CN107393212B - Multi-currency service system and method based on machine learning and distributed architecture - Google Patents
Multi-currency service system and method based on machine learning and distributed architecture Download PDFInfo
- Publication number
- CN107393212B CN107393212B CN201710486545.1A CN201710486545A CN107393212B CN 107393212 B CN107393212 B CN 107393212B CN 201710486545 A CN201710486545 A CN 201710486545A CN 107393212 B CN107393212 B CN 107393212B
- Authority
- CN
- China
- Prior art keywords
- currency
- machine
- index information
- management system
- registered user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F19/00—Complete banking systems; Coded card-freed arrangements adapted for dispensing or receiving monies or the like and posting such transactions to existing accounts, e.g. automatic teller machines
- G07F19/20—Automatic teller machines [ATMs]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F19/00—Complete banking systems; Coded card-freed arrangements adapted for dispensing or receiving monies or the like and posting such transactions to existing accounts, e.g. automatic teller machines
- G07F19/20—Automatic teller machines [ATMs]
- G07F19/211—Software architecture within ATMs or in relation to the ATM network
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a multi-currency service system and a method based on machine learning and distributed architecture, wherein the multi-currency service system comprises a cash recycling machine network and a management system, the cash recycling machine network comprises a plurality of all-in-one machines, the plurality of all-in-one machines are connected with the management system through a network, and the all-in-one machines support the deposit of multi-currency cash and the extraction of the multi-currency cash; the all-in-one machine is used for acquiring a plurality of index information of the transaction date and updating the data of the index information to the management system; the plurality of index information includes an initial stock index information, a cash-in index information, a cash-out index information, and a refill index information; the management system is used for conducting machine learning according to historical data of the index information to determine the optimal currency configuration of the all-in-one machine of the next transaction date. According to the embodiment of the invention, the user realizes multi-currency exchange through the all-in-one machine, the capacity of multi-currency real-time circulation requirements of the multi-cash recycling and recycling all-in-one machine network is improved, and the user experience is improved.
Description
Technical Field
The invention relates to the technical field of multi-currency self-service exchange, in particular to a multi-currency service system and a method based on machine learning and a distributed architecture.
Background
Nowadays, with the global mobility enhancement, the foreign exchange demand of people is expanded, and besides frequent travelers, foreign students without foreign bank accounts, commercial immigration and immigration workers without foreign bank accounts are still the most important drivers of the demand of money service. Briefly, the transaction for foreign exchange redemption includes two parts: some people want to exchange foreign currency with their own currency and some people want to sell their foreign currency in exchange for the own currency, and without a third party, the transactions of both people cannot be handled simply at the same time, except for face-to-face exchanges. Thus, forever, the foreign exchange transactions have been satisfied by storing different currencies through trusted, risk-reducing intermediaries, such as currency service operators, to satisfy the common currency needs of the seller and seller.
In current practice, currency conversion operations occur mostly at airports and are operated by global currency conversion operators to small currency conversion merchants scattered at street corners. Between high interest and other commission fees, users almost always have to pay more for the convenience of redemption. Thus, some customers turn to trust a traditional bank as an option to make a redemption. Most customers expect multiple currencies to be always sufficient. However, due to high cost and settlement risk, it is not practical to maintain currency inventory levels for all branches to meet real-time circulation needs. Alternatively, banks may offer money reservation services to their key customers only to avoid potential liability violations. But often times such transactions are delayed for days when the customer needs to return to a particular branch for cash withdrawal within a limited working time.
Disclosure of Invention
The embodiment of the invention aims to provide a multi-currency service system and method based on machine learning and distributed architecture, which can realize near-real-time optimal currency configuration and distribution of multiple currencies in an all-in-one machine so as to improve the capacity of meeting the real-time circulation requirement of the multiple currencies of a cash recycling system network and improve the user experience.
In order to achieve the above object, an embodiment of the present invention provides a multi-currency service system based on machine learning and a distributed architecture, including: the system comprises a cash recycling system, a cash recycling system and a management system, wherein the cash recycling system comprises a plurality of all-in-one machines, the all-in-one machines are connected with the management system through a network, and the all-in-one machines support the deposit of multi-currency cash and the withdrawal of the multi-currency cash;
the all-in-one machine is used for acquiring a plurality of index information of transaction dates and updating data of the index information to the management system; the plurality of index information includes an initial inventory index information including the number of banknotes of each denomination of each currency initially existing in the integrated machine on a transaction date, a withdrawal cash index information including the number of banknotes of each denomination of each currency deposited in the integrated machine on a transaction date, and a refill index information including the number of banknotes of each denomination of each currency refilled in the integrated machine on a transaction date;
the management system is used for conducting machine learning according to the historical data of the index information to determine the optimal currency configuration of the all-in-one machine on the next transaction date.
Compared with the prior art, the multi-currency service system based on the machine learning and distributed architecture disclosed by the invention can be used for solving the problem of low capacity of meeting the circulation requirement of multiple currencies in real time in the prior art by calculating the optimal currency configuration technical scheme of the all-in-one machine of the cash recycling system based on a plurality of index information and historical data of the all-in-one machine in the cash recycling system network, and has the beneficial effect of greatly improving the capacity of meeting the circulation requirement of the multiple currencies in the cash recycling system network in real time by configuring the optimal currency of the all-in-one machine of the cash recycling system network.
Preferably, the multi-currency service system further comprises a mobile terminal, the mobile terminal is in network connection with the management system, a currency service APP is loaded on the mobile terminal, and the mobile terminal is in network connection with the cash recycling system through the management system;
the mobile terminal is used for generating reserved exchange information according to the operation of a registered user on a currency service APP and sending the reserved exchange information and the current position information of the mobile terminal to the management system, wherein the reserved exchange information comprises currencies to be exchanged and the number of denominations of each currency;
the all-in-one machine is used for distributing the currency to be exchanged to the registered user according to the reserved exchange information.
Preferably, the management system is used for selecting a kiosk which is closest to the position of the mobile terminal from the cash recycling system network according to the position information of the mobile terminal, sending the reserved exchange information to the kiosk, and sending the position information of the kiosk to the mobile terminal which is logged in with the currency service APP of the registered user;
and the all-in-one machine closest to the position of the mobile terminal is used for distributing the currency to be exchanged to the registered user according to the reserved exchange information.
Preferably, the management system is further configured to determine potential redemption requirements of the registered user based on the location information of the registered user's mobile terminal and the registered user's historical transaction data.
Preferably, the mobile terminal is further configured to upload the login time and the interest area of the registered user on the currency service APP to the management system, and the management system is further configured to determine the potential exchange demand of the registered user according to the login time and the interest area of the registered user on the currency service APP.
Preferably, the management system is further configured to determine currency configuration and distribution requirements of the kiosk according to the potential redemption requirements of the registered user and historical data of the plurality of index information.
Preferably, the all-in-one machine comprises a coin change dispenser, wherein the coin change dispenser is used for supporting change exchange of coins with different denominations of different currencies;
the integrated machine is used for converting change to be converted into digital currency and recharging the digital currency to an account of a registered user when no coin exists in the coin change machine.
Preferably, the kiosk is adapted to top up the converted digital currency to an electronic wallet account or points account of the registered user's third party stored value payment instrument.
The embodiment of the invention also provides a method which is realized by adopting the multi-currency service system based on the machine learning and distributed architecture and comprises the following steps:
the all-in-one machine acquires a plurality of index information of the transaction date and updates the data of the index information to the management system; the plurality of index information includes an initial inventory index information including the number of banknotes of each denomination of each currency initially existing in the integrated machine on a transaction date, a withdrawal cash index information including the number of banknotes of each denomination of each currency deposited in the integrated machine on a transaction date, and a refill index information including the number of banknotes of each denomination of each currency refilled in the integrated machine on a transaction date;
and the management system performs machine learning according to the historical data of the index information to determine the optimal currency configuration of the all-in-one machine on the next transaction date.
Compared with the prior art, the method disclosed by the invention has the advantages that the problem of low capability of meeting the circulation requirement of multiple currencies in real time in the prior art is solved by calculating the optimal currency configuration technical scheme of the all-in-one machine of the next transaction date according to the multiple index information and the historical data of the all-in-one machine in the cash recycling system network, and the beneficial effect that the capability of meeting the circulation requirement of multiple currencies in real time can be greatly improved by configuring the optimal currency of the all-in-one machine of the cash recycling system network is obtained.
Preferably, in the step of performing machine learning by the management system according to the historical data of the index information to determine the optimal currency configuration of the all-in-one machine on the next transaction date, the method includes:
the management system is used for determining the potential exchange requirement of a registered user according to the login time and the interest area of the registered user on a currency service APP, the position information of the mobile terminal of the registered user and the historical transaction data of the registered user, and determining the currency type configuration and distribution requirement of the all-in-one machine according to the potential exchange requirement of the registered user and the historical data of the plurality of index information.
Drawings
Fig. 1 is a schematic structural diagram of a multi-currency service system according to an embodiment of the present invention.
FIG. 2 is a schematic structural diagram of a recycling system of a multi-currency service system according to an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of a machine learning engine of a multi-currency service system according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of an application function structure of a mobile terminal of a multi-currency service system according to an embodiment of the present invention.
Fig. 5 is a schematic mechanism diagram of an all-in-one machine of a multi-currency server in the embodiment of the invention.
Fig. 6 is a flow chart of a method of an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1 and 2, a multi-currency service system based on a machine learning and distributed architecture according to an embodiment of the present invention includes a cash recycling system, a mobile terminal, and a management system. The cash recycling system comprises a cash recycling system and a cash recycling system, wherein the cash recycling system comprises a plurality of all-in-one machines, the mobile terminal and the management system are connected through the Internet, and the all-in-one machines support the deposit of multi-currency cash and the withdrawal of the multi-currency cash. The mobile terminal is loaded with a currency service APP, and is in network connection with the cash recycling system through the management system.
The all-in-one machine is used for acquiring a plurality of index information of transaction dates and updating data of the index information to the management system; the plurality of index information includes an initial inventory index information including the number of banknotes of each denomination of each currency initially existing in the integrated machine on a transaction date, a withdrawal cash index information including the number of banknotes of each denomination of each currency deposited in the integrated machine on a transaction date, and a refill index information including the number of banknotes of each denomination of each currency refilled in the integrated machine on a transaction date.
The management system is used for conducting machine learning according to the historical data of the index information to determine the optimal currency configuration of the all-in-one machine on the next transaction date.
According to the embodiment of the invention, the capacity of meeting the circulation requirement of multiple currencies of the cash recycling system network in real time is greatly improved by configuring the optimal currencies of the all-in-one machine of the cash recycling system network, namely, the capacity of matching the currency configuration distribution and the user requirement in each all-in-one machine is greatly improved.
Specifically, the cash recycling system network can be connected with a currency service APP on a mobile terminal of a registered user through the network, so that the user can exchange currency. The kiosk is typically placed in close proximity to where the user has frequent currency exchange needs. For example, the foreign student or the business immigration who needs to exchange the home currency into the local currency for the payment with the local cash, or the foreign traveler who needs to exchange the foreign currency into the local currency, etc. Those locations may be university campuses, travel agencies, train stations, and airports. Currency conversion is just one function of the kiosk of various embodiments, and the kiosk may convert input currency to output currency in cash. The multi-currency service system generates revenue by charging a user a service fee, which may be charged by reserving a portion of the currency provided to the user. In some embodiments, the user may also withdraw cash from a kiosk that includes a similar manner of operation as an ATM (automatic Teller machine) by an account-to-cash process. The user may be required to authenticate or identify himself by means of a card account number or PIN (personal identification number) before the user can obtain the service.
The kiosk may also store or withdraw cash in multiple currencies based on cash-to-cash, cash-to-account, account-to-cash transactions.
In the embodiment of the invention, the transaction initiated by the currency service APP on the mobile terminal and further confirmed by the all-in-one machine is mediated by the machine learning engine, and the machine learning engine is communicated with the management system at the back end through a wired or wireless secure broadband internet. When the management system collects large amounts of data from the internet or social media, predictive analysis can be made through machine learning to estimate travel trends and monetary demands. The management system comprises a back-end management server, a membership server and a transaction server.
The back-end management server comprises a plurality of modules which are used for processing remote monitoring, data exchange and transaction management between the cash recycling system network and the currency service APP client. Monitoring state, running state and event notification of the monitoring all-in-one machine are used for sending information about transaction state to registered users and sending warning information about system maintenance problems to managers. The membership server is used to manage user registration and authentication management. During the registration process, the user can be identified by providing certain specific features, in particular the payment details and the mobile phone SIM number of the user, and other personal information of the user such as gender, age group, occupation, address and/or hobby, etc. can also be submitted. The latest update of the money service transaction should be notified to the registered user by a short message or an email.
In embodiments of the present invention, the management system may also detect malicious events and potential fraud in advance as a preventative security measure. This is supported by the big data acquisition and analysis flow of the machine learning engine. The management system is used for demand management by predicting the cash demand of a user to make appropriate currency configuration and dispensing. After the transaction is over, the transaction server sends and receives real-time transactions based on rules and steps relating to cash transfer data established for final settlement that the participants agree to exchange. The step of transaction redemption includes the steps of: transaction pairing, transaction ordering, data integrity checking, and payment data aggregation for communication. These back-end management modules are connected in sequence to a variety of application programming interfaces with third party digital processors including electronic wallet processors, payment processors and cryptocurrency processors, respectively.
In embodiments of the present invention, different currency combinations may be configured and distributed in different kiosks in a quick and efficient manner. The multi-currency service system of the embodiment of the invention is based on distributed network nodes, and the nodes are all-in-one machines. A distributed architecture refers to the connection of processing units or storage devices through a network, rather than being uniformly connected to a single, common processing system. The distributed architecture of the embodiment of the invention is composed of distributed network nodes, a consensus mechanism is adopted to ensure fault-tolerant communication and concurrent control is adopted to ensure accountability and non-repudiation of transactions, each all-in-one machine is connected with the distributed network nodes and communicates with the currency level with four pieces of index information, the initial stock index information, the cash deposit index information, the cash withdrawal index information and the refill index information are dynamically updated to the network for other all-in-one machines to access, and a manager can establish a currency distribution plan through the updated data and appoint to transmit the currency to each all-in-one machine node in a predetermined or temporary mode.
Specifically, as shown in fig. 2, the automated teller machine network includes a plurality of machines 1, 2, 3, etc., each of which includes a plurality of denominations of money of a plurality of denominations, i.e., includes money 1, 2, 3, 4, etc., the money 1 may be a dollar, the money 2 may be a renminbi, the money 3 may be a hong Kong, and the money 4 may be a yen, and each of which includes a plurality of denominations of dollars, a renminbi, a hong Kong, and a yen, and a user may exchange money for, for example, the renminbi and the hong Kong at the same time through the dollars, so that, for each of the machines, four status indicators are involved for each of the denominations of money, the initial stock of money, the cash deposit, the cash withdrawal, and the refill of each denomination. Therefore, by acquiring the beginning stock index information of a plurality of denominations of each currency of the transaction date of the all-in-one machine, depositing the cash index information, taking out the cash index information and refilling the index information, machine learning is performed according to the historical data of the plurality of index information to determine the optimal currency configuration of the all-in-one machine of the next transaction date. Thus, user experience is improved.
For a currency management system comprising a core process of user authentication, exchange transaction, currency configuration, currency distribution and user notification, the multi-currency service system has the following advantages:
1. currency circulation of different all-in-one machine nodes is fully improved. By configuring and dispensing different currencies in time, each all-in-one node has the best currency configuration for the user to use, and based on the machine learning engine, the currency demand can be predicted to a more accurate degree without the need to passively wait for the user's demand.
2. The money transfer fee can be designed in a more cost-effective manner. The more systematic and unquestionably less associated costs occur the more the way money transport is managed due to the inevitable need for service of the securicar. Therefore, the cost is reduced, and the user experience is improved.
In the embodiment of the present invention, as shown in fig. 3, the idea of machine learning is to perform an automatic modeling process for state estimation and pattern determination in historical data and future data through a process of data arrangement, feature extraction state estimation, and pattern prediction. There are several different machine learning algorithms for building the learning-based prediction module, and two widely adopted machine learning methods are supervised learning and unsupervised learning. The multi-currency service system generates a large amount of data through various related information and channels of transaction. From currency conversion data from a transactional kiosk, exchange rate inspection data from a currency service APP, travel preference data from social media, airline hotel reservation data from partners, to data from other means with similar profiles, machine learning is to develop analytical techniques to develop insights and discovery patterns and make key predictions of potential needs of customers by utilizing the large amount of structural and non-structural, real-time and historical data described above.
As shown in fig. 4, in the embodiment of the present invention, the money service APP loaded on the mobile terminal may be used by the user to perform money service transactions of checking exchange rate, reserving foreign exchange transactions, confirming orders, checking order history information, comparing exchange rate, tracking flow, finding a kiosk, checking inventory, exchanging points, and purchasing cryptocurrency. The user needs to register with the multi-currency service system to gain access to the relevant transaction functions described above. This requires compliance with the real-name program required by regulatory authorities. The purpose is to monitor the customers' transactions against their expected behavior and registry files to identify and block any suspicious transactions such as money laundering or malicious attacks.
The registered user is authorized to use the monetary service APP to obtain the above mentioned monetary service with usage data and activity logs for analysis purposes. Non-registered users may access some of the non-transaction functions described above, such as rate queries and non-priority subscription services.
The mobile terminal can be used for generating reservation exchange information according to the operation of a registered user on the currency APP, and sending the reservation exchange information and the position information of the mobile terminal to the management system, wherein the reservation exchange information comprises currencies to be exchanged and the number of denominations of each currency. The management system is used for selecting an all-in-one machine closest to the position of the mobile terminal from the cash recycling system network according to the position information of the mobile terminal, sending the reserved exchange information to the all-in-one machine, and sending the position information of the all-in-one machine to the mobile terminal logged in with the currency service APP of the registered user.
The all-in-one machine closest to the position of the mobile terminal is used for distributing the currency to be exchanged to the registered user according to the reserved exchange information.
The management system is further used for determining potential redemption requirements of the registered user according to the position information of the mobile terminal of the registered user and the historical transaction data of the registered user. The mobile terminal is further used for uploading the login time and the interest area of the registered user on the currency service APP to the management system, and the management system is further used for determining the potential exchange requirement of the registered user according to the login time and the interest area of the registered user on the currency service APP. The management system is further used for determining currency configuration and distribution requirements of the all-in-one machine according to potential exchange requirements of the registered users and historical data of the index information.
Specifically, the activity log and the area of interest for each login time of the registered user will be recorded for analysis purposes. The mobile terminal comprises a smart phone, a palm computer, a tablet computer, a notebook computer, and an iOS-based or Android-based device, and operates on GPRS (general Packet Radio service), CDMA (code division multiple access), HSDPA (high Speed Downlink Packet access), LTE (Long term evolution) and/or other Wi-Fi (Wireless Fidelity) networks, and provides a data network for connecting and communicating the mobile terminal and the management system. The mobile terminal comprises hardware and software for positioning, wherein the hardware and software for positioning adopt an indoor positioning technology or an outdoor positioning technology, the outdoor positioning technology comprises a GPS technology, and the outdoor positioning technology comprises a cellular base station positioning technology or a distance sensing technology such as ultrasonic detection or low-power Bluetooth beacon communication.
The kiosk may automatically identify the location of a registered user having a mobile terminal in proximity thereto and process a redemption transaction or a provisional transaction of the proximate user with a predetermined request. In addition, the current location of a registered user may also be tracked as the registered user travels to a different country. The data can be aggregated and updated to the servers of the management system for analysis, which is very useful for understanding the travel patterns and monetary needs of the customers.
As shown in fig. 5, the embodiment of the present invention further provides a kiosk, which, in addition to the main functions of supporting cash storage and cash withdrawal, also includes a plurality of functions of supporting banknote validation, multi-currency identification, multi-currency escrow, and cash recycling. The all-in-one machine comprises a main processor, a touch screen, an identity reader, a receipt printer, a high-definition camera, a bank card reader, a stored value card reader, a card picker, a code keyboard, a paper currency receiver, a paper currency extractor, a cash deposit box, a cash withdrawal box and a coin change-making device. The main processor, the touch screen, the identity reader, the receipt printer and the high-definition camera are all arranged in an independent shell adopting strengthening measures so as to prevent damage and theft. Some form of double security door lock and burglar alarm arrangement needs to be implemented to protect the security of the entire unit. In addition, the bank card reader and the password keyboard can process bank card payment. Other forms of prepaid or loyalty cards may be processed contactlessly by stored value card readers. The card taker may be used to support the issuance of bank or membership cards. The banknote acceptor and the banknote extractor both include a banknote validating function, and the banknote acceptor includes a cash deposit box for storing deposited cash, and the cash withdrawal box is for outputting banknotes of different denominations of different currencies for withdrawal by a customer. When the cash deposit box or the cash withdrawal box is separated from the currency detecting device, the cash deposit box or the cash withdrawal box is written into a built-in chip of the all-in-one machine so as to perform better background cash management. In the case of banknote storage, the multi-escrow container may hold some banknotes and allow the user to agree to reject the banknotes to end the transaction.
The coin change dispenser supports change exchange for different denominations of coins of different currencies. The integrated machine is used for converting change to be converted into digital currency and recharging the digital currency to an account of a registered user when no coin exists in the coin change machine. The all-in-one machine is used for charging the converted digital currency to an electronic wallet account or a point account of a third-party stored-value payment tool of the registered user.
In embodiments of the invention where the kiosk determines the output currency based on the user profile of the input currency, exchange rate and other conditions, the output banknotes will typically have different denominations and coins will inevitably be part of the change of transaction. When the all-in-one machine receives the exchange request, the change can be exchanged through the coin change machine. The out-of-service is not desirable to the user when there is no coin in the coin changer, and therefore, the best approach is to be able to exchange change to be exchanged for digital money to be charged to the account of the registered user, for example, the exchanged digital money can be charged to the e-wallet account or points account of the registered user's third party stored value payment instrument.
In summary, the multi-currency service system based on machine learning and distributed architecture in the embodiment of the present invention can support exchange of different currencies, perform optimal currency configuration on an all-in-one machine, improve the ability to meet the circulation demand of the multi-currency, meet the user demand, and improve the user experience.
As shown in fig. 6, an embodiment of the present invention further provides a method implemented by using the multi-currency service system based on machine learning and distributed architecture, including the following steps:
601, the all-in-one machine acquires a plurality of index information of a transaction date and updates the data of the index information to the management system; the plurality of index information includes an initial inventory index information including the number of banknotes of each denomination of each currency initially existing in the integrated machine on a transaction date, a withdrawal cash index information including the number of banknotes of each denomination of each currency deposited in the integrated machine on a transaction date, and a refill index information including the number of banknotes of each denomination of each currency refilled in the integrated machine on a transaction date;
The method disclosed by the invention can greatly improve the capacity of meeting the circulation requirement of multiple currencies in real time by configuring the optimal currency of the all-in-one machine of the cash recycling system network, and improve the user experience.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (10)
1. A multi-currency service system based on machine learning and distributed architecture, comprising: the system comprises a cash recycling system, a cash recycling system and a management system, wherein the cash recycling system comprises a plurality of all-in-one machines, the all-in-one machines are connected with the management system through a network, and the all-in-one machines support the deposit of multi-currency cash and the withdrawal of the multi-currency cash;
the all-in-one machine is used for acquiring a plurality of index information of transaction dates and updating data of the index information to the management system; the plurality of index information includes an initial inventory index information including the number of banknotes of each denomination of each currency initially existing in the integrated machine on a transaction date, a withdrawal cash index information including the number of banknotes of each denomination of each currency deposited in the integrated machine on a transaction date, and a refill index information including the number of banknotes of each denomination of each currency refilled in the integrated machine on a transaction date;
the management system is used for performing machine learning according to the historical data of the index information to determine the optimal currency configuration of the all-in-one machine on the next transaction date;
wherein the management system collects large amounts of data from the internet or social media, making predictive analytics through machine learning to estimate monetary demand.
2. The multi-currency service system based on machine learning and distributed architecture as claimed in claim 1, further comprising a mobile terminal, wherein the mobile terminal is connected to the management system via a network, the mobile terminal is loaded with a currency service APP, and the mobile terminal is connected to the cash recycling machine via the management system via the network;
the mobile terminal is used for generating reserved exchange information according to the operation of a registered user on a currency service APP and sending the reserved exchange information and the current position information of the mobile terminal to the management system, wherein the reserved exchange information comprises currencies to be exchanged and the number of denominations of each currency;
the all-in-one machine is used for distributing the currency to be exchanged to the registered user according to the reserved exchange information.
3. The multi-currency service system based on the machine learning and distributed architecture as claimed in claim 2, wherein the management system is configured to select a kiosk closest to the location of the mobile terminal from the cash recycling kiosk network according to the location information of the mobile terminal, and to send the reservation exchange information to the kiosk, and to send the location information of the kiosk to the mobile terminal logged in with the currency service APP of the registered user;
and the all-in-one machine closest to the position of the mobile terminal is used for distributing the currency to be exchanged to the registered user according to the reserved exchange information.
4. The multi-currency service system based on machine learning and distributed architecture of claim 3, wherein the management system is further configured to determine potential redemption requirements of the registered user based on location information of the registered user's mobile terminal and historical transaction data of the registered user.
5. The multi-currency service system based on machine learning and distributed architecture as claimed in claim 4, wherein the mobile terminal is further configured to upload the login time and the area of interest of the registered user on the currency service APP to the management system, and the management system is further configured to determine the potential exchange needs of the registered user according to the login time and the area of interest of the registered user on the currency service APP.
6. The multi-currency service system based on machine learning and distributed architecture of claim 5, wherein the management system is further configured to determine the currency configuration and distribution requirements of the kiosk based on the potential redemption requirements of the registered user and historical data of the plurality of metrics.
7. The multi-currency service system based on machine learning and distributed architecture of claim 1, wherein the all-in-one machine comprises a coin changer for supporting change exchanges of different denominations of coins of different currencies;
the integrated machine is used for converting change to be converted into digital currency and recharging the digital currency to an account of a registered user when no coin exists in the coin change machine.
8. The multi-currency service system based on machine learning and distributed architecture of claim 7, wherein the kiosk is adapted to top up converted digital currency to an electronic wallet account or points account of a registered user's third party stored value payment instrument.
9. A multi-currency service method based on machine learning and distributed architecture, which is implemented by the multi-currency service system based on machine learning and distributed architecture according to any one of claims 1 to 8, and comprises the following steps:
the all-in-one machine acquires a plurality of index information of the transaction date and updates the data of the index information to the management system; the plurality of index information includes an initial inventory index information including the number of banknotes of each denomination of each currency initially existing in the integrated machine on a transaction date, a withdrawal cash index information including the number of banknotes of each denomination of each currency deposited in the integrated machine on a transaction date, and a refill index information including the number of banknotes of each denomination of each currency refilled in the integrated machine on a transaction date;
and the management system performs machine learning according to the historical data of the index information to determine the optimal currency configuration of the all-in-one machine on the next transaction date.
10. The multi-currency service method based on machine learning and distributed architecture as claimed in claim 9, wherein in the step of the management system performing machine learning based on historical data of the plurality of metric information to determine the best currency configuration of the kiosk for the next transaction date, comprising:
the management system is used for determining the potential exchange requirement of a registered user according to the login time and the interest area of the registered user on a currency service APP, the position information of the mobile terminal of the registered user and the historical transaction data of the registered user, and determining the currency type configuration and distribution requirement of the all-in-one machine according to the potential exchange requirement of the registered user and the historical data of the plurality of index information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710486545.1A CN107393212B (en) | 2017-06-23 | 2017-06-23 | Multi-currency service system and method based on machine learning and distributed architecture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710486545.1A CN107393212B (en) | 2017-06-23 | 2017-06-23 | Multi-currency service system and method based on machine learning and distributed architecture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107393212A CN107393212A (en) | 2017-11-24 |
| CN107393212B true CN107393212B (en) | 2020-03-03 |
Family
ID=60332001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710486545.1A Active CN107393212B (en) | 2017-06-23 | 2017-06-23 | Multi-currency service system and method based on machine learning and distributed architecture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107393212B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108648077B (en) * | 2018-04-24 | 2021-04-13 | 东莞市波动赢机器人科技有限公司 | Information digitalization processing method, electronic device and computer storage medium |
| CN109919587A (en) * | 2019-02-26 | 2019-06-21 | 神州融安科技（北京）有限公司 | A kind of system and method for supporting to exchange between digital cash and commodity money |
| CN110415462B (en) * | 2019-07-31 | 2021-09-03 | 中国工商银行股份有限公司 | ATM equipment banknote adding optimization method and device |
| CN113422773A (en) * | 2021-06-22 | 2021-09-21 | 上海和数软件有限公司 | Block chain digital asset exchange machine |
| CN114677816B (en) * | 2021-08-02 | 2023-09-08 | 中国人民银行数字货币研究所 | Digital currency deposit and withdrawal machine |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101023453A (en) * | 2004-09-22 | 2007-08-22 | 菲多斯有限公司 | Methods and apparatus for facilitating a currency exchange transaction |
| CN104809826A (en) * | 2015-04-30 | 2015-07-29 | 深圳市怡化时代科技有限公司 | Cash distribution method for financial self-service equipment |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9430918B2 (en) * | 2014-09-23 | 2016-08-30 | Moneygram International, Inc. | Receipt generation service |
| CN104899985B (en) * | 2015-04-30 | 2017-08-29 | 深圳市怡化时代科技有限公司 | A kind of self-service method for processing business and system |
| CN105741440B (en) * | 2015-07-22 | 2018-11-09 | 深圳怡化电脑股份有限公司 | Banknote currency type exchanges self-service device operation method and device |
| CN105447685A (en) * | 2015-12-22 | 2016-03-30 | 王义海 | Paper money, coin and cash service device and method of internet payment (financial) account |
| CN105844813A (en) * | 2016-03-17 | 2016-08-10 | 中国工商银行股份有限公司 | Self-service multi-currency exchange device and system |
| CN106384433B (en) * | 2016-09-28 | 2019-04-30 | 北京映翰通网络技术股份有限公司 | Commodity distribution management system and method for automatic vending machine |
-
2017
- 2017-06-23 CN CN201710486545.1A patent/CN107393212B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101023453A (en) * | 2004-09-22 | 2007-08-22 | 菲多斯有限公司 | Methods and apparatus for facilitating a currency exchange transaction |
| CN104809826A (en) * | 2015-04-30 | 2015-07-29 | 深圳市怡化时代科技有限公司 | Cash distribution method for financial self-service equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107393212A (en) | 2017-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107393212B (en) | Multi-currency service system and method based on machine learning and distributed architecture | |
| US9996985B2 (en) | Distribution and enablement of reloadable prepaid cards in transit | |
| AU2010271244B2 (en) | Predictive techniques in transit alerting | |
| US10803454B2 (en) | ATM network resource exchange system | |
| AU2010271245B2 (en) | Reloadable prepaid card distribution, reload, and registration in transit | |
| US20110165836A1 (en) | Id application for nfc phone | |
| US8096398B2 (en) | Efficient movement and storage of funds | |
| CN109344987A (en) | Foreign currency reserves enchashment method, client, server, the pre- Ark of the Covenant of foreign currency and system | |
| CN109409548A (en) | Foreign currency reserves enchashment method, client, server, the pre- Ark of the Covenant of foreign currency and system | |
| US20240135769A1 (en) | Universal removeable cassette array with cash rebalancing | |
| US20230386286A1 (en) | Universal built-in cassette array with cash rebalancing | |
| KR20220046009A (en) | Method and system for providing foreign currency exchange services based on net exchange demand between countries | |
| US11373476B1 (en) | Systems and methods for cash-till optimization | |
| US11935369B2 (en) | Universal cassette for recirculation of bills in multiple systems | |
| US11663874B1 (en) | Mixed denomination cash cassette currency counter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |