Disclosure of Invention
One of the objectives of the present invention is to provide a method and system for automatically reimbursing travel fees based on bill identification and configuration engine.
The embodiment of the invention provides a method for automatically reimbursing travel expenses based on bill identification and configuration engine, which comprises the following steps:
step S1: recognizing text information on a bill file provided by an employee based on an OCR technology;
step S2: restoring the travel information of the employee based on the text information;
step S3: a parameter engine is configured, and based on the parameter engine, the reimbursement cost and travel subsidy suitable for the travel information are calculated;
step S4: and the reimbursement expense is reimbursed, and meanwhile, the travel subsidy is issued.
Preferably, the step S2: based on the text information, restoring the travel information of the employee, including:
converting the text information into bill structured information based on a preset conversion rule;
extracting time, place, personnel, vehicles, expense types, money amounts and taxes in the bill structured information;
determining a first trip item based on the time, location, person, and the vehicle, the first trip item comprising: origin-destination time, origin-destination place, trip person and trip tool;
determining the first stroke item of the same pedestrian as the first stroke item, taking the first stroke item as a second stroke item, and meanwhile, carrying out time sequence sequencing on the second stroke item based on the corresponding origin-destination time to obtain the complete stroke of the pedestrian;
merging the complete strokes of the pedestrians with the same complete stroke to obtain a merged stroke;
and summarizing the combined travel to obtain the travel information of the staff.
Preferably, the step S3: and a parameter engine is configured, based on the parameter engine, the suitable reimbursement cost and travel subsidy of the travel information are calculated, and the parameter engine comprises the following steps:
determining a plurality of cost items for the pedestrian based on the person, the fee type, and the amount, the plurality of cost items including: traffic, lodging and miscellaneous fees;
summarizing the amount corresponding to the same expense item to obtain the expense to be reimbursed of the pedestrian;
acquiring a plurality of difference standard systems of a company, and configuring a difference standard checking engine based on the difference standard systems;
inputting the expense to be reimbursed corresponding to each expense item of the pedestrian into the difference standard checking engine, and checking whether the expense meets the difference standard for reimbursement;
if yes, summarizing the expense to be reimbursed to obtain the reimbursement expense of the pedestrian;
acquiring a subsidy system of the company, and configuring a subsidy calculation engine based on the subsidy system;
and inputting the travel information into the subsidy calculation engine to obtain the travel subsidy of each pedestrian.
Preferably, the step S4: and reimbursing the reimbursement expense, and simultaneously issuing travel subsidies, wherein the reimbursing expense reimbursement expense is composed of the following steps:
summarizing the reimbursement cost of each pedestrian to obtain a cost sum;
summarizing the travel subsidies of the pedestrians to obtain subsidy sums;
summarizing the expense sum and the subsidy sum to obtain a dispensing amount;
acquiring the budget amount of the company;
if the issued amount is less than or equal to the budget amount, calling an interface of a bill checking platform, and checking the authenticity and/or repeated reimbursement of the bill information;
if the bill information is qualified, generating an electronic reimbursement bill by supplementing the bill, the complete journey of the pedestrian, reimbursement cost and travel based on a preset reimbursement bill template;
sending the electronic reimbursement bill to a client corresponding to the leader;
and after receiving the agreement request replied by the client, sending the electronic reimbursement bill to a financial department for accounting reimbursement and issuing.
Preferably, the method for automatically reimbursing travel expenses based on the ticket recognition and configuration engine further comprises:
performing travel simulation based on the travel information to obtain a simulation cost sum, and performing authenticity verification on the cost sum based on the simulation cost sum;
wherein, carry out the stroke simulation based on the stroke information, include:
extracting a first travel item in the travel information and taking the first travel item as a second travel item;
acquiring a preset journey simulation engine, inputting a second journey item into the journey simulation engine, and acquiring a first simulation cost;
acquiring additional information provided by an employee, wherein the additional information comprises: a plurality of first special terms encountered in the trip;
if a second special item corresponding to the second trip item exists in the additional information, analyzing the second special item to obtain a special reason and a special degree;
acquiring preset special big data, and determining whether the special reason and the special degree are real or not based on the special big data;
if yes, determining the commonly corresponding up-regulation amplitude of the special reason and the special degree based on a preset up-regulation amplitude library;
based on the up-regulation amplitude, the first simulation cost is up-regulated to obtain a second simulation cost;
replacing the first simulation cost with the second simulation cost;
summarizing the first simulation cost and the second simulation cost to obtain a simulation cost sum;
and verifying the authenticity of the cost sum based on the simulated cost sum, comprising:
and determining a first difference between the simulation cost sum and the cost sum, if the difference is within a preset first difference range, determining that the cost sum is real, otherwise, completing verification.
Preferably, the determining whether the specific reason and the specific degree are true based on the specific big data includes:
analyzing the special reason to obtain a plurality of reason items, wherein the reason items comprise: a second time, a second location, and a cause type;
formulating a rule based on a preset time interval, and formulating a time interval corresponding to the second time;
formulating a corresponding area of the second location based on a preset area formulation rule;
determining at least one retrieval mode corresponding to the reason type based on a preset retrieval mode library;
based on a retrieval mode, retrieving event data occurring in the area in the time interval from the special big data;
extracting a plurality of first event items in the event data, the first event items including: the system comprises a first publisher, a first publishing mode, a first publishing platform and a first event;
determining at least one first characteristic corresponding to the reason type based on a preset characteristic library;
performing feature extraction on the first event to obtain a plurality of second features;
matching the first characteristic with the second characteristic, and if the first characteristic and the second characteristic are matched, taking the corresponding first event item as a second event item;
extracting a first publisher in the second event item and using the first publisher as a second publisher;
extracting a first publishing mode in the second event item, and using the first publishing mode as a second publishing mode, wherein the second publishing mode comprises: direct publishing and indirect publishing;
extracting a first publishing platform in the second event item, and using the first publishing platform as a second publishing platform;
when the second issuing mode is direct issuing, acquiring a first verification value of a second issuer, and acquiring a second verification value of a second issuing platform;
if the first verification value is greater than or equal to a preset first verification value threshold value and/or the second verification value is greater than or equal to a preset second verification value threshold value, rejecting a corresponding second event item;
meanwhile, when the second issuing mode is indirect issuing, the corresponding second event item is removed;
when second event items needing to be removed in the second event items are all removed, a plurality of third event items are obtained;
extracting a first event in the third event item and taking the first event as a second event;
determining a model base based on a preset degree, and determining a degree determination model corresponding to the reason type;
inputting the second event into a degree determination model to obtain a true degree;
and determining a second difference value between the special degree and the real degree, if the second difference value is within a preset second difference value range, the special reason and the degree are both real, otherwise, the special reason and the degree are not real, and finishing the determination.
The embodiment of the invention provides a system for automatically reimbursing travel expenses based on bill identification and configuration engine, which comprises:
the recognition module is used for recognizing text information on the bill file provided by the employee based on an OCR technology;
the restoring module is used for restoring the travel information of the staff based on the text information;
the calculation module is used for configuring a parameter engine and calculating the reimbursement cost and travel subsidy suitable for the travel information based on the parameter engine;
and the reimbursement module is used for reimbursing the reimbursement cost and issuing the travel subsidy.
Preferably, the reduction module performs the following operations:
converting the text information into bill structured information based on a preset conversion rule;
extracting time, place, personnel, vehicles, expense types, money amounts and taxes in the bill structured information;
determining a first trip item based on the time, location, person, and the vehicle, the first trip item comprising: origin-destination time, origin-destination place, trip person and trip tool;
determining the first stroke item of the same pedestrian as the first stroke item, taking the first stroke item as a second stroke item, and meanwhile, carrying out time sequence sequencing on the second stroke item based on the corresponding origin-destination time to obtain the complete stroke of the pedestrian;
merging the complete strokes of the pedestrians with the same complete stroke to obtain a merged stroke;
and summarizing the combined travel to obtain the travel information of the staff.
Preferably, the calculation module performs the following operations:
determining a plurality of cost items for the pedestrian based on the person, the fee type, and the amount, the plurality of cost items including: traffic, lodging and miscellaneous fees;
summarizing the amount corresponding to the same expense item to obtain the expense to be reimbursed of the pedestrian;
acquiring a plurality of difference standard systems of a company, and configuring a difference standard checking engine based on the difference standard systems;
inputting the expense to be reimbursed corresponding to each expense item of the pedestrian into the difference standard checking engine, and checking whether the expense meets the difference standard for reimbursement;
if yes, summarizing the expense to be reimbursed to obtain the reimbursement expense of the pedestrian;
acquiring a subsidy system of the company, and configuring a subsidy calculation engine based on the subsidy system;
and inputting the travel information into the subsidy calculation engine to obtain the travel subsidy of each pedestrian.
Preferably, the reimbursement module performs the following operations:
summarizing the reimbursement cost of each pedestrian to obtain a cost sum;
summarizing the travel subsidies of the pedestrians to obtain subsidy sums;
summarizing the expense sum and the subsidy sum to obtain a dispensing amount;
acquiring the budget amount of the company;
if the issued amount is less than or equal to the budget amount, calling an interface of a bill checking platform, and checking the authenticity and/or repeated reimbursement of the bill information;
if the bill information is qualified, generating an electronic reimbursement bill by supplementing the bill, the complete journey of the pedestrian, reimbursement cost and travel based on a preset reimbursement bill template;
sending the electronic reimbursement bill to a client corresponding to the leader;
and after receiving the agreement request replied by the client, sending the electronic reimbursement bill to a financial department for accounting reimbursement and issuing.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
The embodiment of the invention provides a method for automatically reimbursing travel expenses based on bill identification and configuration engine, as shown in fig. 1, comprising the following steps:
step S1: recognizing text information on a bill file provided by an employee based on an OCR technology;
step S2: restoring the travel information of the employee based on the text information;
step S3: a parameter engine is configured, and based on the parameter engine, the reimbursement cost and travel subsidy suitable for the travel information are calculated;
step S4: and the reimbursement expense is reimbursed, and meanwhile, the travel subsidy is issued.
The working principle and the beneficial effects of the technical scheme are as follows:
based on OCR [ Optical Character Recognition, Optical Character Recognition ] techniques, identify employee-provided tickets [ for example: text information on electronic invoice, paper invoice image, etc.; determining the travel information of the employee based on the text information; determining a corresponding reimbursement fee based on the configured engine; automatically reimburse the staff;
the embodiment of the invention determines the travel journey of the staff based on the bill provided by the staff, automatically determines the reimbursement cost, does not need the user to manually fill in the bill information and the journey information, improves the convenience, and simultaneously avoids the problem of filling information distortion.
The embodiment of the present invention provides a method for automatically reimbursing travel fees based on a ticket recognition and configuration engine, as shown in fig. 2, in step S2: based on the text information, restoring the travel information of the employee, including:
step S201: converting the text information into bill structured information based on a preset conversion rule;
step S202: extracting time, place, personnel, vehicles, expense types, money amounts and taxes in the bill structured information;
step S203: determining a first trip item based on the time, location, person, and the vehicle, the first trip item comprising: origin-destination time, origin-destination place, trip person and trip tool;
step S204: determining the first stroke item of the same pedestrian as the first stroke item, taking the first stroke item as a second stroke item, and meanwhile, carrying out time sequence sequencing on the second stroke item based on the corresponding origin-destination time to obtain the complete stroke of the pedestrian;
step S205: merging the complete strokes of the pedestrians with the same complete stroke to obtain a merged stroke;
step S206: and summarizing the combined travel to obtain the travel information of the staff.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset conversion rule is specifically as follows: converting text information on a ticket into information that can be utilized [ ticket structuring information ], such as: the text information on the train ticket is '07/23/07/43/23/2021/Suzhou to Hankou', 239.00 Yuan ',' current car on the limited riding day 'and' Suzhou sale ', etc., which is converted into' 07/23/2021/07/43/Suzhou, 239.00 Yuan ', passenger/certain, train, transportation fee';
each trip item contains time, location, and/or cost; sequencing the places in the schedule items based on time; the two adjacent places are the departure place and the destination, and the corresponding time is the departure time and the arrival time; for example: the employee's train ticket proves that he went from train sitting in Suzhou to Hankou on 07:43 at 23.07: 07: 24.07: 2021, and when he went to Hankou, he went from Hankou train station to a hotel for 3 places and 2 trips;
the embodiment of the invention automatically determines the travel information of the staff based on the bills provided by the staff, has reasonable setting and improves the convenience.
The embodiment of the invention provides a method for automatically reimbursing travel expenses based on bill identification and configuration engine, which comprises the following steps of S3: and a parameter engine is configured, based on the parameter engine, the suitable reimbursement cost and travel subsidy of the travel information are calculated, and the parameter engine comprises the following steps:
determining a plurality of cost items for the pedestrian based on the person, the fee type, and the amount, the plurality of cost items including: traffic, lodging and miscellaneous fees;
summarizing the amount corresponding to the same expense item to obtain the expense to be reimbursed of the pedestrian;
acquiring a plurality of difference standard systems of a company, and configuring a difference standard checking engine based on the difference standard systems;
inputting the expense to be reimbursed corresponding to each expense item of the pedestrian into the difference standard checking engine, and checking whether the expense meets the difference standard for reimbursement;
if yes, summarizing the expense to be reimbursed to obtain the reimbursement expense of the pedestrian;
acquiring a subsidy system of the company, and configuring a subsidy calculation engine based on the subsidy system;
and inputting the travel information into the subsidy calculation engine to obtain the travel subsidy of each pedestrian.
The working principle and the beneficial effects of the technical scheme are as follows:
the standard difference system specifically comprises: for example, defining a company difference standard level system, a cabin space allowing passengers to take transportation vehicles such as planes and trains, and lodging limits of various types of cities, defining a post level system of personnel, and binding the post level system with company personnel and the company difference standard system; the auxiliary system specifically comprises: for example, preset subsidy types, calculation formulas of various subsidies, triggering conditions of various subsidies, issuing standards and issuing adjustments of various subsidies for various companies; the configuration engine may be based on a machine learning algorithm.
The subsidy standards of the employees at different posts are different, so that whether the final reimbursement cost meets the corresponding difference standard of the employee or not is determined, the setting is reasonable, and the actual use requirement is better met; need not financial affairs and check by oneself whether every expense of reimbursing satisfies company's travel standard, promoted the convenience.
The embodiment of the invention provides a method for automatically reimbursing travel expenses based on bill identification and configuration engine, which comprises the following steps of S4: and reimbursing the reimbursement expense, and simultaneously issuing travel subsidies, wherein the reimbursing expense reimbursement expense is composed of the following steps:
summarizing the reimbursement cost of each pedestrian to obtain a cost sum;
summarizing the travel subsidies of the pedestrians to obtain subsidy sums;
summarizing the expense sum and the subsidy sum to obtain a dispensing amount;
acquiring the budget amount of the company;
if the issued amount is less than or equal to the budget amount, calling an interface of a bill checking platform, and checking the authenticity and/or repeated reimbursement of the bill information;
if the bill information is qualified, generating an electronic reimbursement bill by supplementing the bill, the complete journey of the pedestrian, reimbursement cost and travel based on a preset reimbursement bill template;
sending the electronic reimbursement bill to a client corresponding to the leader;
and after receiving the agreement request replied by the client, sending the electronic reimbursement bill to a financial department for accounting reimbursement and issuing.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset reimbursement bill template specifically comprises the following steps: a template, on which bills, travel, reimbursement fee and collection account can be filled;
after the reimbursement bill is generated, the reimbursement bill is directly issued to the financial department, and the reimbursement bill is calculated by the financial department, so that the workload of staff of the financial department is reduced.
The embodiment of the invention provides a method for automatically reimbursing travel expenses based on bill identification and configuration engine, which also comprises the following steps:
performing travel simulation based on the travel information to obtain a simulation cost sum, and performing authenticity verification on the cost sum based on the simulation cost sum;
wherein, carry out the stroke simulation based on the stroke information, include:
extracting a first travel item in the travel information and taking the first travel item as a second travel item;
acquiring a preset journey simulation engine, inputting a second journey item into the journey simulation engine, and acquiring a first simulation cost;
acquiring additional information provided by an employee, wherein the additional information comprises: a plurality of first special terms encountered in the trip;
if a second special item corresponding to the second trip item exists in the additional information, analyzing the second special item to obtain a special reason and a special degree;
acquiring preset special big data, and determining whether the special reason and the special degree are real or not based on the special big data;
if yes, determining the commonly corresponding up-regulation amplitude of the special reason and the special degree based on a preset up-regulation amplitude library;
based on the up-regulation amplitude, the first simulation cost is up-regulated to obtain a second simulation cost;
replacing the first simulation cost with the second simulation cost;
summarizing the first simulation cost and the second simulation cost to obtain a simulation cost sum;
and verifying the authenticity of the cost sum based on the simulated cost sum, comprising:
and determining a first difference between the simulation cost sum and the cost sum, if the difference is within a preset first difference range, determining that the cost sum is real, otherwise, completing verification.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset journey simulation engine specifically comprises: an engine generated by learning a large number of records of manually-performed travel simulation based on a machine learning algorithm, wherein the engine can perform simulation based on travel information, such as: the journey is from a company to a Nanjing railway station, the journey from the company to the Nanjing railway station for the passengers to take by the staff is simulated, and the taking cost is calculated based on the distance of the journey and the calculation standard of the local taxi taking cost; the preset special big data is specifically as follows: for example, the special situation is traffic jam, and the big data is various data of traffic jam places (for example, traffic jam data issued by a third-party navigation platform, a traffic jam photo and positioning issued on a microblog); the preset up-regulation amplitude library specifically comprises the following steps: the database stores the corresponding up-regulation amplitudes of different special reasons and special degrees; the preset first difference range specifically includes: for example, ± 1.2;
staff in the journey, can meet special circumstances and cause the expense to exceed the normal use expense inevitably [ for example: the staff only need to integrate special things met by the staff into additional information and upload the additional information, determine the special reasons and special degrees in the additional information based on big data to verify the authenticity, and if the verification is passed, reasonably simulate the cost to be adjusted up based on the special reasons and special degrees;
the embodiment of the invention verifies the cost and authenticity of the user, and avoids the situation that the employee gives a third-party staff a malicious invoice; meanwhile, the expense is increased due to special conditions encountered by staff in the traveling process, and under general conditions, reimbursers cannot verify the special conditions, so that the problem can be solved by the embodiment of the invention; in addition, the setting is reasonable, the cost and the real misjudgment can not be caused, and the method is more accurate; and finally, the method is more applicable under the trend of data sharing.
The embodiment of the invention provides a method for automatically reimbursing travel expenses based on bill identification and configuration engine, which determines whether a special reason and a special degree are true or not based on special big data, and comprises the following steps:
analyzing the special reason to obtain a plurality of reason items, wherein the reason items comprise: a second time, a second location, and a cause type;
formulating a rule based on a preset time interval, and formulating a time interval corresponding to the second time;
formulating a corresponding area of the second location based on a preset area formulation rule;
determining at least one retrieval mode corresponding to the reason type based on a preset retrieval mode library;
based on a retrieval mode, retrieving event data occurring in the area in the time interval from the special big data;
extracting a plurality of first event items in the event data, the first event items including: the system comprises a first publisher, a first publishing mode, a first publishing platform and a first event;
determining at least one first characteristic corresponding to the reason type based on a preset characteristic library;
performing feature extraction on the first event to obtain a plurality of second features;
matching the first characteristic with the second characteristic, and if the first characteristic and the second characteristic are matched, taking the corresponding first event item as a second event item;
extracting a first publisher in the second event item and using the first publisher as a second publisher;
extracting a first publishing mode in the second event item, and using the first publishing mode as a second publishing mode, wherein the second publishing mode comprises: direct publishing and indirect publishing;
extracting a first publishing platform in the second event item, and using the first publishing platform as a second publishing platform;
when the second issuing mode is direct issuing, acquiring a first verification value of a second issuer, and acquiring a second verification value of a second issuing platform;
if the first verification value is greater than or equal to a preset first verification value threshold value and/or the second verification value is greater than or equal to a preset second verification value threshold value, rejecting a corresponding second event item;
meanwhile, when the second issuing mode is indirect issuing, the corresponding second event item is removed;
when second event items needing to be removed in the second event items are all removed, a plurality of third event items are obtained;
extracting a first event in the third event item and taking the first event as a second event;
determining a model base based on a preset degree, and determining a degree determination model corresponding to the reason type;
inputting the second event into a degree determination model to obtain a true degree;
and determining a second difference value between the special degree and the real degree, if the second difference value is within a preset second difference value range, the special reason and the degree are both real, otherwise, the special reason and the degree are not real, and finishing the determination.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset time interval formulation rule specifically comprises the following steps: for example, the time is 15:00, the range is expanded, and the set time interval is 14:30 to 15: 30; the preset region formulation rule is specifically as follows: within 500 meters of the perimeter of the site; the preset retrieval mode library specifically comprises the following steps: a database, in which search modes corresponding to different reason types are stored, for example: the reason type is traffic jam, and the retrieval mode is to retrieve data issued by a traffic third-party platform; the preset feature library specifically comprises: a database, in which the characteristics corresponding to different reason types are stored; the preset first verification value threshold specifically includes: for example, 80; the preset second verification value threshold specifically includes: for example, 85; the preset degree determination model library specifically comprises: a database, in which degree determination models corresponding to different reason types are stored [ models generated after learning a large number of manual degree determination records by using a machine learning algorithm ]; the preset second difference range specifically includes: plus or minus 2;
the time and the place provided by the staff are not necessarily accurate, the range is expanded, and the time area and the area are respectively manufactured; determining a retrieval mode, and performing retrieval in a targeted manner to reduce the retrieval difficulty; when the first event item meeting the condition is searched out based on the searching mode, determining a second event item corresponding to the reason type; prepare to determine from the second event item whether the specific cause and degree of specificity are authentic, cull indirect publications [ for example: the workload of the second event items carried by other platforms is reduced, the second event items with verification values larger than or equal to the threshold of the verification values are removed, and the problem of misjudgment caused by the fact that the second event items with low reliability are brought into judgment is avoided; when the second event items needing to be removed are all removed, the second event items are set to be real due to special reasons, the degree is determined based on the model, and if the degree has small error with the special degree provided by the staff, the special degree is also set to be real; the additional information provided by the staff is subjected to authenticity verification, so that the staff is prevented from working falsely, and meanwhile, the system is more intelligent.
The embodiment of the invention provides a method for automatically reimbursing travel expenses based on bill identification and configuration engine, wherein the first verification value is obtained by the following steps:
determining a plurality of first guarantors corresponding to the second publisher and a first guaranty mode between the second publisher and the first guarantor based on a preset guarantor library, where the first guaranty mode includes: active and passive warranties;
when the first guarantee mode is active guarantee, the corresponding first guarantee person is used as a second guarantee person;
when the first guarantee mode is a passive guarantee mode, the corresponding first guarantee person is used as a third guarantee person;
respectively acquiring a first malicious value of the second publisher, a second malicious value of the second guarantor and a third malicious value of the third guarantor, and calculating the first verification value based on the first malicious value, the second malicious value and the third malicious value, wherein a calculation formula is as follows:
where ρ is1Is the first verification value, σ1Is the first malicious value, σ2,iSaid second maliciousness value, σ, for the ith said second guarantor3,iSaid third malicious value, n, for the ith said third guarantor1Is the total number of said second guarantor, n2Theta is a predetermined constant, mu, for the total number of the third guarantor1、μ2And mu3Is a preset weight value, mu1>μ2>μ3,ε1,i、ε2,i、ε3,iAnd n3Is an intermediate variable, and is and, else is other, σ1,0Is a preset first malicious value threshold, sigma2,0The second malicious value threshold is preset;
the second verification value is obtained by the following steps:
determining a plurality of first guarantee platforms corresponding to the second issuing platform and a second guarantee mode between the second issuing platform and the first guarantee platform based on a preset guarantee platform library, wherein the second guarantee mode comprises the following steps: active and passive warranties;
when the second guarantee mode is active guarantee, the corresponding first guarantee platform is used as a second guarantee platform;
when the second guarantee mode is a passive guarantee mode, the corresponding first guarantee platform is used as a third guarantee platform;
respectively obtaining a fourth malicious value of the second release platform, a fifth malicious value of the second guarantee platform and a sixth malicious value of the third guarantee platform, and calculating the first verification value based on the fourth malicious value, the fifth malicious value and the sixth malicious value, wherein a calculation formula is as follows:
where ρ is2Is the second verification value, σ4Is the fourth malicious value, σ5,iFor the fifth malicious value, σ, of the ith said second vouching platform6,iSaid sixth malicious value, n, for the ith said third vouching platform4For the total number of said second vouching platforms, n5Theta is a preset constant mu for the total number of the third guarantee platforms4、μ5And mu6Is a preset weight value, mu4>μ5>μ6,ε4,i、ε5,i、ε6,iAnd n6Is an intermediate variable, and is and, else is other, σ3,0Is a preset third malicious value threshold, sigma4,0Is a preset fourth malicious value threshold.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset guarantor library specifically comprises: a database, in which the corresponding guarantors of different users are stored, and the guaranty relationship between different users and guaranties is also stored, the guaranty relationship is divided into active guaranty [ a user sets a user as his or her own guarantor, and the other party agrees ] and passive guaranty [ two users belong to the same group, for example: company, home, etc.); the same principle is used for a preset guarantee platform library; the preset constants are specifically: for example, 2; the preset first malicious value threshold specifically includes: for example, 97; the preset second malicious value threshold specifically includes: for example, 98; the preset third malicious value threshold specifically includes: for example, 95; the preset fourth malicious value threshold specifically is: for example, 98;
when malicious values of a user, a guarantor, a publishing platform, and a guaranty platform are obtained, malicious events that occur on the user or the platform [ for example: issue false data, etc. ]; the verification value is calculated based on different malicious values, so that the working efficiency of the system is greatly improved; according to the embodiment of the invention, a guarantee system is adopted, and under the trend of data sharing, each user establishes a guarantee relationship to generate constraint force, so that the cost and the cost of generating malicious events by the user are increased, and the occurrence rate of the malicious events is reduced; meanwhile, the active guarantee and the passive guarantee are analyzed separately, and the method is more reasonable.
The embodiment of the invention provides a system for automatically reimbursing travel expenses based on bill identification and configuration engine, as shown in fig. 3, comprising:
the recognition module 1 is used for recognizing text information on a bill file provided by an employee based on an OCR technology;
the restoring module 2 is used for restoring the travel information of the staff based on the text information;
the calculation module 3 is used for configuring a parameter engine and calculating the reimbursement cost and travel subsidy suitable for the travel information based on the parameter engine;
and the reimbursement module 4 is used for reimbursing the reimbursement cost and issuing the travel subsidy.
The working principle and the beneficial effects of the technical scheme are as follows:
based on OCR [ Optical Character Recognition, Optical Character Recognition ] techniques, identify employee-provided tickets [ for example: text information on electronic invoice, paper invoice image, etc.; determining the travel information of the employee based on the text information; determining a corresponding reimbursement fee based on the configured engine; automatically reimburse the staff;
the embodiment of the invention determines the travel journey of the staff based on the bill provided by the staff, automatically determines the reimbursement cost, does not need the user to manually fill in the bill information and the journey information, improves the convenience, and simultaneously avoids the problem of filling information distortion.
The embodiment of the invention provides a system for automatically reimbursing travel expenses based on bill identification and configuration engine, wherein a recovery module 2 executes the following operations:
converting the text information into bill structured information based on a preset conversion rule;
extracting time, place, personnel, vehicles, expense types, money amounts and taxes in the bill structured information;
determining a first trip item based on the time, location, person, and the vehicle, the first trip item comprising: origin-destination time, origin-destination place, trip person and trip tool;
determining the first stroke item of the same pedestrian as the first stroke item, taking the first stroke item as a second stroke item, and meanwhile, carrying out time sequence sequencing on the second stroke item based on the corresponding origin-destination time to obtain the complete stroke of the pedestrian;
merging the complete strokes of the pedestrians with the same complete stroke to obtain a merged stroke;
and summarizing the combined travel to obtain the travel information of the staff.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset conversion rule is specifically as follows: converting text information on a ticket into information that can be utilized [ ticket structuring information ], such as: the text information on the train ticket is '07/23/07/43/2021/Ou', 'Suzhou to Hankou', '239.00 Yuan', 'Suzhou sale' and the like, and is converted into '07/23/07/43/23/2021/Suzhou/239.00 Yuan';
each trip item contains time, location, and/or cost; sequencing the places in the schedule items based on time; the two adjacent places are the departure place and the destination, and the corresponding time is the departure time and the arrival time; for example: the employee's train ticket proves that he went from train sitting in Suzhou to Hankou on 07:43 at 23.07: 07: 24.07: 2021, and when he went to Hankou, he went from Hankou train station to a hotel for 3 places and 2 trips;
the embodiment of the invention automatically determines the travel information of the staff based on the bills provided by the staff, has reasonable setting and improves the convenience.
The embodiment of the invention provides a system for automatically reimbursing travel expenses based on bill identification and configuration engine, wherein a computing module 3 executes the following operations:
determining a plurality of cost items for the pedestrian based on the person, the fee type, and the amount, the plurality of cost items including: traffic, lodging and miscellaneous fees;
summarizing the amount corresponding to the same expense item to obtain the expense to be reimbursed of the pedestrian;
acquiring a plurality of difference standard systems of a company, and configuring a difference standard checking engine based on the difference standard systems;
inputting the expense to be reimbursed corresponding to each expense item of the pedestrian into the difference standard checking engine, and checking whether the expense meets the difference standard for reimbursement;
if yes, summarizing the expense to be reimbursed to obtain the reimbursement expense of the pedestrian;
acquiring a subsidy system of the company, and configuring a subsidy calculation engine based on the subsidy system;
and inputting the travel information into the subsidy calculation engine to obtain the travel subsidy of each pedestrian.
The working principle and the beneficial effects of the technical scheme are as follows:
the standard difference system specifically comprises: for example, defining a company difference standard level system, a cabin space allowing passengers to take transportation vehicles such as planes and trains, and lodging limits of various types of cities, defining a post level system of personnel, and binding the post level system with company personnel and the company difference standard system; the auxiliary system specifically comprises: for example, preset subsidy types, calculation formulas of various subsidies, triggering conditions of various subsidies, issuing standards and issuing adjustments of various subsidies for various companies; the configuration engine may be based on a machine learning algorithm.
The subsidy standards of the employees at different posts are different, so that whether the final reimbursement cost meets the corresponding difference standard of the employee or not is determined, the setting is reasonable, and the actual use requirement is better met; need not financial affairs and check by oneself whether every expense of reimbursing satisfies company's travel standard, promoted the convenience.
The embodiment of the invention provides a system for automatically reimbursing travel expenses based on bill identification and configuration engine, wherein the reimbursement module 4 executes the following operations:
summarizing the reimbursement cost of each pedestrian to obtain a cost sum;
summarizing the travel subsidies of the pedestrians to obtain subsidy sums;
summarizing the expense sum and the subsidy sum to obtain a dispensing amount;
acquiring the budget amount of the company;
if the issued amount is less than or equal to the budget amount, calling an interface of a bill checking platform, and checking the authenticity and/or repeated reimbursement of the bill information;
if the bill information is qualified, generating an electronic reimbursement bill by supplementing the bill, the complete journey of the pedestrian, reimbursement cost and travel based on a preset reimbursement bill template;
sending the electronic reimbursement bill to a client corresponding to the leader;
and after receiving the agreement request replied by the client, sending the electronic reimbursement bill to a financial department for accounting reimbursement and issuing.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset reimbursement bill template specifically comprises the following steps: a template, on which bills, travel, reimbursement fee and collection account can be filled;
after the reimbursement bill is generated, the reimbursement bill is directly issued to the financial department, and the reimbursement bill is calculated by the financial department, so that the workload of staff of the financial department is reduced.
The embodiment of the invention provides a system for automatically reimbursing travel expenses based on bill identification and configuration engine, which also comprises:
the verification module is used for carrying out travel simulation based on the travel information to obtain simulation cost sum and carrying out authenticity verification on the cost sum based on the simulation cost sum;
wherein the first verification module performs the following operations:
extracting a first travel item in the travel information and taking the first travel item as a second travel item;
acquiring a preset journey simulation engine, inputting a second journey item into the journey simulation engine, and acquiring a first simulation cost;
acquiring additional information provided by an employee, wherein the additional information comprises: a plurality of first special terms encountered in the trip;
if a second special item corresponding to the second trip item exists in the additional information, analyzing the second special item to obtain a special reason and a special degree;
acquiring preset special big data, and determining whether the special reason and the special degree are real or not based on the special big data;
if yes, determining the commonly corresponding up-regulation amplitude of the special reason and the special degree based on a preset up-regulation amplitude library;
based on the up-regulation amplitude, the first simulation cost is up-regulated to obtain a second simulation cost;
replacing the first simulation cost with the second simulation cost;
summarizing the first simulation cost and the second simulation cost to obtain a simulation cost sum;
and verifying the authenticity of the cost sum based on the simulated cost sum, comprising:
and determining a first difference between the simulation cost sum and the cost sum, if the difference is within a preset first difference range, determining that the cost sum is real, otherwise, completing verification.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset journey simulation engine specifically comprises: an engine generated by learning a large number of records of manually-performed travel simulation based on a machine learning algorithm, wherein the engine can perform simulation based on travel information, such as: the journey is from a company to a Nanjing railway station, the journey from the company to the Nanjing railway station for the passengers to take by the staff is simulated, and the taking cost is calculated based on the distance of the journey and the calculation standard of the local taxi taking cost; the preset special big data is specifically as follows: for example, the special situation is traffic jam, and the big data is various data of traffic jam places (for example, traffic jam data issued by a third-party navigation platform, a traffic jam photo and positioning issued on a microblog); the preset up-regulation amplitude library specifically comprises the following steps: the database stores the corresponding up-regulation amplitudes of different special reasons and special degrees; the preset first difference range specifically includes: for example, ± 1.2;
staff in the journey, can meet special circumstances and cause the expense to exceed the normal use expense inevitably [ for example: the staff only need to integrate special things met by the staff into additional information and upload the additional information, determine the special reasons and special degrees in the additional information based on big data to verify the authenticity, and if the verification is passed, reasonably simulate the cost to be adjusted up based on the special reasons and special degrees;
the embodiment of the invention verifies the cost and authenticity of the user, and avoids the situation that the employee gives a third-party staff a malicious invoice; meanwhile, the expense is increased due to special conditions encountered by staff in the traveling process, and under general conditions, reimbursers cannot verify the special conditions, so that the problem can be solved by the embodiment of the invention; in addition, the setting is reasonable, the cost and the real misjudgment can not be caused, and the method is more accurate; and finally, the method is more applicable under the trend of data sharing.
The embodiment of the invention provides a system for automatically reimbursing travel expenses based on bill identification and configuration engine, wherein a verification module executes the following operations:
analyzing the special reason to obtain a plurality of reason items, wherein the reason items comprise: a second time, a second location, and a cause type;
formulating a rule based on a preset time interval, and formulating a time interval corresponding to the second time;
formulating a corresponding area of the second location based on a preset area formulation rule;
determining at least one retrieval mode corresponding to the reason type based on a preset retrieval mode library;
based on a retrieval mode, retrieving event data occurring in the area in the time interval from the special big data;
extracting a plurality of first event items in the event data, the first event items including: the system comprises a first publisher, a first publishing mode, a first publishing platform and a first event;
determining at least one first characteristic corresponding to the reason type based on a preset characteristic library;
performing feature extraction on the first event to obtain a plurality of second features;
matching the first characteristic with the second characteristic, and if the first characteristic and the second characteristic are matched, taking the corresponding first event item as a second event item;
extracting a first publisher in the second event item and using the first publisher as a second publisher;
extracting a first publishing mode in the second event item, and using the first publishing mode as a second publishing mode, wherein the second publishing mode comprises: direct publishing and indirect publishing;
extracting a first publishing platform in the second event item, and using the first publishing platform as a second publishing platform;
when the second issuing mode is direct issuing, acquiring a first verification value of a second issuer, and acquiring a second verification value of a second issuing platform;
if the first verification value is greater than or equal to a preset first verification value threshold value and/or the second verification value is greater than or equal to a preset second verification value threshold value, rejecting a corresponding second event item;
meanwhile, when the second issuing mode is indirect issuing, the corresponding second event item is removed;
when second event items needing to be removed in the second event items are all removed, a plurality of third event items are obtained;
extracting a first event in the third event item and taking the first event as a second event;
determining a model base based on a preset degree, and determining a degree determination model corresponding to the reason type;
inputting the second event into a degree determination model to obtain a true degree;
and determining a second difference value between the special degree and the real degree, if the second difference value is within a preset second difference value range, the special reason and the degree are both real, otherwise, the special reason and the degree are not real, and finishing the determination.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset time interval formulation rule specifically comprises the following steps: for example, the time is 15:00, the range is expanded, and the set time interval is 14:30 to 15: 30; the preset region formulation rule is specifically as follows: within 500 meters of the perimeter of the site; the preset retrieval mode library specifically comprises the following steps: a database, in which search modes corresponding to different reason types are stored, for example: the reason type is traffic jam, and the retrieval mode is to retrieve data issued by a traffic third-party platform; the preset feature library specifically comprises: a database, in which the characteristics corresponding to different reason types are stored; the preset first verification value threshold specifically includes: for example, 80; the preset second verification value threshold specifically includes: for example, 85; the preset degree determination model library specifically comprises: a database, in which degree determination models corresponding to different reason types are stored [ models generated after learning a large number of manual degree determination records by using a machine learning algorithm ]; the preset second difference range specifically includes: plus or minus 2;
the time and the place provided by the staff are not necessarily accurate, the range is expanded, and the time area and the area are respectively manufactured; determining a retrieval mode, and performing retrieval in a targeted manner to reduce the retrieval difficulty; when the first event item meeting the condition is searched out based on the searching mode, determining a second event item corresponding to the reason type; prepare to determine from the second event item whether the specific cause and degree of specificity are authentic, cull indirect publications [ for example: the workload of the second event items carried by other platforms is reduced, the second event items with verification values larger than or equal to the threshold of the verification values are removed, and the problem of misjudgment caused by the fact that the second event items with low reliability are brought into judgment is avoided; when the second event items needing to be removed are all removed, the second event items are set to be real due to special reasons, the degree is determined based on the model, and if the degree has small error with the special degree provided by the staff, the special degree is also set to be real; the additional information provided by the staff is subjected to authenticity verification, so that the staff is prevented from working falsely, and meanwhile, the system is more intelligent.
The embodiment of the invention provides a system for automatically reimbursing travel expenses based on bill identification and configuration engine, wherein a verification module executes the following operations:
determining a plurality of first guarantors corresponding to the second publisher and a first guaranty mode between the second publisher and the first guarantor based on a preset guarantor library, where the first guaranty mode includes: active and passive warranties;
when the first guarantee mode is active guarantee, the corresponding first guarantee person is used as a second guarantee person;
when the first guarantee mode is a passive guarantee mode, the corresponding first guarantee person is used as a third guarantee person;
respectively acquiring a first malicious value of the second publisher, a second malicious value of the second guarantor and a third malicious value of the third guarantor, and calculating the first verification value based on the first malicious value, the second malicious value and the third malicious value, wherein a calculation formula is as follows:
where ρ is1Is the first verification value, σ1Is the first malicious value, σ2,iSaid second maliciousness value, σ, for the ith said second guarantor3,iSaid third malicious value, n, for the ith said third guarantor1Is the total number of said second guarantor, n2Is the third oneTotal number of guarantors, [ theta ] is a predetermined constant, [ mu ]1、μ2And mu3Is a preset weight value, mu1>μ2>μ3,ε1,i、ε2,i、ε3,iAnd n3Is an intermediate variable, and is and, else is other, σ1,0Is a preset first malicious value threshold, sigma2,0The second malicious value threshold is preset;
determining a plurality of first guarantee platforms corresponding to the second issuing platform and a second guarantee mode between the second issuing platform and the first guarantee platform based on a preset guarantee platform library, wherein the second guarantee mode comprises the following steps: active and passive warranties;
when the second guarantee mode is active guarantee, the corresponding first guarantee platform is used as a second guarantee platform;
when the second guarantee mode is a passive guarantee mode, the corresponding first guarantee platform is used as a third guarantee platform;
respectively obtaining a fourth malicious value of the second release platform, a fifth malicious value of the second guarantee platform and a sixth malicious value of the third guarantee platform, and calculating the first verification value based on the fourth malicious value, the fifth malicious value and the sixth malicious value, wherein a calculation formula is as follows:
where ρ is2Is the second verification value, σ4Is the fourth malicious value, σ5,iFor the fifth malicious value, σ, of the ith said second vouching platform6,iSaid sixth malicious value, n, for the ith said third vouching platform4For the total number of said second vouching platforms, n5Theta is a preset constant mu for the total number of the third guarantee platforms4、μ5And mu6Is a preset weight value, mu4>μ5>μ6,ε4,i、ε5,i、ε6,iAnd n6Is an intermediate variable, and is and, else is other, σ3,0Is a preset third malicious value threshold, sigma4,0Is a preset fourth malicious value threshold.
The working principle and the beneficial effects of the technical scheme are as follows:
the preset guarantor library specifically comprises: a database, in which the corresponding guarantors of different users are stored, and the guaranty relationship between different users and guaranties is also stored, the guaranty relationship is divided into active guaranty [ a user sets a user as his or her own guarantor, and the other party agrees ] and passive guaranty [ two users belong to the same group, for example: company, home, etc.); the same principle is used for a preset guarantee platform library; the preset constants are specifically: for example, 2; the preset first malicious value threshold specifically includes: for example, 97; the preset second malicious value threshold specifically includes: for example, 98; the preset third malicious value threshold specifically includes: for example, 95; the preset fourth malicious value threshold specifically is: for example, 98;
when malicious values of a user, a guarantor, a publishing platform, and a guaranty platform are obtained, malicious events that occur on the user or the platform [ for example: issue false data, etc. ]; the verification value is calculated based on different malicious values, so that the working efficiency of the system is greatly improved; according to the embodiment of the invention, a guarantee system is adopted, and under the trend of data sharing, each user establishes a guarantee relationship to generate constraint force, so that the cost and the cost of generating malicious events by the user are increased, and the occurrence rate of the malicious events is reduced; meanwhile, the active guarantee and the passive guarantee are analyzed separately, and the method is more reasonable.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.