CN115271137A - Online booking change processing method - Google Patents
Online booking change processing method Download PDFInfo
- Publication number
- CN115271137A CN115271137A CN202210859691.5A CN202210859691A CN115271137A CN 115271137 A CN115271137 A CN 115271137A CN 202210859691 A CN202210859691 A CN 202210859691A CN 115271137 A CN115271137 A CN 115271137A
- Authority
- CN
- China
- Prior art keywords
- hotel
- transferee
- data
- credit
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003672 processing method Methods 0.000 title abstract description 7
- 238000012546 transfer Methods 0.000 claims abstract description 77
- 238000013210 evaluation model Methods 0.000 claims abstract description 19
- 238000011156 evaluation Methods 0.000 claims description 43
- 238000000034 method Methods 0.000 claims description 37
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 230000009193 crawling Effects 0.000 claims description 6
- 238000012821 model calculation Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000007635 classification algorithm Methods 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 230000006698 induction Effects 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000004904 shortening Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 2
- 238000011867 re-evaluation Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
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
- G06Q10/00—Administration; Management
- G06Q10/02—Reservations, e.g. for tickets, services or events
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
- G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0204—Market segmentation
- G06Q30/0205—Location or geographical consideration
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0206—Price or cost determination based on market factors
-
- 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]
- G06Q30/0609—Buyer or seller confidence or verification
-
- 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]
- G06Q30/0639—Item locations
-
- 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
- G06Q50/10—Services
- G06Q50/12—Hotels or restaurants
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Bioethics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an online booking change processing method, which comprises the following steps of firstly, determining associated data of hotel source price; based on the associated data of the hotel room source price, evaluating the hotel room source condition; predicting the probability of the scheduled change of the hotel room source based on the hotel room source condition; determining past hotel room source reservation data of the transferee, and determining loan credit record of the transferee; evaluating the related data based on the credit value of the transferee and analyzing the credit degree of the transferee, wherein the credit degree evaluation model is established, and the credit value of the corresponding transferee is obtained according to the related data of the transferee; determining a preliminary transfer price according to the house source change probability and the credit value of the transferee; and finally, adjusting the final transfer price according to the recommendation degree of the transferor to the hotel room resources required to be transferred, and acquiring the final transfer price by the transferor, so that the room resources are recycled, and the benefit maximization of the hotel and the user is facilitated.
Description
[ technical field ] A
The invention relates to the technical field of information, in particular to an online booking change processing method.
[ background of the invention ]
Tourism is a common thing for people, but tourism is generally planned many days in advance, and the future time has uncertainty, so that the original plan is often disturbed due to sudden things. At this time, the scheduled airline ticket hotel and the like are in disorder due to the plan, and the booking needs to be cancelled. Firstly, great uncertainty is brought to the hotel, and the originally reserved rooms can not be sold in empty rooms. For the traveler user, the loss is also due to the cancellation of the default money representing a portion of the reimbursement. Resulting in a double output state for both the hotel and the user. Thus, if the transfer of hotel rooms can be planned when the plan is not changed, this can be made more controlled. But the hotel changes from not allowing transfer to encounter a plurality of new problems, such as how to determine the transfer price, whether the transfer can attract other users to buy the house resources available for transfer, and how to solve the problem if the transferee changes again after the transfer is completed. These new problems are problems that have not yet been well solved.
[ summary of the invention ]
The invention provides an online booking change processing method, which mainly comprises the following steps:
determining hotel price associated data; based on hotel price associated data, evaluating a house source condition; predicting the probability of the house source reservation change based on the hotel house source condition; determining the credit value evaluation related data of the transferee and protecting privacy; evaluating related data based on the credit value of the transferee, and analyzing the credit degree of the transferee; determining a preliminary transfer price according to the house source change probability and the credit value of the transferee; adjusting the final transfer price according to the recommendation degree of the transferor;
further optionally, the determining hotel price associated data comprises:
firstly, determining hotel price associated data comprising hotel geographic environment data and hotel internal condition data; the hotel geographic environment data comprises living facilities, medical conditions, traffic conditions and associated scenic spots within 1 kilometer of the periphery; the method comprises the steps of obtaining POI point data of hotel geographic environment data by slice crawling a Baidu map; the internal condition data of the hotel comprises hotel source quantity, hotel score and hotel score number; the method comprises the steps that public data of an OTA platform are crawled, and hotel internal condition data are obtained; collecting the house source preset purpose of the user, and carrying out induction and arrangement; and taking the average value of the hotel room source scores at the same price as the hotel room source score with the score of being empty.
Further optionally, the evaluating the house source condition based on the hotel price correlation data comprises:
analyzing the condition of the house source according to different related parameters of the house source; firstly, establishing a house source condition evaluation system, namely K = Ev S1+ Ic S2, wherein K is the condition of a hotel house source, ev is the external value of the hotel, ic is the internal environment condition of the hotel, and S1 and S2 are different weights; normalizing the Ev and Ic values in the selected hotel sample data range to enable the values to be between 0 and 1; finally, evaluating the house source condition according to a house source condition evaluation system; the method comprises the following steps: analyzing the house source external value based on the house source geographic environment data; analyzing the internal conditions of the house sources based on the data of the hotel;
the method for analyzing the house source external value based on the house source geographic environment data specifically comprises the following steps:
first, hotel room source geographic environment data including hotel peripheral living facilities, medical conditions, traffic conditions and associated attractions are determined. The living facilities around the hotel are determined by the POI points of catering, business, shopping, leisure and common scenic spots within 1 kilometer around the hotel. Medical conditions around the hotel are determined by the number of POI points in the hospital within 1 kilometer of the hotel. The traffic conditions around the hotel are determined by the POI points of a bus station, a subway station, a passenger station, a high-speed rail station and an airport within the range of 1 kilometer around the hotel. The related scenic spots around the hotel are determined by the number of POI points in the scenic spots of 4A and above levels within 1 kilometer around the hotel. Analyzing the house source value of the hotel according to various geographic environment condition data around the hotel, and establishing a hotel house source external value model, namely Ev = N1 x Lv + N2 x Mv + N3 x Tv + N4 x Sv, wherein the Ev is the external value of the hotel house source, the N1, N2, N3 and N4 are different weights, the Lv is the number of POI points of living facilities around the hotel, the Mv is the number of POI points of medical facilities around the hotel, the Tv is the number of POI points of traffic facilities around the hotel, and the Sv is the number of POI points of related scenic spots around the hotel; when the demand of the transferor is matched with the number of corresponding POI points within 1 kilometer around the hotel room source, the corresponding weight is taken as 1, otherwise, the corresponding weight is taken as 0; when the target requirement of the transferor is not similar to the specific type of POI point, taking the standardized weight of N1, N2, N3 and N4; and finally, calculating the external value of the target hotel room source according to the model.
Based on hotel's own data, analysis house source internal condition specifically includes:
firstly, determining self data of the hotel room sources, including the amount of the hotel room sources, the hotel scores and the hotel score number. Analyzing the reservation condition of the hotel according to each condition data of the hotel, and establishing a hotel room source internal condition model, namely Ic = M1 Rn + M2 He + M3 Ed, wherein Ev is the external value of the hotel room source, M1, M2 and M3 are different weights, rn is the number of the hotel room source, he is the hotel score, and Ed is the hotel evaluation number; carrying out normalization standardization processing on each index, and unifying dimensions; and finally, calculating the internal conditions of the target hotel according to the model.
Further optionally, predicting the probability of the house source booking change based on the hotel house source condition comprises:
predicting the probability of the house source reservation change according to the relevant parameters of the hotel house source condition; firstly, determining a reference standard for judging the condition of a room source of a hotel, wherein the reference standard comprises an external value and an internal condition of the room source of the hotel; secondly, establishing a prediction model of the scheduled change probability of the hotel room source, namely: v = (1-K) × 100%; the V is the house source change probability caused by the quality degree of the house source condition; k is an excellent evaluation index of the drugstore house source obtained by integrating various external and internal conditions; finally, pre-estimating the change probability of the house source according to a prediction model of the change probability of the house source of the hotel; the transfer probability of the hotel room source applying for transfer is determined to be V =1.
Further optionally, the determining the transferee reputation value rating related data and privacy protecting comprises:
firstly, determining relevant data of the credit value evaluation of the transferee, wherein the relevant data comprises past hotel reservation records of the transferee, personal information of the transferee and loan records; secondly, storing the credit value evaluation related data of the transferee and the constraint information in a database, enabling the related data in the database to be uniquely corresponding to the identity information of the transferee, and inputting the unique identity information to call the related data of the transferee; finally, the privacy information of the transferee is encrypted by using a homomorphic encryption method, the data is operated on the premise that the encrypted privacy information is not decrypted, and the operation result obtains the result same as the plaintext operation through a decryption algorithm; the method comprises the following steps: determining past predetermined data of the transferee; determining a lender credit record;
the determining of the past scheduled data of the transferee specifically includes:
firstly, determining past reservation data of a hotel room source, wherein the past reservation data comprises hotel reservations of a transferee and a successful check-in record, and a hotel reservation cancellation record and a hotel reservation non-cancellation and non-check-in record are determined; crawling OTA platform data to obtain past scheduled change records of a transferee; performing data preprocessing on the acquired user data, classifying the data and extracting reservation condition data, namely index = { the number of times of successful hotel stay in and the number of times of unsuccessful hotel stay in and hotel stay in }, wherein the index is the historical reservation condition of the transferee; and finally, the acquired data is subjected to data cleaning, and missing data, repeated data and error data are deleted.
The determining of the lending credit record of the transferee specifically comprises the following steps:
firstly, determining credit record information of a transferee, including the age, sex, native place, cultural degree, income condition, work information, normal repayment times and overdue times of the borrowed amount of the network lending platform; acquiring a credit record of a transferee through a data warehouse of a bank credit card center; different degrees of missing values present in the raw data acquired by the data warehouse are interpolated. Determining the missing value of the credit record, including the missing value of the debit demographic information and the missing value of the credit record of the borrower. Interpolating a missing value corresponding to the population characteristic information of the lender by adopting a K neighbor classification algorithm, and assigning the missing value corresponding to the credit record of the lender to be 0; and the privacy information of the transferee is subjected to encryption and decryption algorithm processing, so that the information security is ensured.
Further optionally, the evaluating the related data based on the transferee reputation value, wherein analyzing the transferee reputation comprises:
analyzing the credit degree of the transferee based on the past hotel reservation data and the loan credit record of the transferee; classifying and summarizing the data, and matching the corresponding past hotel reservation data with loan credit record data, namely index = { age, gender, cultural degree, income condition, work information, normal repayment times and overdue times of the loan amount of a network loan platform, times of successful check-in and reservation of the hotel, times of unsuccessful check-in and reservation of the hotel }, wherein the index is a credit value evaluation related index of the transferee; preprocessing each variable of the hotel reservation data and the loan credit record of the transferee by using a tail shortening method; normalizing each variable of the hotel reservation data and the loan credit record of the transferee, and unifying dimensions; the method comprises the following steps: establishing a credibility evaluation model; acquiring a corresponding credit value according to the data of the transferee;
the establishing of the credibility evaluation model specifically comprises the following steps:
constructing a credit value evaluation model by using a BP neural network based on the personal information and credit records of the transferee; calculating the weight occupied by each variable of the personal information and credit record of the transferee on the contribution degree of the personal reputation degree value by using an entropy weight method; adjusting training parameters of the BP neural network model based on the calculated personal information of the transferee and the weight of each variable of the credit record; determining a training sample and a testing sample of the neural network according to the reputation value evaluation related index data; carrying out artificial neural network training on the sample by adopting a trainlm function; and testing the reputation value evaluation model through the test set, adjusting parameters, and continuously training to improve the accuracy of the model.
The obtaining of the corresponding reputation value according to the transferee data specifically includes:
firstly, evaluating a credit degree value of a transferee based on a credit degree evaluation model with high accuracy obtained by neural network training; acquiring the sorted and summarized data of the transferee, and inputting the data into a credibility evaluation model for processing; determining data of a transferee, wherein the data comprises past hotel reservation records of the transferee, personal identity information of the transferee and bank loan records; calculating to obtain a credit degree value R of the transferee based on model prediction, and obtaining a corresponding credit degree evaluation result according to a credit degree evaluation standard; the error of the model calculation result is reduced by multiple operations; presenting the reputation degree evaluation result to the transferee; and the transferee applies for re-evaluation on the evaluation result of the personal credibility with objection.
Further optionally, the determining the preliminary transfer price according to the house resource change probability and the transferee reputation value includes:
determining a predetermined transfer house source price finally obtained by the transferee based on the house source change probability and the credit value of the transferee; constructing an assignment price prediction model: TP = IP-IP V1-IP (R-70)/100N 2); the TP is an assignment price; the IP is the initial price of the hotel room source; v is the predicted hotel room source change probability; the R is a credit degree numerical value of the transferee obtained by model calculation; the N1 and the N2 are different weights; the IP V N1 is a reduction value of the hotel room source change probability to the transfer price; the IP (R-70)/100N 2 is the increment/decrement value of the credit value of the transferee to the house source price; and finally, primarily determining the transfer price according to the transfer price prediction model.
Further optionally, the adjusting the final transfer price according to the recommended degree of the transferor includes:
firstly, determining recommendation degree related parameters of a transferor, wherein the recommendation degree related parameters comprise refreshing click frequency of the transferor, a consultation message reply average time interval, a time interval of a specified check-in time and a time interval exceeding the specified check-in time; secondly, adjusting the assignment price prediction model based on the recommendation degree related parameters; constructing a final transfer price prediction model, namely: FP = TP + CF/30 + M1+ TP/RT + M2+ TP + IT/24 + M3-TP + OT/24 + M4; the FP is the transfer price received by the final transferee, and the TP is the transfer price primarily calculated by the model; the CF is the refreshing click frequency of the transferor to the house resources to be transferred, and the unit is times/hour; the RT is the average reply time interval of the transferor to the transferee consultation information, and the unit is minutes; the IT is the time interval of the residence time specified by the house resources, and the unit is hour; the OT is a time interval of specified check-in time, and the unit is hour; the TP CF/30 M1 is the increment value of the transfer price of the refreshing click frequency of the transferor; the TP/RT M2 is an increasing value of the price of the transfer room for the average time interval of the reply of the consultation messages; TP (IT/24) M3 is an increment value of transfer price in a time interval corresponding to the specified check-in time; TP (OT/24) M4 is a reduction value of transfer price for a time interval exceeding a specified check-in time; if the time interval OT exceeding the specified check-in time is more than 6 hours, the house resource transfer information is invalid; and finally, determining the transfer price acquired by the final transferee according to the final transfer price prediction model.
The technical scheme provided by the embodiment of the invention has the following beneficial effects:
the invention obtains the house sources with variation possibility from a large number of hotel house sources, aggregates and recommends the house sources to become another product available for purchase, and transfers the discounted price to the buyer with uncertain travel time according to whether a person actually varies, thereby realizing the reutilization of the house sources and being beneficial to the maximization of benefits of the hotel and the user.
[ description of the drawings ]
FIG. 1 is a flow chart of an online subscription change processing method according to the present invention.
FIG. 2 is a flow chart of the operation of an online reservation change handling method of the present invention.
[ detailed description ] embodiments
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings and specific embodiments.
FIG. 1 is a flow chart of an online subscription change processing method according to the present invention. As shown in fig. 1, a method for processing online subscription change in this embodiment may specifically include:
step 101, determining hotel price associated data.
Referring to fig. 2, first, hotel price associated data, including hotel geographic environment data and hotel internal condition data, is determined. The hotel geographic environment data includes living facilities, medical conditions, traffic conditions and associated scenic spots within 1 kilometer of the surroundings. And (4) acquiring POI point data of the hotel geographic environment data by slicing and crawling the Baidu map. The internal condition data of the hotel comprises hotel source amount, hotel score and hotel score number. The method comprises the steps that public data of an OTA platform are crawled, and hotel internal condition data are obtained; collecting the house source preset purpose of the user, and carrying out induction and arrangement; and taking the average value of the hotel room source scores at the same price as the hotel room source score with the score of being empty. At most 400 POI point data can be returned in the rectangular range of the Baidu map API, for example, the Suzhou region is cut into 40 by 40 slices to crawl the POI point data of the scenic spots; for example, zhang Sanchao hotel room source A aims at going to the West lake of 5A level scenic spot, liqulian hotel room source B aims at visiting a doctor in the Hospital, and Wangwu hotel room source C aims at going to M city for vacation; for example, if the hotel source score for price 158 is null, then the average of all hotel scores for prices 100-200 dollars is taken as its score.
And step 102, evaluating the condition of the house source based on the hotel price associated data.
And analyzing the condition of the house source according to different related parameters of the house source. Firstly, establishing a house source condition evaluation system, namely K = Ev S1+ Ic S2, wherein K is the condition of a hotel house source, ev is the external value of the hotel, ic is the internal environment condition of the hotel, and S1 and S2 are different weights; normalizing the Ev and Ic values in the selected hotel sample data range to enable the values to be between 0 and 1; finally, evaluating the house source condition according to a house source condition evaluation system; the weight satisfies the condition of S1+ S2= 1; for example, the evaluation system of the hotel room source condition is excellent (0.90-1), good (0.80-0.89), general (0.60-0.79) and poor (0-0.59). For example, if the external value parameter of the hotel room source a is 0.66, the parameter of the hotel internal environmental condition is 0.72, i.e., ev =0.66, ic =0.72, and the weights S1 and S2 are 0.45 and 0.55, respectively, then the hotel room source status K =0.687. For example, hotel room source status values of 0.433,0.671,0.853, and 0.956 respectively indicate that the status of the room source a is poor, the status of the room source B is general, the status of the room source C is good, and the status of the room source D is excellent.
And analyzing the house source external value based on the house source geographic environment data.
Firstly, determining hotel room source geographic environment data comprising living facilities around a hotel, medical conditions, traffic conditions and associated scenic spots. The living facilities around the hotel are determined by the POI points of catering, business, shopping, leisure and common scenic spots within 1 kilometer around the hotel. Medical conditions around the hotel are determined by the number of POI points in the hospital within 1 kilometer of the hotel. The traffic conditions around the hotel are determined by the POI points of the bus station, the subway station, the passenger station, the high-speed rail station and the airport within the range of 1 kilometer around the hotel. The related scenic spots around the hotel are determined by the number of POI points in the scenic spots of 4A and above levels within 1 kilometer around the hotel. Analyzing the house source value of the hotel according to various geographic environment condition data around the hotel, and establishing a hotel house source external value model, namely Ev = N1 Lv + N2 Mv + N3 Tv + N4 Sv, wherein the Ev is the external value of the hotel house source, the N1, N2, N3 and N4 are different weights, the Lv is the number of POI points of living facilities around the hotel, the Mv is the number of POI points of medical facilities around the hotel, the Tv is the number of POI points of traffic facilities around the hotel, and the Sv is the number of POI points of related scenic spots around the hotel; when the demand of the transferor is matched with the number of corresponding POI points within 1 kilometer around the hotel room source, the corresponding weight is taken as 1, otherwise, the corresponding weight is taken as 0; when the target requirement of the transferor is not similar to the specific type of POI point, taking the standardized weight of N1, N2, N3 and N4; and finally, calculating the external value of the target hotel room source according to the model. The weight satisfies the condition of N1+ N2+ N3+ N4= 1; for example, when the goal of the zhang san booked hotel source is to go to M city for vacation, 5 clothing stores, 3 restaurants, 1 billiard hall, 1 hospital, 2 bus stations, 1 subway station, 0 4A and above scenic spots are located in one kilometer around the hotel source a, lv =9, mv =1, tv =3, sv =0, the weights N1, N2, N3, N4 are sequentially 0.22,0.16,0.38,0.24, and the external value Ev =3.28 of the hotel source a; for example, the purpose of the lie four reservation hotel source is to go to a west lake scenic spot of M city, 5 clothing stores, 3 restaurants, 1 billiard hall, 1 hospital, 2 bus stations, 1 subway station, 0 seat 4A and above scenic spots are located within one kilometer around the hotel source B, lv =9, mv =1, tv =3, sv =0, the weights N1, N2, N3, N4 are sequentially 0,1, and the external value Ev =0 of the house source a. For example, in the normalization processing of the sample data of the hotel room source, if the maximum value of Ev in the sample is 8 and the Ev value of the hotel room source B is 3.5, the normalization processing is performed to obtain the normalized Ev value of the hotel room source B as 0.4375.
And analyzing the internal conditions of the house source based on the data of the hotel.
Firstly, determining self data of hotel room sources, including hotel room source amount, hotel scores and hotel score. Analyzing reservation conditions of the hotel according to various condition data of the hotel, and establishing a hotel room source internal condition model, namely Ic = M1+ Rn + M2+ He + M3 + Ed, wherein Ev is the external value of the hotel room source, M1, M2 and M3 are different weights, rn is the number of the hotel room sources, he is the hotel score, and Ed is the hotel evaluation number; carrying out normalization standardization processing on each index, and unifying dimensions; and finally, calculating the internal conditions of the target hotel according to the model. The weight satisfies the condition of M1+ M2+ M3= 1; normalized formula X' = X/X (MAX); the preset threshold value of the hotel score is 5 points; for example, if the room source amount of hotel room source C is 50, the hotel score is 4.8, the number of hotel evaluations is 980, the maximum value of the sample statistical room source amount is 120, the maximum value of the hotel evaluations is 5, and the maximum value of the number of hotel evaluations is 2000, then Rn =0.42, he =0.96, ed =0.49 can be obtained after normalization; for example, if the amount of the house source C of the hotel is 50, the hotel score is 4.8, and the hotel score number is 980, rn =0.42, he =0.96, ed =0.49, the weights N1, N2, N3, N4 are 0.27,0.40,0.33, and the intrinsic value Ic of the house source C =0.66.
And 103, predicting the probability of the house source reservation change based on the state of the house source of the hotel.
Predicting the probability of the house source reservation change according to the relevant parameters of the hotel house source condition; firstly, determining the judgment reference standard of the hotel room source condition, including the external value and the internal condition of the hotel room source. Secondly, establishing a prediction model of the scheduled change probability of the hotel room source, namely: v = (1-K) × 100%; and V is the house source change probability caused by the quality degree of the house source condition. K is an excellent evaluation index of the bouillon source obtained by integrating various external and internal conditions; and finally, pre-estimating the change probability of the house source according to the prediction model of the change probability of the house source of the hotel. The transfer probability V =1 of the hotel room source applying for transfer is determined; for example, the evaluation system for the hotel room source condition is excellent (0.90-1), good (0.80-0.89), general (0.60-0.79) and poor (0-0.59). For example, the status K =0.66 of the hotel room source a, the status of the hotel room source is general, and the variation probability V =34% of the hotel room source.
And step 104, determining the reputation value of the transferee to evaluate related data and protecting privacy.
Firstly, determining data related to evaluation of a credit value of a transferee, wherein the data includes past hotel reservation records of the transferee, personal information of the transferee and loan records; secondly, storing the credit value evaluation related data of the transferee and the constraint information in a database, enabling the related data in the database to be uniquely corresponding to the identity information of the transferee, and inputting unique identity information to call the related data of the transferee; and finally, encrypting the private information of the transferee by using a homomorphic encryption method, operating the data on the premise of not decrypting the encrypted private information, and obtaining the same result as the plaintext operation through a decryption algorithm according to the operation result. For example: the core of the homomorphic encryption method is that the encrypted data is directly used for operation, and the same result as the result obtained by directly using the unencrypted data can be obtained through the operation result through a decryption algorithm. If a homomorphic encryption function F and a plaintext M exist, M is encrypted to obtain a ciphertext F (M) = M, and the ciphertext M is transmitted from a transmitting party to a receiving party, the receiving party does not need to decrypt the ciphertext M, but directly uses the ciphertext M to perform operation C to obtain an operation result C (M) = N, and then performs decryption calculation F on N to obtain a result N = F (N). On the other hand, since the result N, i.e., C (M) = N, can be obtained by directly performing the operation C using the plaintext M, the receiver can normally calculate the data to obtain the correct result without knowing the plaintext content because F (C (F (M))) = C (M), i.e., F (C (M)) = C (F (M)), in the operation.
Past reservation data for the transferee is determined.
Firstly, determining past reservation data of a hotel source, wherein the past reservation data comprises hotel reservation and successful check-in records of a transferee, and the hotel reservation cancellation record and the hotel reservation non-cancellation and check-in records are recorded; crawling OTA platform data to obtain past scheduled change records of a transferee; performing data preprocessing on the acquired user data, classifying the data and extracting reservation condition data, namely index = { the number of times of successful hotel stay in and the number of times of unsuccessful hotel stay in and hotel stay in }, wherein the index is the historical reservation condition of the transferee; and finally, the acquired data is subjected to data cleaning, and missing data, repeated data and error data are deleted. Hotels that cannot be cancelled that exceed a predetermined cancellation deadline set by the hotel; for example, zhang san has 10 hotels reserved and successfully checked in records, has 3 hotels reserved but then cancelled records, has 1 hotel reserved and has no record of cancellation or check in, zhang san = {10,4}.
A lender's lending credit record is determined.
Firstly, determining credit record information of a transferee, including the age, sex, native place, cultural degree, income condition, work information, normal repayment times and overdue times of the borrowed amount of the network lending platform; acquiring a credit record of a transferee through a data warehouse of a bank credit card center; different degrees of missing values present in the raw data acquired by the data warehouse are interpolated. Determining the missing value of the credit record, including the missing value of the debit demographic information and the missing value of the credit record of the borrower. Interpolating a missing value corresponding to the population characteristic information of the lender by adopting a K neighbor classification algorithm, and assigning the missing value corresponding to the credit record of the lender to be 0; the privacy information of the transferee is subjected to encryption and decryption algorithm processing, and the information security is ensured; for example, for the income situation of the zhangsan missing, the K-nearest neighbor classification algorithm is adopted for interpolation, firstly, K samples closest to zhangsan feature information are searched, secondly, the mode of the corresponding variable values of the K samples is used as the population information feature value of zhangsan missing, and zhangsan population feature information is interpolated.
And 105, evaluating the related data based on the credit value of the transferee, and analyzing the credit degree of the transferee.
Analyzing the credit degree of the transferee based on the past hotel reservation data and the loan credit record of the transferee; classifying and summarizing the data, and matching the reservation data of the corresponding past hotel with the credit loan record data, namely index = { age, sex, cultural degree, income condition, work information, normal repayment times and overdue times of the borrowed amount of the network lending platform, times of successfully booking the hotel by check in and times of unsuccessfully booking the hotel by check in }, wherein the index is a related index for evaluating the credit value of the transferee; preprocessing all variables of hotel reservation data and loan credit records of the transferee by using a tail shortening method; normalizing each variable of the hotel reservation data and the loan credit record of the transferee, and unifying dimensions; the tail-narrowing method set the outlier level to 5% and replaced the observed values above 0.95 quantile with 0.95 quantile. For example, the number of overdue times of Zhang III loan is 50, far beyond the normal level, the number of overdue times of the sample at 0.95 quantile of the index is 10, and therefore the number of overdue times of Zhang III is taken as 10; age (20-25 years, 45 years old or older) =0, (36-40 years) =1, (31-35 years old, 41-45 years old) =2, (26-30 years old) =3; sex male =0, female =1; the culture degree is not =0, the specialty =1, the family =2, the master and above =3; work information is temporarily not provided =0, private owner =1, salary family =2, (8 k-15 k) =3, official =4; for example, zhang three 35 years old, male, academic department, monthly income 10000, officer, loan normal repayment times 13 times, overdue times 1 time, successful hotel check-in times 42 times, unsuccessful hotel check-in times 4 times, maximum monthly income value 35000 after treatment by tail contraction method, maximum loan times 60, maximum overdue times 10 times, maximum successful check-in times 75 times, maximum unsuccessful check-in times 20 times, zhang three = {2,0,2,0.28,4,0.22,0.1,0.56,0.2}.
And establishing a credit evaluation model.
Constructing a credit value evaluation model by using a BP neural network based on the personal information and credit records of the transferee; calculating the weight occupied by each variable of the personal information and credit record of the transferee on the contribution degree of the personal credit degree value by using an entropy weight method; adjusting training parameters of the BP neural network model based on the calculated personal information of the transferee and the weight of each variable of the credit record; determining a training sample and a testing sample of the neural network according to the reputation value evaluation related index data; carrying out artificial neural network training on the sample by adopting a trainlm function; the credit value evaluation model is tested through the test set, parameters are adjusted, and the model is trained continuously, so that the accuracy of the model is improved;
The input layer of the BP neural network includes 9 indices: the system comprises the following steps of age, gender, culture degree, income condition, working information, normal repayment times and overdue times of the borrowed amount of the network lending platform, times of successful hotel entrance and hotel reservation and times of unsuccessful hotel entrance and hotel reservation, wherein n =9 is an output layer, namely a credit value evaluation result, namely m =1 is an output layer, and 5 layers are hidden layers. Adopting a Sigmoid function and a linear function as an activation function in the BP neural network; and adopting the mean square error as an error function of the BP neural network.
And acquiring a corresponding credit value according to the transferee data.
Firstly, evaluating a credit degree value of a transferee based on a credit degree evaluation model with high accuracy obtained by neural network training; acquiring collated and summarized transferee data, and inputting the collated and summarized transferee data into a credibility evaluation model for processing; determining data of a transferee, wherein the data comprises past hotel reservation records of the transferee, personal identity information of the transferee and bank loan records; calculating to obtain a credit degree value R of the transferee based on model prediction, and obtaining a corresponding credit degree evaluation result according to a credit degree evaluation standard; the error of the model calculation result is reduced by multiple operations; presenting the reputation degree evaluation result to the transferee; the transferee applies for and reevaluates the personal reputation evaluation result with objection; for example, the evaluation system of the credit rating of the transferee is excellent (90-100), good (80-89), general (60-79), and poor (0-59); for example, if the value of the credit worthiness of the transferee is 92, 77, 83, and 55, respectively, the credit status of the transferee a is excellent, the credit status of the transferee B is general, the credit status of the transferee C is good, and the credit status of the transferee D is poor.
And step 106, determining an initial transfer price according to the house source change probability and the credit value of the transferee.
Determining a predetermined transfer house source price finally obtained by the transferee based on the house source change probability and the credit value of the transferee; constructing a transfer price prediction model: TP = IP-IP V1-IP (R-70)/100N 2); the TP is an assignment price; the IP is the initial price of the hotel room source; v is the predicted change probability of the hotel room source; the R is a credit degree numerical value of the transferee obtained by model calculation; the N1 and the N2 are different weights; the IP V N1 is a reduction value of the change probability of the hotel room source to the transfer price; the IP (R-70)/100N 2 is the increment/decrement value of the credit value of the transferee to the house source price; finally, primarily determining the transfer price according to the transfer price prediction model; the higher the change probability of the hotel room source is, the lower the transfer price is; the credit degree value R of the transferee is greater than 70, and the higher the credit degree value is, the larger the transfer price reduction value is; the credit degree value R of the transferee is less than 70, and the lower the credit degree value is, the larger the transfer price growth value is; for example, the predicted initial price IP =210 of the hotel room source, the assignment probability V =56%, the value of the credit degree of the transferee R =86, the weight N1=0.2, N2=0.4, and TP =173.0.
And step 107, adjusting the final transfer price according to the recommendation degree of the transferor.
Firstly, determining recommendation degree related parameters of a transferor, wherein the recommendation degree related parameters comprise refreshing click frequency of the transferor, a consultation message reply average time interval, a time interval of a specified check-in time and a time interval exceeding the specified check-in time; secondly, adjusting the assignment price prediction model based on the recommendation degree related parameters; constructing a final transfer price prediction model, namely: FP = TP + CF/30 + M1+ TP/RT + M2+ TP + IT/24 + M3-TP + OT/24 + M4; the FP is the transfer price received by the final transferee, and the TP is the transfer price primarily calculated by the model; the CF is the refreshing click frequency of the transferor to the house resources to be transferred, and the unit is times/hour; the RT is the average reply time interval of the transferor to the transferee consultation information, and the unit is minutes; the IT is the time interval of the residence time specified by the house resources, and the unit is hour; the OT is a time interval of specified check-in time, and the unit is hour; TP CF/30 M1 is the increment value of the transfer price of the refresh click frequency of the transferor; the TP/RT M2 is an increasing value of the price of the transfer room for the average time interval of the reply of the consultation messages; TP (IT/24) M3 is an increment value of transfer price in a time interval from the specified residence time; TP (OT/24) M4 is a reduction value of transfer price for a time interval exceeding a specified check-in time; if the time interval OT exceeding the specified check-in time is more than 6 hours, the house resource transfer information is invalid; finally, determining the transfer price obtained by the final transferee according to the final transfer price prediction model; when the final transfer price exceeds the house source initial price, taking the house source initial transfer price as the house source final transfer price; for example, when the initial transfer price TP =173.04, the refresh click frequency CF of the transferor =18 times/hour, the average time interval RT of the reply of the consultation message =5 minutes, the time interval IT =16 hours from the prescribed check-in, the time interval OT exceeding the prescribed check-in time =0 hour, and the weights M1, M2, M3, M4 are 0.05,0.1, 0.3 in this order, the final transferee obtains the price FP =193.228.
Claims (8)
1. An online subscription change handling method, the method comprising:
determining hotel price associated data; evaluating the condition of the house resources based on the hotel price associated data, wherein the evaluating the condition of the house resources based on the hotel price associated data specifically comprises: analyzing the house source external value based on the house source geographic environment data, and analyzing the house source internal condition based on the hotel data; predicting the probability of the house source reservation change based on the hotel house source condition; determining the reputation value of the transferee to evaluate the related data and protect the privacy, wherein the determining the reputation value of the transferee to evaluate the related data and protect the privacy specifically comprises the following steps: determining past preset data of a transferee and determining a loan credit record of the transferee; evaluating the related data based on the credit value of the transferee, analyzing the credit degree of the transferee, and specifically comprising: establishing a credit rating evaluation model, and acquiring a corresponding credit value according to the data of an acquirer; determining a preliminary transfer price according to the house source change probability and the credit value of the transferee; and adjusting the final transfer price according to the recommendation degree of the transferor.
2. The method of claim 1, wherein the determining hotel price associated data comprises:
firstly, determining hotel price associated data, including hotel geographic environment data and hotel internal condition data; the hotel geographic environment data comprises living facilities, medical conditions, traffic conditions and associated scenic spots within 1 kilometer of the periphery; the method comprises the steps of obtaining POI point data of hotel geographic environment data by carrying out slice crawling on a Baidu map; the internal condition data of the hotel comprises hotel room source amount, hotel score and hotel score number; the method comprises the steps that public data of an OTA platform are crawled, and hotel internal condition data are obtained; collecting the house source preset purpose of the user, and carrying out induction and arrangement; and taking the average value of the hotel room source scores at the same price as the hotel room source score with the score of being empty.
3. The method of claim 1, wherein evaluating the house-source condition based on the hotel price correlation data comprises:
analyzing the condition of the house source according to different related parameters of the house source; firstly, establishing a house source condition evaluation system, namely K = Ev S1+ Ic S2, wherein K is the condition of a hotel house source, ev is the external value of the hotel, ic is the internal environment condition of the hotel, and S1 and S2 are different weights; normalizing the Ev and Ic values in the selected hotel sample data range to enable the values to be between 0 and 1; finally, evaluating the house source condition according to a house source condition evaluation system; the method comprises the following steps: analyzing the house source external value based on the house source geographic environment data; analyzing the internal conditions of the house sources based on the data of the hotel;
the method for analyzing the house source extrinsic value based on the house source geographic environment data specifically comprises the following steps:
firstly, determining hotel room source geographic environment data comprising living facilities around a hotel, medical conditions, traffic conditions and associated scenic spots; wherein, the living facilities around the hotel are determined by the POI points of catering, business, shopping, leisure and common scenic spots within 1 kilometer around the hotel; medical conditions around the hotel are determined by the POI point number of the hospital within 1 kilometer around the hotel; the traffic conditions around the hotel are determined by the POI points of a bus station, a subway station, a passenger station, a high-speed rail station and an airport within 1 kilometer around the hotel; the related scenic spots around the hotel are determined by the number of POI points in the scenic spots of 4A and above levels within 1 kilometer around the hotel; analyzing the house source value of the hotel according to various geographic environment condition data around the hotel, and establishing a hotel house source external value model, namely Ev = N1 x Lv + N2 x Mv + N3 x Tv + N4 x Sv, ev is the external value of the hotel house source, N1, N2, N3 and N4 are different weights, lv is the number of POI points of living facilities around the hotel, mv is the number of POI points of medical facilities around the hotel, tv is the number of POI points of traffic facilities around the hotel, and Sv is the number of POI points of associated scenic spots around the hotel; when the demand of the transferor is matched with the number of corresponding POI points within 1 kilometer around the hotel room source, the corresponding weight is taken as 1, otherwise, the corresponding weight is taken as 0; when the target requirement of the transferor is not similar to the specific type of POI point, taking the standardized weight of N1, N2, N3 and N4; finally, calculating the external value of the target hotel room source according to the model;
based on hotel's own data, analysis house source internal condition specifically includes:
firstly, determining self data of hotel room sources, including hotel room source amount, hotel scores and hotel score number; analyzing the reservation condition of the hotel according to each condition data of the hotel, and establishing a hotel room source internal condition model, namely Ic = M1 Rn + M2 He + M3 Ed, wherein Ev is the external value of the hotel room source, M1, M2 and M3 are different weights, rn is the number of the hotel room source, he is the hotel score, and Ed is the hotel evaluation number; carrying out normalization standardization processing on each index, and unifying dimensions; and finally, calculating the internal conditions of the target hotel according to the model.
4. The method of claim 1, wherein predicting the probability of a room source booking shift based on hotel room source conditions comprises:
predicting the probability of the house source reservation change according to the relevant parameters of the hotel house source condition; firstly, determining a reference standard for judging the condition of a room source of a hotel, wherein the reference standard comprises an external value and an internal condition of the room source of the hotel; secondly, establishing a prediction model of the scheduled change probability of the hotel room source, namely: v = (1-K) × 100%; v is the house source change probability caused by the quality degree of the house source condition; k is an excellent evaluation index of the drugstore house source obtained by integrating various external and internal conditions; finally, pre-estimating the change probability of the house source according to a prediction model of the change probability of the house source of the hotel; the transfer probability of the hotel room source applying for transfer is determined to be V =1.
5. The method of claim 1, wherein the determining a transferee reputation value to evaluate related data and to protect privacy comprises:
firstly, determining relevant data of the credit value evaluation of the transferee, wherein the relevant data comprises past hotel reservation records of the transferee, personal information of the transferee and loan records; secondly, storing the credit value evaluation related data of the transferee and the constraint information in a database, enabling the related data in the database to be uniquely corresponding to the identity information of the transferee, and inputting the unique identity information to call the related data of the transferee; finally, encrypting the private information of the transferee by using a homomorphic encryption method, operating the data on the premise of not decrypting the encrypted private information, and obtaining the same result as the plaintext operation through a decryption algorithm according to the operation result; the method comprises the following steps: determining past predetermined data of the transferee; determining a lender credit record;
the determining of the past scheduled data of the transferee specifically includes:
firstly, determining past reservation data of a hotel room source, wherein the past reservation data comprises hotel reservations of a transferee and a successful check-in record, and a hotel reservation cancellation record and a hotel reservation non-cancellation and non-check-in record are determined; crawling OTA platform data to obtain past scheduled change records of a transferee; performing data preprocessing on the acquired user data, classifying the data and extracting reservation condition data, namely index = { the number of times of successful hotel stay in and the number of times of unsuccessful hotel stay in and hotel stay in }, wherein the index is the historical reservation condition of the transferee; finally, the acquired data is subjected to data cleaning, and missing data, repeated data and error data are deleted;
the determining of the lending credit record of the transferee specifically comprises the following steps:
firstly, determining credit record information of a transferee, including the age, sex, native place, cultural degree, income condition, work information, normal repayment times and overdue times of the borrowed amount of the network lending platform; acquiring a credit record of a transferee through a data warehouse of a bank credit card center; interpolating missing values of different degrees existing in the original data acquired by the data warehouse; determining missing values of credit records, including missing values of debit and credit population characteristic information and missing values of credit records of a borrower; the method comprises the following steps that a K neighbor classification algorithm is adopted to interpolate a missing value corresponding to the population characteristic information of a lender, and the missing value corresponding to a credit record of the lender is assigned to be 0; and the privacy information of the transferee is subjected to encryption and decryption algorithm processing, so that the information security is ensured.
6. The method of claim 1, wherein the analyzing a transferee reputation based on the transferee reputation value rating related data comprises:
analyzing the credit degree of the transferee based on the past hotel reservation data and the loan credit record of the transferee; classifying and summarizing the data, and matching the corresponding past hotel reservation data with loan credit record data, namely index = { age, gender, cultural degree, income condition, work information, normal repayment times and overdue times of the loan amount of a network loan platform, times of successful check-in and reservation of the hotel, times of unsuccessful check-in and reservation of the hotel }, wherein the index is a credit value evaluation related index of the transferee; preprocessing all variables of hotel reservation data and loan credit records of the transferee by using a tail shortening method; normalizing each variable of the hotel reservation data and the loan credit record of the transferee, and unifying dimensions; the method comprises the following steps: establishing a credit rating evaluation model; acquiring a corresponding credit value according to the data of the transferee;
the establishing of the credibility evaluation model specifically comprises the following steps:
constructing a credit value evaluation model by using a BP neural network based on the personal information and credit records of the transferee; calculating the weight occupied by each variable of the personal information and credit record of the transferee on the contribution degree of the personal credit degree value by using an entropy weight method; adjusting training parameters of the BP neural network model based on the calculated personal information of the transferee and the weight of each variable of the credit record; determining a training sample and a testing sample of the neural network according to the reputation value evaluation related index data; carrying out artificial neural network training on the sample by adopting a trainlm function; the credit value evaluation model is tested through the test set, parameters are adjusted, and the model is trained continuously, so that the accuracy of the model is improved;
the obtaining of the corresponding reputation value according to the transferee data specifically includes:
firstly, evaluating a credit degree value of a transferee based on a credit degree evaluation model with high accuracy obtained by neural network training; acquiring collated and summarized transferee data, and inputting the collated and summarized transferee data into a credibility evaluation model for processing; determining data of a transferee, wherein the data comprises past hotel reservation records of the transferee, personal identity information of the transferee and bank loan records; calculating to obtain a credit degree value R of the transferee based on model prediction, and obtaining a corresponding credit degree evaluation result according to a credit degree evaluation standard; the error of the model calculation result is reduced by multiple operations; presenting the reputation degree evaluation result to the transferee; and the transferee applies for re-evaluation on the evaluation result of the personal credibility with objection.
7. The method of claim 1, wherein determining a preliminary assignment price based on the premises variability probability and the transferee reputation value comprises:
determining a predetermined transfer house source price finally obtained by the transferee based on the house source change probability and the credit value of the transferee; constructing a transfer price prediction model: TP = IP-IP V1-IP (R-70)/100N 2); TP is transfer price; IP is the initial price of the hotel room source; v is the predicted hotel room source change probability; r is a credit degree numerical value of the transferee obtained by model calculation; n1 and N2 are different weights; IP V N1 is a deduction value of the change probability of the hotel room source to the transfer price; IP (R-70)/100 n2 is the credit value of the transferee's increase/decrease value for the house source price; and finally, primarily determining the transfer price according to the transfer price prediction model.
8. The method of claim 1, wherein adjusting the final transfer price based on the referral level comprises:
firstly, determining recommendation degree related parameters of a transferor, wherein the recommendation degree related parameters comprise refreshing click frequency of the transferor, a consultation message reply average time interval, a time interval of a specified check-in time and a time interval exceeding the specified check-in time; secondly, adjusting the assignment price prediction model based on the recommendation degree related parameters; constructing a final transfer price prediction model, namely: FP = TP + CF/30 + M1+ TP/RT + M2+ TP + IT/24 + M3-TP + OT/24 + M4; FP is the transfer price received by the final transferee, and TP is the transfer price primarily calculated by the model; CF is the refreshing click frequency of the transferor to the house resources to be transferred, and the unit is times/hour; RT is the average reply time interval of the transferor to the transferee consultation information, and the unit is minutes; IT is a time interval of the house source specified check-in time, and the unit is hour; OT is the time interval specifying the check-in time in hours; TP CF/30 M1 is the increment value of the transfer price of the refresh click frequency of the transferor; TP/RT M2 is the increment value of the price of the transfer room for the average reply time interval of the consultation messages; TP (IT/24) M3 is an increment value of the transfer price in the time interval from the specified check-in time; TP (OT/24) M4 is a reduction in transfer price for a time interval exceeding a specified check-in time; if the time interval OT exceeding the specified check-in time is more than 6 hours, the house resource transfer information is invalid; and finally, determining the transfer price acquired by the final transferee according to the final transfer price prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210859691.5A CN115271137A (en) | 2022-07-21 | 2022-07-21 | Online booking change processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210859691.5A CN115271137A (en) | 2022-07-21 | 2022-07-21 | Online booking change processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115271137A true CN115271137A (en) | 2022-11-01 |
Family
ID=83767932
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210859691.5A Pending CN115271137A (en) | 2022-07-21 | 2022-07-21 | Online booking change processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115271137A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1746920A (en) * | 2004-09-10 | 2006-03-15 | 鹏元资信评估有限公司 | Personnel credit system |
| US20160335563A1 (en) * | 2015-05-14 | 2016-11-17 | Xchange Hotels, Inc. | Method and apparatus for reselling room reservations |
| CN106934553A (en) * | 2017-03-17 | 2017-07-07 | 重庆市个人信用管理有限责任公司 | A kind of farmer credit evaluation method and device |
| CN109685547A (en) * | 2018-11-29 | 2019-04-26 | 杭州中房信息科技有限公司 | A kind of home price automatic evaluation method based on big data |
| CN110414751A (en) * | 2018-04-26 | 2019-11-05 | 观相科技（上海）有限公司 | A kind of hotel industry addressing evaluation system and evaluation method based on geographical location |
| CN112634018A (en) * | 2020-12-23 | 2021-04-09 | 百维金科(上海)信息科技有限公司 | Overdue monitoring method for optimizing recurrent neural network based on ant colony algorithm |
| CN113343309A (en) * | 2021-08-02 | 2021-09-03 | 北京东方通软件有限公司 | Natural person database privacy security protection method and device and terminal equipment |
-
2022
- 2022-07-21 CN CN202210859691.5A patent/CN115271137A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1746920A (en) * | 2004-09-10 | 2006-03-15 | 鹏元资信评估有限公司 | Personnel credit system |
| US20160335563A1 (en) * | 2015-05-14 | 2016-11-17 | Xchange Hotels, Inc. | Method and apparatus for reselling room reservations |
| CN106934553A (en) * | 2017-03-17 | 2017-07-07 | 重庆市个人信用管理有限责任公司 | A kind of farmer credit evaluation method and device |
| CN110414751A (en) * | 2018-04-26 | 2019-11-05 | 观相科技（上海）有限公司 | A kind of hotel industry addressing evaluation system and evaluation method based on geographical location |
| CN109685547A (en) * | 2018-11-29 | 2019-04-26 | 杭州中房信息科技有限公司 | A kind of home price automatic evaluation method based on big data |
| CN112634018A (en) * | 2020-12-23 | 2021-04-09 | 百维金科(上海)信息科技有限公司 | Overdue monitoring method for optimizing recurrent neural network based on ant colony algorithm |
| CN113343309A (en) * | 2021-08-02 | 2021-09-03 | 北京东方通软件有限公司 | Natural person database privacy security protection method and device and terminal equipment |
Non-Patent Citations (1)
| Title |
|---|
| "《摩拜单车更新信用分*** 对信用差者半小时收费100元》", pages 1, Retrieved from the Internet <URL:http://finance.people.com.cn/n1/2018/0224/c1004-29832019.html> * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Atkin et al. | The returns to face-to-face interactions: Knowledge spillovers in Silicon Valley | |
| Cai et al. | Price determinants of Airbnb listings: evidence from Hong Kong | |
| Antecol et al. | Labor supply differences between married heterosexual women and partnered lesbians: A semi‐parametric decomposition approach | |
| Galster et al. | Residential satisfaction: Compositional and contextual correlates | |
| Aiken et al. | Machine learning and mobile phone data can improve the targeting of humanitarian assistance | |
| Falk et al. | Modelling the cancellation behaviour of hotel guests | |
| US20180308188A1 (en) | Computerized system and method for real estate searches and procurement | |
| Osei Asibey et al. | Analysing the influence of health insurance status on peoples’ health seeking behaviour in rural Ghana | |
| US10559150B1 (en) | Secured electronic communication system | |
| CN112508075B (en) | DBSCAN clustering method based on transverse federation and related equipment thereof | |
| Baumle et al. | Legislating the family: The effect of state family laws on the presence of children in same‐sex households | |
| Ding et al. | Estimating socioeconomic status via temporal-spatial mobility analysis-a case study of smart card data | |
| Albaladejo et al. | The effects of motivations to go to the country on rural accommodation choice: A hybrid discrete choice model | |
| CN116681220A (en) | Smart city civil management method and system based on Internet of things | |
| Lubua Dr | The influence of socioeconomic factors to the use of mobile phones in the agricultural sector of Tanzania | |
| CN116805039B (en) | Feature screening method, device, computer equipment and data disturbance method | |
| Young | Toward an understanding of wilderness participation | |
| CN115271137A (en) | Online booking change processing method | |
| Jin et al. | Research on the evaluation model of rural information demand based on big data | |
| CN114882974A (en) | Psychological diagnosis database access artificial intelligence verification system and method | |
| CN112685654B (en) | Student identification method and device, computing equipment and readable computer storage medium | |
| Gath et al. | The potential for linked administrative data to provide household and family information | |
| US20200134514A1 (en) | Booked-blocked classifier | |
| US20130317887A1 (en) | Dynamic campaign supporter system for mobilizing campaign supporters | |
| KR102379625B1 (en) | System for providing sharing economy based riding sharing service |
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 |