated after the confirmation for characterizing after user places an order, when checking house type information with hotel, hotel does not have OTA platforms
Inform that house type described in OTA platforms needs to appreciate, after OTA platforms are completed to acknowledgement of orders, hotel is informed described in OTA platforms
House type needs situation about appreciating.
S104, the overseas hotel's house type selected according to the service deficiency Probabilistic Prediction Model prediction user service
Shortage probability;
The expression formula of the service deficiency Probabilistic Prediction Model is:
Wherein,
Score represents the service deficiency probability, and T represents a number for decision tree, and t represents the t decision tree, ft(xi) table
Show the function of decision tree, w represents leaf weight, yiModel actual value is represented,The model predication value of t-1 tree, θ before expression
Partial derivative is represented, l represents residual error function, and γ and β are adjustable parameter,For by the model actual value and institute
The residual error function that model predication value defines is stated, for calculating the difference band between the model actual value and the model predication value
The loss come;giRepresent residual error function pairFirst derivative, hiRepresent residual error function pairSecond dervative, n represent
The quantity of training sample, i represent i-th of training sample, the integer of i span between 1~n.
The overseas hotel occupancy to be monitored can be covered to try to achieve to maximize by the service deficiency Probabilistic Prediction Model
The house type state of day.For example, the present invention the service deficiency probabilistic forecasting mould for overseas hotel past 30 days or 180 days
The type of booking rooms for inside having the order that service deficiency occurs is predicted, it is possible to achieve 80% order in covering day order, from
80% order is 60 days in advance predetermined knowable to the distribution situation of predetermined number of days in advance of overseas hotel house type, therefore can be used
The service deficiency probabilistic forecasting mould prediction of the present invention had the overseas hotel that service deficiency occurs within past 30 days or 180 days
Move in room availability corresponding to day within the 60th day after house type and lower odd-numbered day.
S105, judge whether the service deficiency probability is more than given threshold, when being judged as YES, to the overseas hotel
House type takes corresponding Intervention Strategy.
Wherein, the Intervention Strategy includes directly closing house type, house type is postponed in the sequence of OTA platforms, sends mail
Or by outgoing call intervention notifications user and it is automatically reminded at least one of overseas hotel's progress secondary-confirmation house type.
Specifically, when overseas hotel's house type of user's selection belongs to the high house type of generation service deficiency probability, OTA platforms
Sequence of the house type in OTA platforms is postponed, service deficiency probability is higher, and it is bigger to occur the risk of service deficiency, and sequence is more rearward;
Or directly close house type.
By setting different threshold intervals, the urgency that house type to that service deficiency occur makes differentiation, and passes through EBK
The logical hotel that is automatically reminded to of (Ebooking, the predetermined menu manager of electronics) or agency carries out secondary-confirmation.Wherein, EBK refers to be supplied to
Hotel is used as the system for safeguarding room availability, room rate, stock and hotel information;Agency is logical to be referred to specially be supplied to OTA platforms to provide wine
The system that the supplier of shop resource uses.Supplier can manage the content of order, house type and clearing etc., and supplier only bears
Duty is placed an order and received orders, then order is given into hotel, without being responsible for confirming price and having room without room.
Embodiment 2
As shown in Fig. 2 the present invention also provides a kind of forecasting system of the overseas hotel's house type service deficiency of OTA platforms.
The forecasting system includes first acquisition unit 1, second acquisition unit 2, model and establishes unit 3, the and of predicting unit 4
Processing unit 5.
Before the first acquisition unit 1 places an order for user, user profile is predicted;The user profile selects including user
Overseas hotel's house type information and move in day information;
The second acquisition unit is used to obtain order history corresponding with the user profile according to the user profile
Information, the same base attribute information for moving in a day historical information, supplier's channel historical information and house type;
Wherein, the order history information includes setting number of days of the overseas hotel's house type of user's selection before the lower odd-numbered day
The interior order numbers information that service deficiency occurs;
It is described to take day with moving in overseas wine same move in that day history packet included belonging to overseas hotel's house type
Be engaged in defect order numbers information and/or move in whether day is weekend information;
The same order numbers information for moving in day generation service deficiency is used to be characterized in the setting time before the lower odd-numbered day
In the order inside to place an order, the identical quantity on order for moving in day generation service deficiency after the lower odd-numbered day;
For example, nearly 30 days of lower odd-numbered day same moved in day that (with female hotel, female hotel represents the actual entity wine moved in of user
Shop) occur service deficiency order numbers refer to:Such as current lower odd-numbered day is on October 1st, 2017, and order moves in the time as 2017
On October 5, in.Now, be 1 day-September of September 30 days by single time under statistics, and with the order (October 1 was placed an order)
Corresponding house type (house type corresponds to same female hotel), and move in all orders that day is October 5 and take
The order numbers for defect of being engaged in.
Supplier's channel historical information corresponds to the attribute information of supplier including house type and/or the history of supplier is ordered
The situation of service deficiency occurs in list;Wherein, the attribute information of supplier includes whether supplier belongs to what is directly signed with OTA platforms
The class information that information and/or OTA platforms are set to supplier;
Wherein, a source of houses part for OTA platforms derives from and the direct cooperation in hotel, another part derive from acquisition hotel
House type sells the third party of power, i.e. supplier.The attribute information of supplier includes the different channels of supplier;Different channels are main
Including straight label, direct-connected supplier of group, cooperation supplier and other OTA suppliers etc..Carried for the supplier of different channels
Type of buying a house in installments is different with the mode of management room availability.
The base attribute information of the house type includes whether house type is the information for retaining room;
Wherein, the reservation room represents that hotel is supplied to the room that OTA platforms are freely sold, and OTA platforms will can be reserved
Room sold with different house type, and need not be confirmed with hotel.
The model establish unit be used for by XGBoost algorithms to the order history information, it is described together move in calendar
The base attribute information of history information, supplier's channel historical information and the house type is respectively processed, and establishes overseas wine
The service deficiency Probabilistic Prediction Model of shop house type;
Wherein, the data that model training is not involved in by using test data are pre- to the service deficiency probability of foundation
Surveying model needs to carry out model checking, and the prediction knot by recall rate and accuracy rate to the service deficiency Probabilistic Prediction Model
Fruit is evaluated.The interpretational criteria of model is in the case where recall rate is certain, and accuracy rate is higher, and the service deficiency probability is pre-
The prediction result for surveying model is more accurate.
The service deficiency full room before including validating that, confirm after full room, to shop without room, to shop without it is predetermined, confirm before appreciate
With at least one of rise in price after confirmation;
Full room is used to characterizing after user places an order before the confirmation, described in whether OTA platforms are examined with overseas hotel and had
Before overseas hotel house type, OTA platforms show the situation in the full room of overseas hotel house type;
For OTA platforms, do not know that hotel is actual usually and whether also have room, therefore after user places an order, OTA platforms
Need to be checked with hotel and (retain some special circumstances that need not be checked with hotel such as room except room belongs to), then
User's order is confirmed, confirmation i.e. herein represent the confirmations that are carried out to user's order of OTA.
Full room is used to characterizing after user places an order before the confirmation, and OTA platforms are apprised of when checking house type information with hotel
The situation in the full room of the house type;
Wherein, the result in the full room of the house type is informed user by OTA platforms after the full room of the house type is apprised of, convenient
User carries out cancelling confirmation or carries out other operations to order.
Full room is used to characterizing after user places an order after the confirmation, and when checking house type information with hotel, hotel accuses OTA platforms
Know that house type described in OTA platforms has room, after OTA platforms are completed to acknowledgement of orders, the full room of house type described in OTA platforms is informed in hotel
Situation.
Described to be used to characterizing after user places an order without room to shop, preparation does not have the overseas hotel when moving in the overseas hotel
The situation of house type;
It is described to shop without after being intended for characterizing and being used for predetermined order, preparation does not have user pre- when moving in the overseas hotel
Determine the situation of sequence information;
Appreciated before the confirmation for characterizing after user places an order, OTA platforms are apprised of when checking house type information with hotel
The price of the house type needs to go up;
Wherein, the price of the house type is needed to go up by OTA platforms after the price for being apprised of the house type needs to go up
Result feed back to user, facilitate user to carry out cancelling confirmation to order or carry out other operations.
Appreciated after the confirmation for characterizing after user places an order, when checking house type information with hotel, hotel does not have OTA platforms
Inform that house type described in OTA platforms needs to appreciate, after OTA platforms are completed to acknowledgement of orders, hotel is informed described in OTA platforms
House type needs situation about appreciating.
The predicting unit is used for the overseas wine that user's selection is predicted according to the service deficiency Probabilistic Prediction Model
The service deficiency probability of shop house type;
The expression formula of the service deficiency Probabilistic Prediction Model is:
Wherein,
Score represents the service deficiency probability, and T represents a number for decision tree, and t represents the t decision tree, ft(xi) table
Show the function of decision tree, w represents leaf weight, yiModel actual value is represented,The model predication value of t-1 tree, θ before expression
Partial derivative is represented, l represents residual error function, and γ and β are adjustable parameter,For by the model actual value and institute
The residual error function that model predication value defines is stated, for calculating the difference band between the model actual value and the model predication value
The loss come;giRepresent residual error function pairFirst derivative, hiRepresent residual error function pairSecond dervative, n represent
The quantity of training sample, i represent i-th of training sample, the integer of i span between 1~n.
The overseas hotel occupancy to be monitored can be covered to try to achieve to maximize by the service deficiency Probabilistic Prediction Model
The house type state of day.For example, the present invention the service deficiency probabilistic forecasting mould for overseas hotel past 30 days or 180 days
The type of booking rooms for inside having the order that service deficiency occurs is predicted, it is possible to achieve 80% order in covering day order, from
80% order is 60 days in advance predetermined knowable to the distribution situation of predetermined number of days in advance of overseas hotel house type, therefore can be used
The service deficiency probabilistic forecasting mould prediction of the present invention had the overseas hotel that service deficiency occurs within past 30 days or 180 days
Move in room availability corresponding to day within the 60th day after house type and lower odd-numbered day.
The processing unit is used to judge whether the overseas hotel house type service deficiency probability is more than given threshold, is sentencing
When breaking to be, corresponding Intervention Strategy is taken to the overseas hotel house type.
Wherein, the Intervention Strategy includes directly closing house type, house type is postponed in the sequence of OTA platforms, sends mail
Or by outgoing call intervention notifications user and it is automatically reminded at least one of overseas hotel's progress secondary-confirmation house type.
Specifically, when overseas hotel's house type of user's selection belongs to the high house type of generation service deficiency probability, OTA platforms
Sequence of the house type in OTA platforms is postponed, service deficiency probability is higher, and it is bigger to occur the risk of service deficiency, and sequence is more rearward;
Or directly close house type.
By setting different threshold intervals, the urgency that house type to that service deficiency occur makes differentiation, and by EBK or
The logical hotel that is automatically reminded to of agency carries out secondary-confirmation.Wherein, EBK refer to be supplied to hotel be used as safeguarding room availability, room rate, stock and
The system of hotel information;The system that the logical supplier for referring to specially be supplied to OTA platforms to provide hotel's resource of agency uses.Supply
Business can manage the content of order, house type and clearing etc., and supplier is only responsible for placing an order and receiving orders, then order is given to
Hotel, without being responsible for confirming price and having room without room.
Although the foregoing describing the embodiment of the present invention, it will be appreciated by those of skill in the art that these
It is merely illustrative of, protection scope of the present invention is defined by the appended claims.Those skilled in the art is not carrying on the back
On the premise of principle and essence from the present invention, various changes or modifications can be made to these embodiments, but these are changed
Protection scope of the present invention is each fallen within modification.
Claims (10)
1. a kind of Forecasting Methodology of the overseas hotel's house type service deficiency of OTA platforms, it is characterised in that the Forecasting Methodology includes:
S1, before user places an order, predict user profile;The user profile include user selection overseas hotel's house type information and
Move in day information;
S2, obtained according to the user profile corresponding with user profile order history information, move in together a day historical information,
The base attribute information of supplier's channel historical information and house type;
S3, by XGBoost algorithms to the order history information, described with moving in a day historical information, supplier's channel
The base attribute information of historical information and the house type is respectively processed, and the service deficiency probability for establishing overseas hotel's house type is pre-
Survey model;
S4, the service deficiency of the overseas hotel's house type selected according to the service deficiency Probabilistic Prediction Model prediction user are general
Rate;
S5, judge whether the service deficiency probability is more than given threshold, when being judged as YES, the overseas hotel house type is adopted
Take corresponding Intervention Strategy.
2. the Forecasting Methodology of the overseas hotel's house type service deficiency of OTA platforms as claimed in claim 1, it is characterised in that described
The expression formula of service deficiency Probabilistic Prediction Model is:
<mrow>
<mi>s</mi>
<mi>c</mi>
<mi>o</mi>
<mi>r</mi>
<mi>e</mi>
<mo>=</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mo>&lsqb;</mo>
<msub>
<mi>g</mi>
<mi>i</mi>
</msub>
<msub>
<mi>f</mi>
<mi>t</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mfrac>
<mn>1</mn>
<mn>2</mn>
</mfrac>
<msub>
<mi>h</mi>
<mi>i</mi>
</msub>
<msubsup>
<mi>f</mi>
<mi>t</mi>
<mn>2</mn>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
<mo>+</mo>
<mrow>
<mo>(</mo>
<mi>&gamma;</mi>
<mi>T</mi>
<mo>+</mo>
<mfrac>
<mn>1</mn>
<mn>2</mn>
</mfrac>
<mi>&beta;</mi>
<mo>|</mo>
<mo>|</mo>
<mi>w</mi>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mo>)</mo>
</mrow>
</mrow>
Wherein,
Score represents the service deficiency probability, and T represents a number for decision tree, and t represents the t decision tree, ft(xi) represent to determine
The function of plan tree, w represent leaf weight, yiModel actual value is represented,The model predication value of t-1 tree before expression,Table
Showing partial derivative, l represents residual error function, and γ and β are adjustable parameter,For by the model actual value and described
The residual error function that model predication value defines, giRepresent residual error function pairFirst derivative, hiRepresent residual error function pair
Second dervative, n represents the quantity of training sample, and i represents i-th of training sample, and i span is whole between 1~n
Number.
3. the Forecasting Methodology of the overseas hotel's house type service deficiency of OTA platforms as claimed in claim 1, it is characterised in that described
Intervention Strategy includes directly closing house type, house type is postponed in the sequence of OTA platforms, sends mail or by outgoing call intervention notifications
User carries out at least one of secondary-confirmation house type with overseas hotel is automatically reminded to.
4. the Forecasting Methodology of the overseas hotel's house type service deficiency of OTA platforms as claimed in claim 1, it is characterised in that described
Order history information includes overseas hotel's house type of user's selection and service deficiency occurs in the setting number of days before the lower odd-numbered day
Order numbers information;
It is described service occurs day to lack with moving in day history packet and include overseas wine belonging to overseas hotel's house type same move in
Sunken order numbers information and/or move in whether day is weekend information;
Same move in occurs the order numbers information of service deficiency day and is used under being characterized in the setting time before the lower odd-numbered day
In single order, the identical quantity on order for moving in day generation service deficiency after the lower odd-numbered day;
Supplier's channel historical information is corresponded in the attribute information of supplier and/or the History Order of supplier including house type
The situation of service deficiency occurs;Wherein, the attribute information of supplier includes whether supplier belongs to the information directly signed with OTA platforms
And/or the class information that OTA platforms are set to supplier;
The base attribute information of the house type includes whether house type is the information for retaining room;Wherein, the reservation room is overseas wine
Shop is supplied to the house type of OTA platform presells.
5. the Forecasting Methodology of the overseas hotel's house type service deficiency of OTA platforms as claimed in claim 1, it is characterised in that described
Service deficiency full room before including validating that, confirm after full room, to shop without room, to shop without it is predetermined, confirm before appreciate and appreciate after confirming
At least one of;
Full room is used to characterizing after user places an order before the confirmation, and OTA platforms are apprised of described when checking house type information with hotel
The situation in the full room of house type;
Full room is used to characterizing after user places an order after the confirmation, and OTA platforms when checking house type information with hotel, inform by hotel
House type described in OTA platforms has room, and after OTA platforms are completed to acknowledgement of orders, the full room of house type described in OTA platforms is informed in hotel
Situation.
Described to be used to characterizing after user places an order without room to shop, preparation does not have the overseas hotel house type when moving in the overseas hotel
Situation;
It is described to shop without after being intended for characterizing and being used for predetermined order, preparation does not have user's predetermined order when moving in the overseas hotel
The situation of single information;
Appreciated before the confirmation for characterizing after user places an order, OTA platforms are apprised of described when checking house type information with hotel
The price of house type needs to go up;
Appreciated after the confirmation for characterizing after user places an order, when checking house type information with hotel, hotel does not accuse OTA platforms
Know that house type described in OTA platforms needs to appreciate, after OTA platforms are completed to acknowledgement of orders, house type described in OTA platforms is informed in hotel
Need situation about appreciating.
6. a kind of forecasting system of the overseas hotel's house type service deficiency of OTA platforms, it is characterised in that the forecasting system includes the
One acquiring unit, second acquisition unit, model establish unit, predicting unit and processing unit;
Before the first acquisition unit places an order for user, user profile is predicted;The user profile includes the sea of user's selection
Outer hotel's house type information and move in day information;
The second acquisition unit be used for according to the user profile obtain corresponding with user profile order history information,
With the base attribute information for moving in a day historical information, supplier's channel historical information and house type;
The model establishes unit and is used to believe the order history information, the same day history of moving in by XGBoost algorithms
The base attribute information of breath, supplier's channel historical information and the house type is respectively processed, and establishes overseas hotel room
The service deficiency Probabilistic Prediction Model of type;
The predicting unit is used for the overseas hotel room that user's selection is predicted according to the service deficiency Probabilistic Prediction Model
The service deficiency probability of type;
The processing unit is used to judge whether the overseas hotel house type service deficiency probability is more than given threshold, is being judged as
When being, corresponding Intervention Strategy is taken to the overseas hotel house type.
7. the forecasting system of the overseas hotel's house type service deficiency of OTA platforms as claimed in claim 6, it is characterised in that described
The expression formula of service deficiency Probabilistic Prediction Model is:
<mrow>
<mi>s</mi>
<mi>c</mi>
<mi>o</mi>
<mi>r</mi>
<mi>e</mi>
<mo>=</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mo>&lsqb;</mo>
<msub>
<mi>g</mi>
<mi>i</mi>
</msub>
<msub>
<mi>f</mi>
<mi>t</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mfrac>
<mn>1</mn>
<mn>2</mn>
</mfrac>
<msub>
<mi>h</mi>
<mi>i</mi>
</msub>
<msubsup>
<mi>f</mi>
<mi>t</mi>
<mn>2</mn>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
<mo>+</mo>
<mrow>
<mo>(</mo>
<mi>&gamma;</mi>
<mi>T</mi>
<mo>+</mo>
<mfrac>
<mn>1</mn>
<mn>2</mn>
</mfrac>
<mi>&beta;</mi>
<mo>|</mo>
<mo>|</mo>
<mi>w</mi>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mo>)</mo>
</mrow>
</mrow>
Wherein,
Score represents the service deficiency probability, and T represents a number for decision tree, and t represents the t decision tree, ft(xi) represent to determine
The function of plan tree, w represent leaf weight, yiModel actual value is represented,The model predication value of t-1 tree before expression,Table
Showing partial derivative, l represents residual error function, and γ and β are adjustable parameter,For by the model actual value and described
The residual error function that model predication value defines, giRepresent residual error function pairFirst derivative, hiRepresent residual error function pair
Second dervative, n represents the quantity of training sample, and i represents i-th of training sample, and i span is whole between 1~n
Number.
8. the forecasting system of the overseas hotel's house type service deficiency of OTA platforms as claimed in claim 6, it is characterised in that described
Intervention Strategy includes directly closing house type, house type is postponed in the sequence of OTA platforms, sends mail or by outgoing call intervention notifications
User carries out at least one of secondary-confirmation house type with overseas hotel is automatically reminded to.
9. the forecasting system of the overseas hotel's house type service deficiency of OTA platforms as claimed in claim 6, it is characterised in that described
Order history information includes overseas hotel's house type of user's selection and service deficiency occurs in the setting number of days before the lower odd-numbered day
Order numbers information;
It is described service occurs day to lack with moving in day history packet and include overseas wine belonging to overseas hotel's house type same move in
Sunken order numbers information and/or move in whether day is weekend information;
Same move in occurs the order numbers information of service deficiency day and is used under being characterized in the setting time before the lower odd-numbered day
In single order, the identical quantity on order for moving in day generation service deficiency after the lower odd-numbered day;
Supplier's channel historical information is corresponded in the attribute information of supplier and/or the History Order of supplier including house type
The situation of service deficiency occurs;Wherein, the attribute information of supplier includes whether supplier belongs to the information directly signed with OTA platforms
And/or the class information that OTA platforms are set to supplier;
The base attribute information of the house type includes whether house type is the information for retaining room;
Wherein, the reservation room represents that hotel is supplied to the room that OTA platforms are freely sold.
10. the forecasting system of the overseas hotel's house type service deficiency of OTA platforms as claimed in claim 6, it is characterised in that described
Service deficiency full room before including validating that, confirm after full room, to shop without room, to shop without it is predetermined, confirm before appreciate and appreciate after confirming
At least one of;
Full room is used to characterizing after user places an order before the confirmation, and OTA platforms are apprised of described when checking house type information with hotel
The situation in the full room of house type;
Full room is used to characterizing after user places an order after the confirmation, and OTA platforms when checking house type information with hotel, inform by hotel
House type described in OTA platforms has room, and after OTA platforms are completed to acknowledgement of orders, the full room of house type described in OTA platforms is informed in hotel
Situation.
Described to be used to characterizing after user places an order without room to shop, preparation does not have the overseas hotel house type when moving in the overseas hotel
Situation;
It is described to shop without after being intended for characterizing and being used for predetermined order, preparation does not have user's predetermined order when moving in the overseas hotel
The situation of single information;
Appreciated before the confirmation for characterizing after user places an order, OTA platforms are apprised of described when checking house type information with hotel
The price of house type needs to go up;
Appreciated after the confirmation for characterizing after user places an order, when checking house type information with hotel, hotel does not accuse OTA platforms
Know that house type described in OTA platforms needs to appreciate, after OTA platforms are completed to acknowledgement of orders, house type described in OTA platforms is informed in hotel
Need situation about appreciating.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710994934.5A CN107679674B (en) | 2017-10-23 | 2017-10-23 | OTA platform overseas hotel room type service defect prediction method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710994934.5A CN107679674B (en) | 2017-10-23 | 2017-10-23 | OTA platform overseas hotel room type service defect prediction method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107679674A true CN107679674A (en) | 2018-02-09 |
CN107679674B CN107679674B (en) | 2021-06-11 |
Family
ID=61140383
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710994934.5A Active CN107679674B (en) | 2017-10-23 | 2017-10-23 | OTA platform overseas hotel room type service defect prediction method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107679674B (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109118056A (en) * | 2018-07-19 | 2019-01-01 | 携程计算机技术(上海)有限公司 | The processing method and system of service deficiency |
CN109376891A (en) * | 2018-11-27 | 2019-02-22 | 携程计算机技术(上海)有限公司 | The determination method and system of the full room state of people place |
CN109685234A (en) * | 2018-12-27 | 2019-04-26 | 携程计算机技术(上海)有限公司 | The processing method and system subscribed for hotel's house type |
CN110543964A (en) * | 2019-07-19 | 2019-12-06 | 深圳市跨越新科技有限公司 | XGboost model-based abnormal singleweight early warning method and system |
CN110738054A (en) * | 2019-10-14 | 2020-01-31 | 携程计算机技术(上海)有限公司 | Method, system, electronic device and storage medium for identifying hotel information in mail |
CN111080417A (en) * | 2019-12-27 | 2020-04-28 | 携程计算机技术(上海)有限公司 | Processing method for improving booking smoothness rate, model training method and system |
CN111127109A (en) * | 2019-12-27 | 2020-05-08 | 携程计算机技术(上海)有限公司 | Prediction method of different city heat values, model training method and system |
CN111144598A (en) * | 2019-12-27 | 2020-05-12 | 携程计算机技术(上海)有限公司 | Prediction method of OTA reserved room reliability, model training method and system |
CN111445046A (en) * | 2020-03-18 | 2020-07-24 | 携程计算机技术(上海)有限公司 | Hotel reservation information processing method and system, electronic equipment and storage medium |
CN111523958A (en) * | 2020-03-10 | 2020-08-11 | 携程计算机技术(上海)有限公司 | Hotel fax return processing method, system, electronic equipment and medium |
CN111832788A (en) * | 2019-04-23 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Service information generation method and device, computer equipment and storage medium |
CN111861801A (en) * | 2020-07-21 | 2020-10-30 | 携程计算机技术(上海)有限公司 | Hotel full room prediction method, system, equipment and storage medium |
CN113139746A (en) * | 2021-05-13 | 2021-07-20 | 深圳他米科技有限公司 | Hotel stay-continuing method, device, equipment and storage medium based on artificial intelligence |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103197983A (en) * | 2013-04-22 | 2013-07-10 | 东南大学 | Service component reliability online time sequence predicting method based on probability graph model |
CN103455858A (en) * | 2013-09-23 | 2013-12-18 | 东南大学 | Service-oriented system quality dynamic early-warning method |
WO2015198503A1 (en) * | 2014-06-27 | 2015-12-30 | 株式会社パラダイムシフト | Revenue management system and revenue management method |
CN106503824A (en) * | 2016-11-11 | 2017-03-15 | 成都昊腾科技有限公司 | A kind of hotel service system and many hotel service total systems |
CN106780173A (en) * | 2016-12-01 | 2017-05-31 | 携程计算机技术(上海)有限公司 | OTA hotels inventory management method and system |
-
2017
- 2017-10-23 CN CN201710994934.5A patent/CN107679674B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103197983A (en) * | 2013-04-22 | 2013-07-10 | 东南大学 | Service component reliability online time sequence predicting method based on probability graph model |
CN103455858A (en) * | 2013-09-23 | 2013-12-18 | 东南大学 | Service-oriented system quality dynamic early-warning method |
WO2015198503A1 (en) * | 2014-06-27 | 2015-12-30 | 株式会社パラダイムシフト | Revenue management system and revenue management method |
CN106503824A (en) * | 2016-11-11 | 2017-03-15 | 成都昊腾科技有限公司 | A kind of hotel service system and many hotel service total systems |
CN106780173A (en) * | 2016-12-01 | 2017-05-31 | 携程计算机技术(上海)有限公司 | OTA hotels inventory management method and system |
Non-Patent Citations (1)
Title |
---|
金艺兰: "旅游度假区服务质量对顾客满意度影响研究", 《中国优秀硕士学位论文全文数据库 经济与管理科学辑》 * |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109118056B (en) * | 2018-07-19 | 2022-02-08 | 携程计算机技术(上海)有限公司 | Service defect processing method and system |
CN109118056A (en) * | 2018-07-19 | 2019-01-01 | 携程计算机技术(上海)有限公司 | The processing method and system of service deficiency |
CN109376891A (en) * | 2018-11-27 | 2019-02-22 | 携程计算机技术(上海)有限公司 | The determination method and system of the full room state of people place |
CN109376891B (en) * | 2018-11-27 | 2022-06-21 | 携程计算机技术(上海)有限公司 | Method and system for determining full house state of residential accommodation |
CN109685234A (en) * | 2018-12-27 | 2019-04-26 | 携程计算机技术(上海)有限公司 | The processing method and system subscribed for hotel's house type |
CN109685234B (en) * | 2018-12-27 | 2023-09-05 | 携程计算机技术(上海)有限公司 | Processing method and system for hotel room reservation |
CN111832788A (en) * | 2019-04-23 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Service information generation method and device, computer equipment and storage medium |
CN111832788B (en) * | 2019-04-23 | 2024-03-29 | 北京嘀嘀无限科技发展有限公司 | Service information generation method, device, computer equipment and storage medium |
CN110543964A (en) * | 2019-07-19 | 2019-12-06 | 深圳市跨越新科技有限公司 | XGboost model-based abnormal singleweight early warning method and system |
CN110738054A (en) * | 2019-10-14 | 2020-01-31 | 携程计算机技术(上海)有限公司 | Method, system, electronic device and storage medium for identifying hotel information in mail |
CN111127109A (en) * | 2019-12-27 | 2020-05-08 | 携程计算机技术(上海)有限公司 | Prediction method of different city heat values, model training method and system |
CN111144598A (en) * | 2019-12-27 | 2020-05-12 | 携程计算机技术(上海)有限公司 | Prediction method of OTA reserved room reliability, model training method and system |
CN111080417A (en) * | 2019-12-27 | 2020-04-28 | 携程计算机技术(上海)有限公司 | Processing method for improving booking smoothness rate, model training method and system |
CN111523958A (en) * | 2020-03-10 | 2020-08-11 | 携程计算机技术(上海)有限公司 | Hotel fax return processing method, system, electronic equipment and medium |
CN111445046A (en) * | 2020-03-18 | 2020-07-24 | 携程计算机技术(上海)有限公司 | Hotel reservation information processing method and system, electronic equipment and storage medium |
CN111861801A (en) * | 2020-07-21 | 2020-10-30 | 携程计算机技术(上海)有限公司 | Hotel full room prediction method, system, equipment and storage medium |
CN113139746A (en) * | 2021-05-13 | 2021-07-20 | 深圳他米科技有限公司 | Hotel stay-continuing method, device, equipment and storage medium based on artificial intelligence |
CN113139746B (en) * | 2021-05-13 | 2023-11-14 | 深圳他米科技有限公司 | Hotel check-in method, device, equipment and storage medium based on artificial intelligence |
Also Published As
Publication number | Publication date |
---|---|
CN107679674B (en) | 2021-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107679674A (en) | The Forecasting Methodology and system of the overseas hotel's house type service deficiency of OTA platforms | |
Rezaei et al. | Port performance measurement in the context of port choice: an MCDA approach | |
Karlaftis et al. | Containership routing with time deadlines and simultaneous deliveries and pick-ups | |
Yap et al. | Dynamics of liner shipping service scheduling and their impact on container port competition | |
Li et al. | Investigating the effect of carbon tax and carbon quota policy to achieve low carbon logistics operations | |
CN110225081A (en) | Information of charging pile supplying system and method | |
CN107889069A (en) | A kind of Short Message Service Gateway system of selection, device, server and readable storage medium storing program for executing | |
CN108038637A (en) | Logistic information management system based on cloud platform | |
CN109214732A (en) | Method, device and equipment for selecting logistics objects and determining logistics line overload | |
CN109902930A (en) | A kind of auxiliary facility planning auxiliary system based on real population index | |
Wang et al. | Transshipment hub port selection for shipping carriers in a dual hub-port system | |
CN106600020A (en) | System used for scheduling station pickup services and method thereof | |
CN108960731A (en) | Express mail collection methods in full-automatic logistics transportation | |
CN107730019A (en) | User based on user's portrait retrieves method and system | |
Pomfret et al. | Global value-chains and connectivity in developing Asia–With application to the Central and West Asian region | |
Wang et al. | The effect of distance on cargo flows: a case study of Chinese imports and their hinterland destinations | |
CN108764800A (en) | Method for realizing rapid delivery of packages based on crowdsourcing public transportation system | |
CN107222871A (en) | A kind of wireless private network electric power base station planning methods of TD LTE 230 | |
CN110490391A (en) | A kind of calculation method of multimodal transport traffic program | |
CN107798411A (en) | The evaluation process method and server of a kind of regular bus route | |
Zeng et al. | Optimization of electric bus scheduling for mixed passenger and freight flow in an urban-rural transit system | |
CN105825284A (en) | Airborne logistics self-service space booking system | |
Macharis et al. | Location Analysis Model for Belgian Intermodal Terminals: towards an integration of the modal choice variables | |
CN107609961A (en) | A kind of auto form delivering system and method based on logistics | |
CN208314847U (en) | A kind of logistics transportation 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |