A method and device for verifying consistency of travel expense itinerary and automatically calculating subsidies, an electronic device and a storage medium

Abstract

This invention provides a method, apparatus, electronic device, and storage medium for verifying the consistency of travel expense itineraries and automatically calculating subsidies, relating to the technical field of computer science. This method constructs a full-link travel spatiotemporal trajectory chain by collecting multimodal travel data and performs cross-validation of spatiotemporal consistency across vouchers. This effectively overcomes the spatiotemporal logic blind spots caused by data fragmentation and coarse-grained rules in traditional technologies. Thus, while achieving accurate and automatic identification of travel violations, the method utilizes a time-slicing dynamic calculation mechanism to ensure the granularity and accuracy of subsidy calculation. Ultimately, this achieves a transformation of travel business from "manual, extensive review" to "automated, refined compliance management throughout the entire lifecycle," significantly improving review efficiency and reducing the risk of corporate fund loss.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, electronic device, and storage medium for verifying the consistency of travel expense itinerary trajectory and automatically calculating subsidies. Background Technology

[0002] Travel expenses are typically the second largest controllable cost for businesses (after salaries). In large companies, employees travel frequently and have complex itineraries (connecting multiple locations, changes mid-journey). Finance staff face immense pressure to verify information: Did the employee actually go to the city? Were the taxi receipts for the business trip? Did they receive excessive subsidies (e.g., the hotel included breakfast, but the employee received the full meal allowance; or they ate on the plane or in the taxi and received a meal allowance)? Currently, most companies still rely on employees manually filling out "business trip allowance application forms," ​​with finance personnel manually verifying flight tickets and invoice dates—a process that is inefficient and prone to errors.

[0003] The traditional methods of verifying and checking travel expenses have the following technical problems: Manual calculation and verification: Employees calculate the number of days and subsidies in Excel, and the finance department visually compares the invoices.

[0004] Problems: Extremely inefficient and difficult to detect hidden issues such as "reimbursing taxi fares in Shanghai while on a business trip to Beijing" (if the dates match). Unable to handle complex deduction logic (such as deductions including early departures).

[0005] Simple rule control: The OA system only controls the "start and end dates of the business trip".

[0006] Problem: The system approves applications as long as the invoice date is within the application period. It cannot identify specific time conflicts (e.g., a flight in the morning with a taxi receipt for the same morning). It cannot verify the continuity of the itinerary.

[0007] TMC platform single data: Enterprises use business travel platforms (such as Ctrip Business Travel).

[0008] Problem: TMC only has booking data, not actual expense reports (such as taxi receipts hailed on the street or hotel bookings not made through TMC). The data is fragmented, making it impossible to form a complete chain of evidence.

[0009] In summary, traditional travel verification technologies suffer from several problems: manual verification is inefficient and prone to errors; system rules can only perform coarse-grained date matching and cannot identify spatiotemporal logical conflicts; and incomplete data coverage of business travel platforms leads to a disconnect between online and offline consumption records. These issues make it difficult for companies to achieve refined management and automatic compliance verification of the entire lifecycle of their travel business. Summary of the Invention

[0010] In view of this, the purpose of this invention is to provide a method, device, electronic device and storage medium for verifying the consistency of travel expense itinerary trajectory and automatically calculating subsidies, so as to alleviate the technical problem that traditional travel audit technology makes it impossible for enterprises to achieve refined management and automatic compliance verification of the entire life cycle of travel business.

[0011] In a first aspect, the present invention provides a method for verifying the consistency of travel expense itineraries and automatically calculating subsidies, including: Collect multimodal travel data and generate standard trajectory point data based on the multimodal travel data; Construct a full-link business travel spatiotemporal trajectory chain based on the aforementioned standard trajectory point data; Perform cross-validation of cross-document spatiotemporal consistency based on the aforementioned end-to-end travel spatiotemporal trajectory chain; If the verification is successful, the travel subsidy amount will be dynamically calculated based on the preset travel policy rules and the full-link travel time-space trajectory chain, and the final total travel subsidy amount will be obtained by summarizing. A travel compliance report containing calculation details is generated based on the final total travel allowance and the travel allowance amount for each time period, and a payment instruction is generated based on the final total travel allowance.

[0012] Furthermore, standard trajectory point data is generated based on the multimodal travel data, including: By connecting to business travel platforms and ride-hailing platforms through application programming interfaces (APIs), structured order data is obtained, and the structured order data is marked as the benchmark fact data with the highest confidence level. Optical character recognition technology is used to identify the ticket images uploaded by users and extract the take-off and landing times and locations of transportation tickets, check-in and check-out information and room type characteristics of accommodation tickets, and time and location information of consumption invoices. With user authorization, collect the check-in coordinates of mobile office applications or parse the EXIF ​​geolocation information of uploaded image files as auxiliary data; A global city geographic database is established, and the location descriptions in the above multi-source data are uniformly mapped to standard city codes. All time data are uniformly converted into Coordinated Universal Time timestamps for storage, generating the standard trajectory point data containing timestamps, location coordinates, event types, and credential sources.

[0013] Furthermore, based on the aforementioned standard trajectory point data, a full-link travel spatiotemporal trajectory chain is constructed, including: Extract the traffic order data from the standard trajectory point data as key nodes, construct a time axis skeleton, and divide the time axis skeleton into multiple road segment intervals; The accommodation order data in the standard trajectory point data is extracted as the anchor point of the stay and superimposed on the time axis skeleton to determine the basic stay period; Using a spatiotemporal interpolation algorithm and based on preset state determination rules, the blank time periods on the time axis skeleton, excluding the key nodes and the basic dwell time periods, are filled with states to generate a continuous time series that includes the transit state, the destination state, and the activity state. Detect whether there is a time interval and missing traffic credentials between adjacent key nodes. If so, mark the corresponding time period as an unknown displacement state and associate it with a supplementary explanation request. Based on the key nodes, the basic dwell time period, the filled continuous time series, and the marked unknown displacement states, the full-link travel spatiotemporal trajectory chain is assembled and generated.

[0014] Furthermore, cross-document spatiotemporal consistency cross-validation is performed based on the aforementioned end-to-end travel spatiotemporal trajectory chain, including: Calculate the geographical distance between the location coordinates of the consumption invoice and the city to which the current trajectory segment belongs. If it exceeds the preset threshold of the same city radius, it is judged as an abnormal consumption in a different location. The system checks whether the transaction time of the consumption invoice falls within the on-the-go time period of the major transportation. If so, and the consumption type of the consumption invoice is not airport pick-up and drop-off service, it is determined to be an abnormal on-the-go consumption violation. The system checks whether there are two or more consumption invoices from different locations whose business transactions overlap in time. If so, it is determined to be an anomaly of spatiotemporal logic conflict. The distance between the start and end points of a continuous urban traffic pass is checked. If the distance between the previous end point and the next start point is too far and there is no traffic record in between, it is judged as an abnormal trajectory break.

[0015] Furthermore, based on the preset travel policy rules and the aforementioned end-to-end travel spatiotemporal trajectory chain, the travel allowance amount is dynamically calculated by time slice, including: The entire-link travel time-space trajectory chain is divided into multiple time slices according to natural days; For each time slice, a preset city classification matrix is ​​queried based on the city level of the trajectory within the slice to determine the benchmark subsidy amount for that day. Based on the departure or return time points within the time slice, determine whether it is the first or last day of the business trip, and apply the corresponding first and last day subsidy coefficients to correct the benchmark subsidy amount. If the hotel bill data associated with the test day includes a breakfast service identifier, the breakfast subsidy amount will be deducted from the revised subsidy amount. If there are any reimbursed business entertainment expense vouchers or meeting meal expense vouchers on the day of the inspection, the meal allowance for the corresponding period will be deducted from the remaining subsidy amount. If there is a company vehicle dispatch record on the day of the check, and if so, the urban transportation subsidy amount is deducted from the remaining subsidy amount to obtain the travel subsidy amount.

[0016] Furthermore, based on the city level of the trajectory within the slice, a preset city classification matrix is ​​queried to determine the benchmark subsidy amount for the day, including: If the same natural day involves crossing cities of different levels, the benchmark subsidy amount for that day is determined according to the preset cross-city rule configuration and the city classification matrix.

[0017] Furthermore, a travel compliance report containing calculation details is generated based on the final total travel allowance and the travel allowance amounts for each time period, and a payment instruction is generated based on the final total travel allowance, including: Generate a visual travel trajectory map, and use different colors to mark compliant travel segments and abnormal change points on the travel trajectory map; For documents that fail verification, distinguish between hard and soft defects; hard defects are directly removed and the reimbursement process is blocked; for soft defects, an explanation submission entry is generated and the calculation is retried after manual approval. The travel compliance report is generated based on the travel trajectory map, the hard injury anomaly, the soft injury anomaly, the daily benchmark subsidy amount, the subsidy coefficient, the deduction details, and the travel subsidy amount. The payment instruction is then generated based on the final total travel subsidy amount.

[0018] Secondly, the present invention also provides a travel expense itinerary consistency verification and automatic subsidy calculation device, comprising: The acquisition and generation unit is used to acquire multimodal travel data and generate standard trajectory point data based on the multimodal travel data; The construction unit is used to construct a full-link travel spatiotemporal trajectory chain based on the standard trajectory point data; The verification unit is used to perform cross-document spatiotemporal consistency verification based on the end-to-end travel spatiotemporal trajectory chain; The calculation and aggregation unit is used to dynamically calculate the travel subsidy amount by time slice based on the preset travel system rules and the full-link travel spatiotemporal trajectory chain if the verification is successful, and to aggregate the final total travel subsidy amount. The generation unit is used to generate a travel compliance report containing calculation details based on the final total travel allowance and the travel allowance amount for each time period, and to generate a payment instruction based on the final total travel allowance.

[0019] Thirdly, the present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method described in the first aspect.

[0020] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the method described in the first aspect.

[0021] This invention provides a method for verifying the consistency of travel expense itineraries and automatically calculating subsidies, comprising: collecting multimodal travel data and generating standard trajectory point data based on the multimodal travel data; constructing a full-link travel spatiotemporal trajectory chain based on the standard trajectory point data; performing cross-validation of spatiotemporal consistency across vouchers based on the full-link travel spatiotemporal trajectory chain; if the validation passes, dynamically calculating the travel subsidy amount by time slice based on preset travel policy rules and the full-link travel spatiotemporal trajectory chain, and summarizing to obtain the final total travel subsidy amount; generating a travel compliance report containing calculation details based on the final total travel subsidy amount and the travel subsidy amount for each time period, and generating a payment instruction based on the final total travel subsidy amount. As described above, the travel expense itinerary consistency verification and automatic subsidy calculation method of the present invention effectively overcomes the spatial-temporal logic blind spots caused by data fragmentation and coarse-grained rules in traditional technologies by collecting multimodal travel data to construct a full-link travel spatiotemporal trajectory chain and performing cross-validation of spatiotemporal consistency across vouchers. This achieves accurate and automatic identification of travel violations while ensuring the granularity and accuracy of subsidy calculation through a time-slicing dynamic calculation mechanism. Ultimately, it achieves the transformation of travel business from "manual and extensive review" to "automated and refined compliance management throughout the entire lifecycle," significantly improving review efficiency and reducing the risk of corporate fund loss. It also alleviates the technical problem that traditional travel review technology makes it impossible for enterprises to achieve refined management and automatic compliance verification of the entire lifecycle of travel business. Attached Figure Description

[0022] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating a method for verifying the consistency of travel expense itineraries and automatically calculating subsidies, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a travel expense itinerary consistency verification and automatic subsidy calculation device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0024] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Traditional travel auditing technologies cannot enable companies to conduct refined management and automatic compliance verification of the entire lifecycle of their travel business.

[0026] Based on this, the travel expense itinerary consistency verification and automatic subsidy calculation method of the present invention constructs a full-link travel spatiotemporal trajectory chain by collecting multimodal travel data and performs cross-validation of spatiotemporal consistency across vouchers. This effectively overcomes the spatiotemporal logic blind spots caused by data fragmentation and coarse-grained rules in traditional technologies. Thus, while achieving accurate and automatic identification of travel violations, the dynamic calculation mechanism using time slicing ensures the granularity and accuracy of subsidy calculation. Ultimately, this achieves the transformation of travel business from "manual and extensive review" to "automated and refined compliance management throughout the entire lifecycle," significantly improving review efficiency and reducing the risk of corporate fund loss.

[0027] To facilitate understanding of this embodiment, a detailed description of the method for verifying the consistency of travel expense itineraries and calculating automatic subsidies disclosed in this embodiment of the invention will be provided first.

[0028] Example 1: According to an embodiment of the present invention, an embodiment of a method for verifying the consistency of travel expense itinerary trajectory and automatically calculating subsidies is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] Figure 1 This is a flowchart of a method for verifying the consistency of travel expense itineraries and automatically calculating subsidies according to an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes the following steps: Step S102: Collect multimodal travel data and generate standard trajectory point data based on the multimodal travel data; In this step, the system first obtains raw travel-related data from multiple heterogeneous data sources through pre-built data interfaces or file parsing modules. Multimodal travel data covers different dimensions of information reflecting employee business travel behavior, including but not limited to transportation ticketing data (such as airfare and train ticket orders), accommodation registration data (such as hotel check-in / check-out records), ride-hailing data, restaurant spending records, meeting attendance records, and location-based services (LBS) attendance data. Due to differences in data formats, time precision, and location description methods from different sources (e.g., airfares use airport codes, hotels use detailed addresses, and ride-hailing apps use latitude and longitude), the system needs to clean, normalize, and semantically map the raw data. Specific implementation methods include using Natural Language Processing (NLP) technology to extract address entities, calling the Geocoding API to convert text addresses into unified latitude and longitude coordinates, and uniformly calibrating all timestamps to the standard time zone. After processing, each individual consumption or behavior record is transformed into standard trajectory point data containing "unique identifier, timestamp, latitude and longitude coordinates, behavior type, and associated voucher number" (i.e., including timestamp, location coordinates, event type, and voucher source).

[0030] This step, through the collection and standardization of multimodal travel data, breaks down data silos between online and offline platforms and different platforms, laying a unified data foundation for the subsequent construction of continuous spatiotemporal trajectories (i.e., the full-link travel spatiotemporal trajectory chain), and ensuring the comprehensiveness of data sources and the consistency of formats.

[0031] Step S104: Construct a full-link travel spatiotemporal trajectory chain based on standard trajectory point data; In this step, the system uses an employee's travel application number as an index to concatenate all standard trajectory point data generated during that trip in chronological order. The construction process considers not only linear time sequences but also spatial topological relationship verification to ensure that adjacent trajectory points are physically feasible. The system calculates the time difference and spatial distance between two adjacent trajectory points, and combines this with the average speed threshold of conventional transportation to determine if there is logical connectivity between the two points. If the time interval between two points is insufficient to complete the spatial displacement, it is marked as an abnormal breakpoint; if the connection is smooth, a complete end-to-end travel spatiotemporal trajectory chain with both time and spatial dimensions is formed. This trajectory chain visually reconstructs the employee's complete movement path from departure point to destination and return trip.

[0032] This step constructs a full-link spatiotemporal trajectory (i.e., a full-link business travel spatiotemporal trajectory chain), which transforms previously isolated consumption records into a logically related chain of evidence, enabling a complete reproduction of employees' business travel routes and providing a global perspective for identifying hidden travel conflicts.

[0033] Step S106: Perform cross-validation of cross-validation of spatiotemporal consistency across vouchers based on the end-to-end travel spatiotemporal trajectory chain; In this step, the system utilizes the constructed end-to-end travel spatiotemporal trajectory chain to perform in-depth logical verification of various expense vouchers attached to the chain. The core of the verification lies in "spatiotemporal exclusivity" and "voucher matching degree." The system checks whether employees appear in two locations that are too far apart physically and cannot be reached by normal transportation within the same time period (i.e., spatiotemporal conflict); at the same time, it verifies whether the time and location recorded on the voucher are consistent with the corresponding nodes on the trajectory chain. For example, it checks whether the issuance time of the restaurant invoice is within the employee's local stay time period, and whether the taxi origin coincides with the destination of the previous trip. If it is found that the voucher time does not overlap with the trajectory time, or the voucher location deviates from the trajectory node by more than a preset threshold (e.g., 5 kilometers), the verification is judged as failing. That is, it checks whether the location coordinates of the expense invoice data fall within the geofence corresponding to the trajectory chain of the current time period, and checks whether the timestamp of the expense invoice data has a logical conflict with the state of the trajectory chain of the current time period.

[0034] This step enables automated logical auditing across vouchers, accurately identifying "overlapping" violations (such as duplicate reimbursements and fraudulent trips) that traditional rules cannot detect, significantly improving the accuracy of compliance verification.

[0035] Step S108: If the verification is successful, the travel subsidy amount is dynamically calculated by time slice based on the preset travel policy rules and the full-link travel time-space trajectory chain, and the final total travel subsidy amount is obtained by summarizing. In this step, when the cross-validation of cross-validation for spatiotemporal consistency across vouchers passes, the system activates the subsidy calculation engine. The engine first loads the company's pre-set travel policy rule library (containing subsidy standards for different city levels, subsidy coefficients for different time periods, and subsidy rules for special scenarios). Then, the system performs high-precision "time slicing" on the entire travel spatiotemporal trajectory chain along the time axis, dividing the continuous trip into discrete time periods in hours or half-hours. For each time slice, the system dynamically calculates the subsidy based on the geographical location (determining the city level) and behavioral status (e.g., en route, in accommodation, in a meeting) of the corresponding trajectory point. Finally, the calculation results for all time slices are summed to obtain the final total travel subsidy.

[0036] This step uses time-slice dynamic calculation to achieve refined subsidy accounting, avoiding the unfairness caused by traditional daily calculations such as "half a day of business trip is counted as a day" or "long-haul flights are not fully subsidized", ensuring that every subsidy is based on evidence.

[0037] Step S110: Generate a travel compliance report containing calculation details based on the final total travel allowance and the travel allowance amount for each time period, and generate a payment instruction based on the final total travel allowance.

[0038] In this step, the system automatically generates a structured travel compliance report based on the calculation results of step S108. The report not only includes the final total travel allowance but also details the calculation basis for each time slice, the corresponding trajectory nodes, the matching regulations and clauses, and a list of verified vouchers. For cases that fail verification, the report clearly marks the anomalies and their reasons. The report supports visual display for financial personnel to review or for direct archiving. Simultaneously, if the report status is confirmed as "compliant," the system will automatically generate a payment instruction containing payee information, amount, and summary through the interface with the enterprise's financial system or the bank-enterprise direct connection system, and push it to the payment gateway to execute the payment operation.

[0039] This step automates the entire process from review to payment, and the resulting detailed reports enhance financial transparency, facilitate audit traceability, significantly shorten the reimbursement cycle, and improve employee satisfaction.

[0040] This invention provides a method for verifying the consistency of travel expense itineraries and automatically calculating subsidies, comprising: collecting multimodal travel data and generating standard trajectory point data based on the multimodal travel data; constructing a full-link travel spatiotemporal trajectory chain based on the standard trajectory point data; performing cross-validation of spatiotemporal consistency across vouchers based on the full-link travel spatiotemporal trajectory chain; if the validation passes, dynamically calculating the travel subsidy amount by time slice based on preset travel policy rules and the full-link travel spatiotemporal trajectory chain, and summarizing to obtain the final total travel subsidy amount; generating a travel compliance report containing calculation details based on the final total travel subsidy amount and the travel subsidy amount for each time period, and generating a payment instruction based on the final total travel subsidy amount. As described above, the travel expense itinerary consistency verification and automatic subsidy calculation method of the present invention effectively overcomes the spatial-temporal logic blind spots caused by data fragmentation and coarse-grained rules in traditional technologies by collecting multimodal travel data to construct a full-link travel spatiotemporal trajectory chain and performing cross-validation of spatiotemporal consistency across vouchers. This achieves accurate and automatic identification of travel violations while ensuring the granularity and accuracy of subsidy calculation through a time-slicing dynamic calculation mechanism. Ultimately, it achieves the transformation of travel business from "manual and extensive review" to "automated and refined compliance management throughout the entire lifecycle," significantly improving review efficiency and reducing the risk of corporate fund loss. It also alleviates the technical problem that traditional travel review technology makes it impossible for enterprises to achieve refined management and automatic compliance verification of the entire lifecycle of travel business.

[0041] The above provides a brief overview of the travel expense itinerary consistency verification and automatic subsidy calculation method of the present invention. The specific details involved are described in detail below.

[0042] In an optional embodiment of the present invention, generating standard trajectory point data based on multimodal travel data specifically includes the following steps: (1) Connect to business travel platforms and ride-hailing platforms through application programming interfaces to obtain structured order data, and mark the structured order data as the benchmark fact data with the highest confidence level; Specifically, the system connects to TMC platforms such as business travel agencies (TMCs) and ride-hailing platforms through API interfaces to obtain tamper-proof, structured order data, and marks the structured order data as the benchmark fact data with the highest confidence level.

[0043] (2) Use optical character recognition technology to identify the ticket images uploaded by users and extract the take-off and landing time and location of the transportation ticket, the check-in and check-out information and room type characteristics of the accommodation ticket, and the time and location information of the consumption invoice; Specifically, it uses an OCR engine to recognize PDFs or printed photos. PDFs or printed photos can include: For transportation documents (flight itineraries and train tickets uploaded by employees): extract departure / destination, takeoff / landing time (accurate to the minute), and flight / train number. Pay special attention to identifying local time conversions for flights crossing time zones.

[0044] Accommodation receipt: Check-in / check-out date, hotel address (analyzed as city + administrative district), room type (including whether breakfast is included), and guest name.

[0045] For consumption invoices (hotel bills, taxi receipts, restaurant receipts): retrieve the pick-up / drop-off time for taxi receipts and the invoice date for restaurant receipts (note that the invoice date is often later than the consumption time, and you need to combine the receipt or remarks to help determine the actual time of the transaction).

[0046] (3) With the user’s authorization, collect the check-in coordinates of the mobile office application or parse the EXIF ​​geolocation information of the uploaded image file as auxiliary data; Specifically, LBS auxiliary data (optional): If the company has a mobile attendance application, collect the clock-in coordinates of employees during business trips; or parse the EXIF ​​geolocation information of uploaded photos.

[0047] (4) Establish a global city geographic database, map the location descriptions in the above multi-source data into standard city codes, and convert all time data into Coordinated Universal Time timestamps for storage, generating standard trajectory point data containing timestamps, location coordinates, event types and credential sources.

[0048] Specifically, a global urban geographic database (GeoDB) is established. "PVG", "Shanghai Pudong", and "Hu" are uniformly mapped to the standard city code CN_SH_021. All times are uniformly converted to UTC timestamps for storage, but converted to the "local time of the business trip location" when calculating subsidies. Finally, the above heterogeneous data is cleaned and uniformly mapped to standard trajectory point data containing {timestamp, location coordinates (longitude / latitude / admin area), event type, voucher source}.

[0049] In an optional embodiment of the present invention, a full-link business trip spatio-temporal trajectory chain is constructed based on the standard trajectory point data, which specifically includes the following steps: (1) Extract the large transportation order data in the standard trajectory point data as key nodes, construct a time-axis skeleton, and divide the time-axis skeleton into multiple section intervals; Specifically, constructing the full-link business trip spatio-temporal trajectory chain is the core model of this system. Traditional reimbursement only looks at isolated documents, while this system restores the "process".

[0050] Taking the date approved by the business trip application form as the range and the large transportation (air ticket / train ticket) as the key node, the time axis is divided into several Segments (section intervals, such as: Beijing → Shanghai section, Shanghai stay section, Shanghai → Beijing section).

[0051] (2) Extract the accommodation order data in the standard trajectory point data as the stay anchor points, and superimpose them on the time-axis skeleton to determine the basic stay period; Specifically, the "anchor points" are established based on the accommodation records.

[0052] (3) Using the spatio-temporal interpolation algorithm and based on the preset state determination rules, fill the blank time periods on the time-axis skeleton except for the key nodes and the basic stay period with states, generating a continuous time series including the on-the-way state, the in-hotel state, and the activity state; Specifically, the "details" are filled with intracity transportation and consumption invoices. The system automatically infers the states of the blank time periods (such as: on the way, in the hotel, at work), forming a logically closed-loop User_Travel_Timeline.

[0053] When filling the states, the on-the-way state is from 2 hours before the take-off time to 1 hour after the landing time and is considered on the way. Usually, no catering or intracity taxi fees should be generated during this period (unless it is for pick-up and drop-off at the airport); the in-hotel state is from 22:00 every night to 07:00 the next day from the hotel check-in date to the check-out date; the activity state is the period other than the above times.

[0054] (4) Detect whether there is a time interval and a lack of transportation vouchers between adjacent key nodes. If so, mark the corresponding time period as the unknown displacement state and associate a supplementary explanation request; Specifically, if the transportation system shows that route 1 is in Shanghai and route 3 is in Beijing, but the intermediate transportation tickets are missing, the system will mark a "unknown displacement" trajectory and require the employee to provide supplementary explanations (such as hitchhiking or losing tickets). Otherwise, the subsidy calculation for this period will be suspended.

[0055] (5) Based on key nodes, basic dwell time periods, filled continuous time series and marked unknown displacement states, assemble and generate a full-link travel spatiotemporal trajectory chain.

[0056] In an optional embodiment of the present invention, cross-document spatiotemporal consistency cross-validation is performed based on the end-to-end travel spatiotemporal trajectory chain, specifically including the following steps: (1) Calculate the geographical distance between the location coordinates of the consumption invoice and the city to which the current trajectory segment belongs. If it exceeds the preset same-city radius threshold, it is determined to be an abnormal consumption in another location. Specifically, spatial conflict detection is performed by calculating the geographical inclusion relationship between the location of the consumption invoice and the city of the current trajectory segment.

[0057] For example, an employee's location tracking shows Segment: Shanghai, but they submitted a taxi receipt for City: Suzhou. The system calls the map API to calculate the distance between the two locations. If the distance exceeds a threshold (e.g., across cities), an alert is triggered, requiring proof of cross-city travel.

[0058] In other words, ticket-to-ticket consistency verification: that is, verifying whether there are conflicting consumptions within the same time period (e.g., a restaurant invoice in Shanghai at 12:00 and a taxi invoice in Beijing at 12:30).

[0059] (2) Check whether the time of the transaction of the consumption invoice falls within the time period of the transit status of the major transportation. If so, and the consumption type of the consumption invoice is not airport pick-up and drop-off service, it is judged as an abnormal consumption violation during transit. Specifically, time conflict detection (including logical conflicts) is performed, specifically checking if the invoice date is displayed during the flight. It's important to distinguish between "invoice date" and "consumption date." For e-invoices, the system prioritizes the "transaction time"; if none is available, it relies on invoice remarks or associated itinerary confirmation.

[0060] In other words, ticket-travel consistency verification: that is, verifying whether the pick-up time on the taxi receipt is in the city of business trip and not during flight / high-speed rail travel.

[0061] (3) Check whether there are two or more consumption invoices from different locations whose business time periods overlap. If so, it is determined to be an abnormal spatiotemporal logic conflict. Specifically, time conflict detection (including overlapping conflicts) is performed, which means that the time periods of two taxi tickets overlap (e.g., ticket A is from 10:00 to 10:30 and ticket B is from 10:15 to 10:45, which is judged as a risk of ticket substitution or group booking).

[0062] A hotel stay-travel consistency check is also required: that is, to verify whether the hotel check-in / check-out dates completely cover the nights during the business trip (unless it is noted in the application form that nighttime transportation or staying at relatives' or friends' homes is included).

[0063] (4) Verify the distance between the start and end points of the continuous urban traffic pass. If the distance between the previous end point and the next start point is too far and there is no traffic record in between, it is determined to be an abnormal trajectory break.

[0064] Specifically, a continuity check is performed, which verifies whether the start and end points of urban traffic are continuous. For example, if the previous taxi ticket's destination is location A, and the next ticket's starting point is location B, and the distance between A and B is extremely far with no traffic records in between, this is judged as an abnormal trajectory break.

[0065] In an optional embodiment of the present invention, the travel allowance amount is dynamically calculated based on preset travel policy rules and the entire travel time-space trajectory chain, specifically including the following steps: (1) Divide the entire travel time-space trajectory chain into multiple time slices according to natural days; Specifically, the entire travel time and space trajectory chain is sliced ​​according to "natural day (00:00-24:00)" or "24-hour cycle" to obtain multiple time slices.

[0066] (2) For each time slice, query the preset city classification matrix according to the city level of the trajectory in the slice to determine the benchmark subsidy amount for the day; Specifically, the system has a built-in flexible rules engine (Drools) that digitizes corporate travel policies. This includes: city tier classification (first-tier / second-tier / third-tier), subsidy benchmark (e.g., 300 RMB / day for first-tier cities), time period rules (e.g., departure time before 12:00 is counted as a full day, departure time after 12:00 is counted as a half day), and deduction rules (e.g., breakfast allowance is deducted for hotel accommodations that include breakfast, and meal allowance is deducted for conferences that include meals).

[0067] In practice, different companies have vastly different travel policies, so the system uses parameterized configuration. Specifically, this includes: City Tier Matrix: Maintain a City_Tier_Map. For example: {Beijing: Tier 1, Hangzhou: Tier 2, County-level cities: Tier 3}. Different tiers correspond to different Allowance_Base (base amount, i.e., benchmark subsidy amount).

[0068] Segmented calculation logic: First and last day rules: If the departure time is <12:00, the subsidy coefficient is 1.0; if it is >12:00, the subsidy coefficient is 0.5. The same applies to the return day.

[0069] Cross-city rules: If crossing cities of different tiers within a day (e.g., in a second-tier city in the morning and in a first-tier city in the afternoon), the rules can be configured as "highest tier principle", "lowest tier principle" or "weighted by length of stay".

[0070] Mutually exclusive deduction rules: Rule_Breakfast:IF Hotel_Receipt contains "Breakfast" OR "Bed&Breakfast" THEN deduct 20% of Daily_Allowance. Rule_Banquet:IF Expense_List contains "Business Meal" THEN deduct 40%(Lunch / Dinner) for that day. Rule_Conference:IF Trip_ Type=="Conference" AND Conference ProvideMeal==True THEN allowance=0. When determining the base subsidy amount for a given day, the system queries the city tier for each day to obtain the base subsidy amount (e.g., 300 yuan). If the system crosses different city tiers within the same calendar day, the base subsidy amount for that day is determined based on the preset cross-city rule configuration and the city tier matrix. The system can also automatically identify the "en route" and "stationed" statuses, applying different subsidy standards accordingly.

[0071] (3) Based on the departure or return time point within the time slice, determine whether it is the first or last day of the business trip, and apply the corresponding first and last day subsidy coefficient to correct the benchmark subsidy amount; Specifically, check if there are any major transportation trips on the day, applying the first and last day coefficient. Check if the day is a weekend / holiday (some companies offer extra allowances for business trips on holidays).

[0072] (4) Check the hotel bill data associated with the day. If it contains the breakfast service identifier, deduct the breakfast subsidy amount from the corrected subsidy amount. (5) Check whether there are any reimbursed business entertainment expense vouchers or meeting meal expense vouchers on the same day. If so, deduct the meal allowance for the corresponding period from the remaining subsidy amount. (6) Check whether there is a company dispatch record on the day of the test. If there is, deduct the urban transportation subsidy amount from the remaining subsidy amount to obtain the travel subsidy amount.

[0073] Specifically, scan the catering invoices and expense records for the day. If there are "business entertainment expenses," the corresponding meal allowance will be automatically deducted.

[0074] Scan the hotel bill for the day. If the OCR identifies "breakfast included", deduct the breakfast allowance.

[0075] Scan the administrative vehicle dispatch record. If the company provides a vehicle for pick-up and drop-off, deduct the local transportation allowance.

[0076] Summary output: Total_Subsidy = ∑(Day_i_Base) The system uses a time factor (Time_Factor-Deductions) formula, where Day_i_Base represents the base subsidy amount for day i, Time_Factor represents the subsidy coefficient, and Deductions represents the deductions from the subsidy. The system supports generating calculation drafts to show employees "why they only received 120 yuan instead of 300 yuan that day," reducing communication costs.

[0077] In an optional embodiment of the present invention, a travel compliance report containing calculation details is generated based on the final total travel allowance and the travel allowance amount for each time period, and a payment instruction is generated based on the final total travel allowance, specifically including the following steps: (1) Generate a visual travel trajectory map and mark compliant travel segments and abnormal change points on the travel trajectory map with different colors; (2) For documents that fail verification, distinguish between hard and soft defects; for hard defects, remove them directly and block the reimbursement process; for soft defects, generate an explanation submission entry and re-trigger the calculation after manual approval. (3) Generate a travel compliance report based on the travel trajectory map, hard injury anomalies, soft injury anomalies, daily benchmark subsidy amount, subsidy coefficient, deduction details and travel subsidy amount, and generate a payment instruction based on the final total travel subsidy amount.

[0078] Specifically, the system outputs a travel compliance and subsidy calculation report, displaying detailed calculation formulas. For documents that fail consistency checks (such as ride-hailing in other cities or conflicting travel times), they are automatically marked as "abnormal" and reimbursement is blocked; for the calculated subsidy amount, a payment instruction is directly generated without the need for manual recalculation.

[0079] In practice, a visual trajectory graph is generated on the review interface, with green representing compliant trips and red representing abnormal transition points.

[0080] Abnormal blocking: For "hard errors" (such as invoices from other locations), the system will directly reject the invoice and refuse reimbursement. For "soft errors" (such as broken tracking), the system allows employees to submit explanations, which must be approved by a human supervisor before the system can recalculate.

[0081] Data Accumulation: Accumulate real business travel data into a "business travel map," analyze frequently visited business destinations and hotels to provide data support for companies to negotiate negotiated rates with hotels; analyze employees' actual taxi routes to optimize business travel transportation suggestions.

[0082] This invention proposes a system for verifying the consistency of travel expense itineraries and automatically calculating subsidies. Addressing the pain points of corporate travel reimbursement, such as "falsified itineraries, cumbersome subsidy calculations, and spatiotemporal logical conflicts in invoices," this system aggregates data from major transportation modes (air / rail), accommodation, local transportation (taxi), and optional LBS location data to construct a complete "spatiotemporal trajectory chain" for employee travel. Combined with the company's complex travel policy engine, it achieves a closed-loop process from verifying the authenticity of travel itineraries to automatically calculating subsidies (allowances).

[0083] The core invention points are as follows: 1. A trajectory reconstruction algorithm based on spatio-temporal anchors: Existing expense reimbursement systems are discrete. This invention addresses this by "connecting points into lines." It innovatively proposes using immutable major transportation (airfare / train tickets) as strong anchors and accommodation as weak anchors to construct a continuous spatio-temporal model. This algorithm can automatically handle complex scenarios such as time zone changes, overnight flights, and intercity transfers, filling the gap in traditional systems that only provide "documents" without "itineraries." In other words, it's a full-link travel spatio-temporal trajectory reconstruction technology: integrating multimodal data such as major transportation, accommodation, expense invoices, and LBS positioning, and using time interpolation and spatial mapping algorithms to reconstruct discrete documents into a continuous, visualized employee travel spatio-temporal trajectory map.

[0084] 2. Decoupling Verification Technology for Invoice Issuance Time and Transaction Time: Addressing the verification challenges caused by the "anytime issuance" nature of electronic invoices, this invention introduces a "transaction time inference logic." By associating itineraries (PDF or API data) from ride-hailing platforms with check-in times from hotel bills, it prioritizes transaction time over issue time for trajectory matching, effectively resolving time logic misjudgments caused by reissued invoices. Invoice-Trajectory-Accommodation Multi-Dimensional Consistency Cross-Validation Model: A conflict detection model based on geofencing and time windows is established to verify the consistency of "consumption voucher and trajectory location," "consumption time and travel status," and "accommodation time and business trip period," automatically identifying logical paradoxes.

[0085] 3. Dynamic mutually exclusive deduction subsidy calculation engine: Traditional subsidy calculations mostly rely on manual data entry (employees fill in 3 days' worth of data themselves). The previous system (300 yuan) was prone to errors and couldn't exclude already reimbursed meals. This invention achieves automatic mutual exclusion linkage between "reimbursement items" and "subsidy items." The system automatically identifies the "breakfast included" label on hotel bills and the "hospitality" attribute on restaurant invoices, and automatically deducts these attributes when calculating subsidies, achieving refined cost control and preventing "duplicate enjoyment of benefits." Specifically, it uses a dynamic subsidy deduction engine based on invoice attributes: it uses OCR to identify the "breakfast included" label on hotel bills and the "hospitality" attribute on restaurant invoices, and automatically triggers mutually exclusive deduction logic when calculating daily subsidies to prevent duplicate enjoyment of benefits.

[0086] 4. Geofencing Cross-validation of Multi-Source Heterogeneous Credentials: Utilizing Geographic Information Technology (GIS), this method establishes the inclusion relationship between credentials and cities. It not only verifies city consistency but also checks more granular "reasonableness." For example, it verifies the distance between hotel locations and office / airport locations, identifying risks associated with non-business-related abnormal travel such as detours or visits to scenic spots.

[0087] 5. Transaction Time Inference and Calibration Method: To address the issue of delayed electronic invoice issuance time, an algorithm is used to infer the actual transaction time by associating itinerary, order details, or context time, serving as a benchmark for verification and subsidy calculation.

[0088] 6. Differentiated segmented subsidy calculation logic: Supports automatic calculation of daily subsidies by city level, natural day slice, and cross-city mobility rules (such as choosing the higher / lower rate), and generates a visual bill containing calculation formulas and deduction details.

[0089] The corresponding beneficial effects are as follows: Eliminating fraudulent travel expense claims: Through rigorous verification of time and space trajectories, we accurately intercept violations such as fabricated itineraries, proxy ticket claims, and non-business-related spending. This is expected to reduce the cost of travel violations by 10%-20%.

[0090] Achieving zero-manual subsidy calculation: The complex travel system (city classification, time-based deductions, and deductions for early departures) is coded. The system automatically calculates subsidies based on travel routes, eliminating errors from manual calculations and improving financial audit efficiency by over 90%.

[0091] Improve the employee expense reimbursement experience: Employees no longer need to manually calculate "which day is half a day, which day is a full day, and how much will be deducted," but only need to confirm the calculation results generated by the system. In addition, the system will provide real-time alerts in case of any discrepancies, avoiding the frustration of rejection afterward.

[0092] Strengthened compliance audit capabilities: All subsidy calculations are verifiable (based on actual airfare and bank slip dates), forming a complete chain of evidence. Even in audits conducted years later, the basis for each subsidy disbursement can be clearly traced.

[0093] Optimize travel policy formulation: Based on real travel data, companies can identify unreasonable aspects of their travel policies (such as excessively low subsidies in a certain city that force employees to pay out of pocket), thereby optimizing the standards and improving employee satisfaction.

[0094] Example 2: This invention also provides a travel expense itinerary consistency verification and automatic subsidy calculation device. This travel expense itinerary consistency verification and automatic subsidy calculation device is mainly used to execute the travel expense itinerary consistency verification and automatic subsidy calculation method provided in Embodiment 1 of this invention. The following is a detailed description of the travel expense itinerary consistency verification and automatic subsidy calculation device provided in this invention.

[0095] Figure 2 This is a schematic diagram of a travel expense itinerary consistency verification and automatic subsidy calculation device according to an embodiment of the present invention, as shown below. Figure 2 As shown, the device mainly includes: a data acquisition and generation unit 10, a construction unit 20, a verification unit 30, a calculation and summarization unit 40, and a generation unit 50, wherein: The acquisition and generation unit is used to acquire multimodal travel data and generate standard trajectory point data based on the multimodal travel data; The building unit is used to construct a full-link travel spatiotemporal trajectory chain based on standard trajectory point data; The verification unit is used to perform cross-validation of cross-document spatiotemporal consistency based on the end-to-end travel spatiotemporal trajectory chain; The calculation and aggregation unit is used to dynamically calculate the travel subsidy amount by time slice based on the preset travel policy rules and the whole-link travel time and space trajectory chain if the verification is successful, and to aggregate the final total travel subsidy amount. The generation unit is used to generate a travel compliance report containing calculation details based on the final total travel allowance and the travel allowance amount for each time period, and to generate a payment instruction based on the final total travel allowance.

[0096] This invention provides a travel expense itinerary consistency verification and automatic subsidy calculation device, comprising: collecting multimodal travel data and generating standard trajectory point data based on the multimodal travel data; constructing a full-link travel spatiotemporal trajectory chain based on the standard trajectory point data; performing cross-voucher spatiotemporal consistency cross-verification based on the full-link travel spatiotemporal trajectory chain; if the verification is successful, dynamically calculating the travel subsidy amount by time slice based on preset travel policy rules and the full-link travel spatiotemporal trajectory chain, and summarizing to obtain the final total travel subsidy amount; generating a travel compliance report containing calculation details based on the final total travel subsidy amount and the travel subsidy amount for each time period, and generating a payment instruction based on the final total travel subsidy amount. As described above, the travel expense itinerary consistency verification and automatic subsidy calculation device of the present invention effectively overcomes the spatial-temporal logic blind spots caused by data fragmentation and coarse-grained rules in traditional technologies by collecting multimodal travel data to construct a full-link travel spatiotemporal trajectory chain and performing cross-validation of spatiotemporal consistency across vouchers. This achieves accurate and automatic identification of travel violations while ensuring the granularity and accuracy of subsidy calculation through a time-slicing dynamic calculation mechanism. Ultimately, it achieves the transformation of travel business from "manual and extensive review" to "automated and refined compliance management throughout the entire lifecycle," significantly improving review efficiency and reducing the risk of corporate fund loss. It also alleviates the technical problem that traditional travel review technology makes it impossible for enterprises to achieve refined management and automatic compliance verification of the entire lifecycle of travel business.

[0097] Optionally, the data acquisition and generation unit is also used to: connect to business travel platforms and ride-hailing platforms via application programming interfaces (APIs) to obtain structured order data, and mark the structured order data as the benchmark fact data with the highest confidence level; use optical character recognition (OCR) technology to identify ticket images uploaded by users, and extract the take-off and landing times and locations of transportation tickets, check-in and check-out information and room type characteristics of accommodation tickets, and time and location information of consumption invoices; with the user's authorization, collect the check-in coordinates of mobile office applications or parse the EXIF ​​geographic location information of uploaded image files as auxiliary data; establish a global city geographic database, uniformly map the location descriptions in the above multi-source data to standard city codes, and uniformly convert all time data into Coordinated Universal Time (UTC) timestamps for storage, generating standard trajectory point data containing timestamps, location coordinates, event types, and voucher sources.

[0098] Optionally, the construction unit is also used to: extract large-scale transportation order data from standard trajectory point data as key nodes, construct a time axis skeleton, and divide the time axis skeleton into multiple road segment intervals; extract accommodation order data from standard trajectory point data as dwell anchor points, and overlay them onto the time axis skeleton to determine the basic dwell time period; use a spatiotemporal interpolation algorithm, based on preset state determination rules, to fill the blank time periods on the time axis skeleton except for key nodes and basic dwell time periods, generating a continuous time series including transit status, lodging status, and activity status; detect whether there is a time interval between adjacent key nodes and missing transportation vouchers, and if so, mark the corresponding time period as an unknown displacement status and associate it with a supplementary explanation request; and assemble and generate a full-link travel spatiotemporal trajectory chain based on key nodes, basic dwell time periods, the filled continuous time series, and the marked unknown displacement status.

[0099] Optionally, the verification unit is also used to: calculate the geographical distance between the location coordinates of the consumption invoice and the city to which the current trajectory segment belongs; if it exceeds the preset same-city radius threshold, it is determined to be an abnormal cross-regional consumption; detect whether the business occurrence time of the consumption invoice falls within the on-the-go time period of major transportation; if so, and the consumption type of the consumption invoice is not airport pick-up and drop-off service, it is determined to be an abnormal on-the-go illegal consumption; detect whether there are two or more consumption invoices from different locations whose business occurrence time periods overlap; if so, it is determined to be an abnormal spatiotemporal logic conflict; verify the distance between the start and end points of continuous intra-city transportation vouchers; if the distance between the previous end point and the next start point is too far and there is no traffic record in between, it is determined to be an abnormal trajectory break.

[0100] Optionally, the calculation and aggregation unit is also used to: divide the entire travel time-space trajectory chain into multiple time slices according to natural days; for each time slice, query a preset city classification matrix according to the city level of the trajectory in the slice to determine the benchmark subsidy amount for the day; based on the departure or return time point in the time slice, determine whether it is the first or last day of the business trip, and apply the corresponding first and last day subsidy coefficient to correct the benchmark subsidy amount; check the hotel bill data associated with the day, and if it contains a breakfast service identifier, deduct the breakfast subsidy amount from the corrected subsidy amount; check whether there are reimbursed business entertainment expense vouchers or meeting meal expense vouchers on the day, and if so, deduct the meal allowance amount for the corresponding period from the remaining subsidy amount; check whether there are company vehicle dispatch records on the day, and if so, deduct the urban transportation subsidy amount from the remaining subsidy amount to obtain the travel subsidy amount.

[0101] Optionally, the calculation and aggregation unit is also used to: if the same natural day crosses cities of different levels, determine the benchmark subsidy amount for that day based on the preset cross-city rule configuration and city classification matrix.

[0102] Optionally, the generation unit is also used to: generate a visualized travel trajectory map and mark compliant travel segments and abnormal transition points on the travel trajectory map with different colors; distinguish between hard and soft defects for documents that fail verification; directly remove hard defects and block the reimbursement process; generate an explanation submission portal for soft defects and re-trigger calculation after manual approval; generate a travel compliance report based on the travel trajectory map, hard defects, soft defects, daily benchmark subsidy amount, subsidy coefficient, deduction details and travel subsidy amount, and generate a payment instruction based on the final total travel subsidy amount.

[0103] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0104] like Figure 3 As shown in the embodiment of this application, an electronic device 600 includes a processor 601, a memory 602, and a bus. The memory 602 stores machine-readable instructions that can be executed by the processor 601. When the electronic device is running, the processor 601 and the memory 602 communicate via the bus. The processor 601 executes the machine-readable instructions to perform the steps of the above-described method for verifying the consistency of travel expense itinerary trajectory and calculating automatic subsidies.

[0105] Specifically, the aforementioned memory 602 and processor 601 can be general-purpose memory and processor, without any specific limitations. When the processor 601 runs the computer program stored in the memory 602, it can execute the aforementioned method for verifying the consistency of travel expense itinerary trajectory and calculating automatic subsidies.

[0106] The processor 601 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 601 or by instructions in software form. The processor 601 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 602, and processor 601 reads the information from memory 602 and, in conjunction with its hardware, completes the steps of the above method.

[0107] Corresponding to the above-mentioned method for verifying the consistency of travel expense itineraries and calculating automatic subsidies, this application also provides a computer-readable storage medium storing machine-executable instructions. When the machine-executable instructions are called and run by a processor, the machine-executable instructions cause the processor to perform the steps of the above-mentioned method for verifying the consistency of travel expense itineraries and calculating automatic subsidies.

[0108] The travel expense itinerary consistency verification and automatic subsidy calculation device provided in this application embodiment can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in this application embodiment are the same as those in the foregoing method embodiments. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the foregoing method embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can all be referred to the corresponding processes in the above method embodiments, and will not be repeated here.

[0109] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0110] For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0111] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0112] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0113] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the travel expense itinerary consistency verification and automatic subsidy calculation method described in the various embodiments of this application. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0114] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0115] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for verifying the consistency of travel expense itineraries and automatically calculating subsidies, characterized in that, include: Collect multimodal travel data and generate standard trajectory point data based on the multimodal travel data; Construct a full-link business travel spatiotemporal trajectory chain based on the aforementioned standard trajectory point data; Perform cross-validation of cross-document spatiotemporal consistency based on the aforementioned end-to-end travel spatiotemporal trajectory chain; If the verification is successful, the travel subsidy amount will be dynamically calculated based on the preset travel policy rules and the full-link travel time-space trajectory chain, and the final total travel subsidy amount will be obtained by summarizing. A travel compliance report containing calculation details is generated based on the final total travel allowance and the travel allowance amount for each time period, and a payment instruction is generated based on the final total travel allowance.

2. The method according to claim 1, characterized in that, Standard trajectory point data is generated based on the multimodal travel data, including: By connecting to business travel platforms and ride-hailing platforms through application programming interfaces (APIs), structured order data is obtained, and the structured order data is marked as the benchmark fact data with the highest confidence level. Optical character recognition technology is used to identify the ticket images uploaded by users and extract the take-off and landing times and locations of transportation tickets, check-in and check-out information and room type characteristics of accommodation tickets, and time and location information of consumption invoices. With user authorization, collect the check-in coordinates of mobile office applications or parse the EXIF ​​geolocation information of uploaded image files as auxiliary data; A global city geographic database is established, and the location descriptions in the above multi-source data are uniformly mapped to standard city codes. All time data are uniformly converted into Coordinated Universal Time timestamps for storage, generating the standard trajectory point data containing timestamps, location coordinates, event types, and credential sources.

3. The method according to claim 1, characterized in that, Based on the aforementioned standard trajectory point data, a full-link travel spatiotemporal trajectory chain is constructed, including: Extract the traffic order data from the standard trajectory point data as key nodes, construct a time axis skeleton, and divide the time axis skeleton into multiple road segment intervals; The accommodation order data in the standard trajectory point data is extracted as the anchor point of the stay and superimposed on the time axis skeleton to determine the basic stay period; Using a spatiotemporal interpolation algorithm and based on preset state determination rules, the blank time periods on the time axis skeleton, excluding the key nodes and the basic dwell time periods, are filled with states to generate a continuous time series that includes the transit state, the destination state, and the activity state. Detect whether there is a time interval and missing traffic credentials between adjacent key nodes. If so, mark the corresponding time period as an unknown displacement state and associate it with a supplementary explanation request. Based on the key nodes, the basic dwell time period, the filled continuous time series, and the marked unknown displacement states, the full-link travel spatiotemporal trajectory chain is assembled and generated.

4. The method according to claim 1, characterized in that, Based on the aforementioned end-to-end travel spatiotemporal trajectory chain, cross-document spatiotemporal consistency cross-validation is performed, including: Calculate the geographical distance between the location coordinates of the consumption invoice and the city to which the current trajectory segment belongs. If it exceeds the preset threshold of the same city radius, it is judged as an abnormal consumption in a different location. The system checks whether the transaction time of the consumption invoice falls within the on-the-go time period of the major transportation. If so, and the consumption type of the consumption invoice is not airport pick-up and drop-off service, it is determined to be an abnormal on-the-go consumption violation. The system checks whether there are two or more consumption invoices from different locations whose business transactions overlap in time. If so, it is determined to be an anomaly of spatiotemporal logic conflict. The distance between the start and end points of a continuous urban traffic pass is checked. If the distance between the previous end point and the next start point is too far and there is no traffic record in between, it is judged as an abnormal trajectory break.

5. The method according to claim 1, characterized in that, Based on preset travel policy rules and the aforementioned end-to-end travel time-space trajectory chain, the travel allowance amount is dynamically calculated by time slice, including: The entire-link travel time-space trajectory chain is divided into multiple time slices according to natural days; For each time slice, a preset city classification matrix is ​​queried based on the city level of the trajectory within the slice to determine the benchmark subsidy amount for that day. Based on the departure or return time points within the time slice, determine whether it is the first or last day of the business trip, and apply the corresponding first and last day subsidy coefficients to correct the benchmark subsidy amount. If the hotel bill data associated with the test day includes a breakfast service identifier, the breakfast subsidy amount will be deducted from the revised subsidy amount. If there are any reimbursed business entertainment expense vouchers or meeting meal expense vouchers on the day of the inspection, the meal allowance for the corresponding period will be deducted from the remaining subsidy amount. If there is a company vehicle dispatch record on the day of the check, and if so, the urban transportation subsidy amount is deducted from the remaining subsidy amount to obtain the travel subsidy amount.

6. The method according to claim 5, characterized in that, Based on the city level of the trajectory within the slice, a preset city classification matrix is ​​queried to determine the benchmark subsidy amount for the day, including: If the same natural day involves crossing cities of different levels, the benchmark subsidy amount for that day is determined according to the preset cross-city rule configuration and the city classification matrix.

7. The method according to claim 1, characterized in that, A travel compliance report containing calculation details is generated based on the final total travel allowance and the travel allowance amounts for each time period. A payment instruction is then generated based on the final total travel allowance, including: Generate a visual travel trajectory map, and use different colors to mark compliant travel segments and abnormal change points on the travel trajectory map; For documents that fail verification, distinguish between hard and soft defects; hard defects are directly removed and the reimbursement process is blocked; for soft defects, an explanation submission entry is generated and the calculation is retried after manual approval. The travel compliance report is generated based on the travel trajectory map, the hard injury anomaly, the soft injury anomaly, the daily benchmark subsidy amount, the subsidy coefficient, the deduction details, and the travel subsidy amount. The payment instruction is then generated based on the final total travel subsidy amount.

8. A device for verifying the consistency of travel expense itineraries and automatically calculating subsidies, characterized in that, include: The acquisition and generation unit is used to acquire multimodal travel data and generate standard trajectory point data based on the multimodal travel data; The construction unit is used to construct a full-link travel spatiotemporal trajectory chain based on the standard trajectory point data; The verification unit is used to perform cross-document spatiotemporal consistency verification based on the end-to-end travel spatiotemporal trajectory chain; The calculation and aggregation unit is used to dynamically calculate the travel subsidy amount by time slice based on the preset travel system rules and the full-link travel spatiotemporal trajectory chain if the verification is successful, and to aggregate the final total travel subsidy amount. The generation unit is used to generate a travel compliance report containing calculation details based on the final total travel allowance and the travel allowance amount for each time period, and to generate a payment instruction based on the final total travel allowance.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program thereon, characterized in that, The computer program is executed by the processor to perform the method of any one of claims 1 to 7.

Images