CN108121824A - A kind of chat robots and system towards financial service - Google Patents
A kind of chat robots and system towards financial service Download PDFInfo
- Publication number
- CN108121824A CN108121824A CN201810030096.4A CN201810030096A CN108121824A CN 108121824 A CN108121824 A CN 108121824A CN 201810030096 A CN201810030096 A CN 201810030096A CN 108121824 A CN108121824 A CN 108121824A
- Authority
- CN
- China
- Prior art keywords
- user
- financial
- module
- data
- human
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0281—Customer communication at a business location, e.g. providing product or service information, consulting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of chat robots and system towards financial service, the finance data acquisition module acquisition finance data of the chat robots, and it is sent to financial base module and carries out classification storage;Human-computer interaction module receives financial business consultation information input by user and carries out response;Financial business processing module carries out business scenario classification according to financial business consultation information, and extracts corresponding human-computer interaction strategy in financial base module according to sorted business scenario and be pushed to the human-computer interaction module progress human-computer dialogue;User draws a portrait module according to human-computer dialogue contents extraction user characteristics to build user's portrait;Order module picks out the financial product generation order for meeting user's portrait, and carries out tracing management to the order after user's confirmation.Technical scheme completes the element tasks such as user data acquisition, answer, order outbound by way of human-computer interaction, can promote the operational efficiency of financial service and financial service experience.
Description
Technical field
The present invention relates to financial industry intelligent online service technology fields, and in particular to a kind of chat towards financial service
Robot and system.
Background technology
In general, internet financial business is divided into finance interconnection networking, mobile payment and Third-party payment, P2P nets
Network is provided a loan and crowd raises the different types such as financing.Internet finance occurred in China since 1997, with big data, cloud meter
Continuing to bring out for the technologies such as calculation, block chain, artificial intelligence, has showed " blowout ", explosive growth since 2013.Internet
By the advantage in terms of enterprise schema and information processing capability, open up a market finance chance outside existing channel, reduces friendship
Easy cost, alleviates problem of information asymmetry to a certain extent, improves Efficiency of Capital Allocation, and social and economic activities are generated
Different influences.
In recent years, internet finance rapid development, internet finance reduce transaction cost, can alleviate information asymmetry and ask
Topic improves the allocative efficiency of fund.As country graduallyes relax control the threshold trip that enters of financial industry, many finance of rising in recent years are public
Department, these financing corporations release various diversified credit financing products, finance product and investment product, effectively meet huge
Grassroots's use of funds demand compensates for the deficiency of conventional silver industry.
But with financial product variation, these financing corporations need to accept more and more financial consultation services, manually
Mode, which has, seeks advice from efficient, the service experience of quicklook, and shortcoming is:Human cost height, information asymmetry, professional degree are deposited
Difference, product is deficient the shortcomings of.Manually input is big, and since the professional ability of personnel and mechanism is uneven, has seriously affected use
Family service experience.Lack more effective user data acquisition method simultaneously, largely realize signing and examination & approval with manpower, cause to service
The rising of the beneath and all kinds of costs of efficiency.
The content of the invention
In view of this, it is an object of the invention to overcome the deficiencies of the prior art and provide a kind of chatting towards financial service
Its robot and system complete the basic work such as user data acquisition, answer, order outbound by way of human-computer interaction
Make, experienced with promoting the operational efficiency of financial service and financial service.
In order to achieve the above object, the present invention adopts the following technical scheme that:
A kind of chat robots towards financial service carry out information exchange with a background server, including:Finance data
Acquisition module, financial base module, human-computer interaction module, financial business processing module, user's portrait module and order module,
Wherein,
The finance data acquisition module, for gathering the finance data in internet and the background server, concurrently
It gives the financial base module and carries out classification storage;
The human-computer interaction module, for receiving financial business consultation information input by user and carrying out response;The people
Machine interactive interface is equipped with Text Entry, phonetic entry button and multimedia file send key;
The financial business processing module, for carrying out business scenario point according to financial business consultation information input by user
Class, and corresponding human-computer interaction strategy is extracted in the financial base module according to sorted business scenario and is pushed to institute
Human-computer interaction module is stated, so that the human-computer interaction module carries out human-computer dialogue according to the human-computer interaction strategy;
User's portrait module, for building user's portrait according to human-computer dialogue contents extraction user characteristics;
The order module, for picking out the financial product life for meeting user's portrait from the financial base module
Tracing management is carried out into order, and to the order after user's confirmation.
Preferably, the financial business processing module, is specifically used for:
Semantic analysis is carried out to financial business consultation information input by user, to extract keyword;
According to the keyword, the business scenario that user seeks advice from is judged;
Corresponding business scenario has been searched whether in the financial base module;
It is pushed to if so, extracting the corresponding human-computer interaction strategy of the business scenario from the financial base module
The human-computer interaction module, so that the human-computer interaction module carries out human-computer dialogue according to the human-computer interaction strategy.
Preferably, it is described that semantic analysis is carried out to financial business consultation information input by user, to extract keyword, specifically
For:
After carrying out data prediction to financial business consultation information input by user, the image completed, voice number will be handled
According to text data is changed into, to extract keyword;Wherein, the data prediction includes:Data parsing, data cleansing, data
Extraction, data conversion, data loading.
Preferably, user's portrait module, for that can reflect user from the extracting data after the semantic analysis
Demand, the user characteristic data of credit information are drawn a portrait with building user.
Preferably, the financial business processing module is additionally operable to the user characteristics number of user portrait module output
According to being input in the financial product Matching Model to prestore, if matching result reaches preset value, the order module is controlled from described
The financial product generation order for meeting user's portrait is picked out in financial base module.
Preferably, the financial business processing module builds financial product Matching Model, and profit using machine learning algorithm
Algorithms of different is sought with classification, cluster, correlation rule, the regression analysis in machine learning algorithm, calculates user characteristic data and gold
Melt the matching degree of product Matching Model.
Preferably, the order module includes:
Order application module, for the order application for receiving the financial business application material of user's submission and making a report on;
Air control approval module, for auditing the financial business application material of user's submission, and according to the financial circles after examination & verification
Business application material and order application carry out user credit assessment and payment loan risk assessment, and assessment result is input to and is prestored
Risk evaluation model in, obtain the feature of risk parameter of user;
Module of making loans is paid, it, should by the human-computer interaction module the output phase if the feature of risk parameter is more than threshold value
Response masterplate, otherwise, complete on-line payment and make loans operation;
Order tracing management module, for tracking the refund situation of the order after making loans with management and control payment.
Preferably, machine learning algorithm of the air control approval module based on recurrence, classification, cluster, decision tree etc is built
The risk evaluation model is found, realizes the online examination & approval to user's order.
Preferably, the finance data acquisition module, is specifically used for:
The finance in internet and the background server is gathered by manual entry, reptile instrument, online transmission mode
Data;The finance data includes:Financial common sense, product data, legal knowledge, risk knowledge, mechanism public sentiment;
Data cleansing, data integration, data conversion and hough transformation pretreatment are carried out to the finance data of acquisition;
Pretreated finance data is sent to the financial base module and carries out classification storage.
A kind of chat robots system towards financial service, including background server and above-mentioned towards financial service
Chat robots, wherein,
The background server and the chat robots and user terminal wireless connection, user terminal is to the chatting machine
Device human hair send the financial business consultation information of user or online order application;The chat robots and the background server into
Row information interacts, with the online order application of the financial business consultation information of response user or response user;
Wherein, the user terminal includes:Desktop computer, tablet computer, laptop and smart mobile phone;The use
Family terminal receives financial business consultation information input by user by webpage or user APP.
The present invention at least possesses following advantageous effect using above technical scheme:
As shown from the above technical solution, this chat robots and system towards financial service provided by the invention, energy
It is enough that business scenario classification is carried out according to financial business consultation information input by user, and according to sorted business scenario described
Corresponding human-computer interaction strategy is extracted in financial base module and is pushed to the human-computer interaction module, so that human-computer interaction module
Human-computer dialogue is carried out according to the human-computer interaction strategy;Meanwhile user's portrait module can be used according to human-computer dialogue contents extraction
For family feature to build user's portrait, order module can pick out the gold for meeting user's portrait from the financial base module
Melt product generation order, and tracing management is carried out to the order after user's confirmation.Compared with prior art, it is provided by the invention this
Towards the chat robots and system of financial service, user data acquisition, answer are completed by way of human-computer interaction, is ordered
The element tasks such as single outbound can promote the operational efficiency of financial service and financial service experience.By sentencing in human-computer interaction
Disconnected user characteristics, and the response under matching scene is exported according to the similarity of all kinds of financial scenario features and puts question to masterplate, pass through
Repeatedly gradually completion user precisely draws a portrait for interaction, so as to provide more accurate financial consultation and more High-effective Service experience.
Description of the drawings
It in order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention, for those of ordinary skill in the art, without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of schematic block diagram for chat robots towards financial service that one embodiment of the invention provides;
Fig. 2 is a kind of schematic block diagram for chat robots system towards financial service that one embodiment of the invention provides.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, technical scheme will be carried out below
Detailed description.Obviously, described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art are obtained all on the premise of creative work is not made
Other embodiment belongs to the scope that the present invention is protected.
Below by drawings and examples, technical scheme is described in further detail.
Referring to Fig. 1 and Fig. 2, a kind of chat robots 10 towards financial service of one embodiment of the invention offer, with one
Background server 20 carries out information exchange, including:Finance data acquisition module 101, financial base module 102, human-computer interaction
Module 103, financial business processing module 104, user's portrait module 105 and order module 106, wherein,
The finance data acquisition module 101, for gathering the finance data in internet and the background server, and
It is sent to the financial base module 102 and carries out classification storage;
The human-computer interaction module 103, for receiving financial business consultation information input by user and carrying out response;It is described
Human-computer interaction interface is equipped with Text Entry, phonetic entry button and multimedia file send key;
The financial business processing module 104, for carrying out business field according to financial business consultation information input by user
Scape is classified, and extracts corresponding human-computer interaction strategy in the financial base module 102 according to sorted business scenario
The human-computer interaction module is pushed to, so that the human-computer interaction module carries out human-computer dialogue according to the human-computer interaction strategy;
User's portrait module 105, for building user's portrait according to human-computer dialogue contents extraction user characteristics;
The order module 106, for picking out the finance for meeting user's portrait from the financial base module 102
Product generates order, and carries out tracing management to the order after user's confirmation.
It should be noted that the human-computer interaction module is divided into:Response submodule and enquirement submodule.Response submodule is used
In all kinds of financial business of response (including credit operation) consulting, submodule is putd question to be needed for puing question to user comprising user
Enquirement and financial status is asked to put question to, and returned data submission form allows user to fill in.
As shown from the above technical solution, this chat robots towards financial service provided by the invention, being capable of basis
Financial business consultation information input by user carries out business scenario classification, and is known according to sorted business scenario in the finance
Corresponding human-computer interaction strategy is extracted in knowledge library module and is pushed to the human-computer interaction module, so that human-computer interaction module is according to institute
It states human-computer interaction strategy and carries out human-computer dialogue;Meanwhile user's portrait module can be according to human-computer dialogue contents extraction user characteristics
To build user's portrait, order module can pick out the financial product for meeting user's portrait from the financial base module
Order is generated, and tracing management is carried out to the order after user's confirmation.Compared with prior art, it is provided by the invention this towards gold
Melt the chat robots of service, the bases such as user data acquisition, answer, order outbound are completed by way of human-computer interaction
Work can promote the operational efficiency of financial service and financial service experience.By judging user characteristics in human-computer interaction, and
Response and the enquirement masterplate under matching scene are exported according to the similarity of all kinds of financial scenario features, it is gradually complete by repeatedly interaction
It precisely draws a portrait into user, so as to provide more accurate financial consultation and more High-effective Service experience.
Preferably, the financial business processing module 104, is specifically used for:
Semantic analysis is carried out to financial business consultation information input by user, to extract keyword;
According to the keyword, the business scenario that user seeks advice from is judged;
Corresponding business scenario has been searched whether in the financial base module 102;
It is pushed away if so, extracting the corresponding human-computer interaction strategy of the business scenario from the financial base module 102
The human-computer interaction module 103 is given, so that the human-computer interaction module 103 is man-machine right according to human-computer interaction strategy progress
Words.
Preferably, it is described that semantic analysis is carried out to financial business consultation information input by user, to extract keyword, specifically
For:
After carrying out data prediction to financial business consultation information input by user, the image completed, voice number will be handled
According to text data is changed into, to extract keyword;Wherein, the data prediction includes:Data parsing, data cleansing, data
Extraction, data conversion, data loading.
It should be noted that described carry out semantic analysis to financial business consultation information input by user, including:It will receive
To the image data of non-text data of user's issue, voice data change into text data;The processing bag of pretreatment image
Include optical character identification (OCR), all kinds of artificial neural network technologies (RNN/LSTM/GRU) and connectionism time sorter
(CTC) technical method being combined;Speech processes include feature extraction, identification modeling and model training, decoding and obtain result.
Preferably, user's portrait module 105, for that can reflect use from the extracting data after the semantic analysis
Family demand, the user characteristic data of credit information are drawn a portrait with building user.
The module it should be noted that user draws a portrait by Feature Extraction Technology and combines the machine learning calculation such as cluster, classification
Method extracts user characteristics, and does tagsort, and user characteristics is categorized as:Demand characteristic, credit feature and product are special
Sign.It is matched with various product in financial base module according to user characteristics, to the highest product of matching degree, will not collected
Enquirement masterplate and air control port corresponding to the product feature arrived return to human-computer interaction module.
Preferably, the financial business processing module 104 is additionally operable to the user for exporting user portrait module 105
Characteristic is input in the financial product Matching Model to prestore, if matching result reaches preset value, controls the order module
106 pick out the financial product generation order for meeting user's portrait from the financial base module 102.
It should be noted that financial business processing module is to include scene matching and product based on matching strategy model realization
Matched major function.
Scene matching is calculated by carrying out matching degree calculating to user characteristics and each classification model of place using machine learning
Classification, cluster, correlation rule, regression analysis scheduling algorithm calculate matching degree in method, and the interaction data of user is categorized into affiliated field
The subsequent processing of chat robots is used in scape.The product matching feature that scene includes is by user characteristics and all kinds of productions
Product characteristic model carries out matching primitives, the similarity of product feature is calculated, when user characteristics reaches the matched pre-set threshold value of product
When, the product of high matching degree is pushed to user by system by chat robots.Other scenes also product consulting, financial knowledge,
The expansible scene library such as credit evaluation, assets assessment, order processing, organization evaluation, trade trend.
Preferably, the financial business processing module 104 builds financial product Matching Model using machine learning algorithm,
And algorithms of different is sought using classification, cluster, correlation rule, the regression analysis in machine learning algorithm, calculate user characteristic data
With the matching degree of financial product Matching Model.
Preferably, the order module 106 includes:
Order application module, for the order application for receiving the financial business application material of user's submission and making a report on;
Air control approval module, for auditing the financial business application material of user's submission, and according to the financial circles after examination & verification
Business application material and order application carry out user credit assessment and payment loan risk assessment, and assessment result is input to and is prestored
Risk evaluation model in, obtain the feature of risk parameter of user;
Module of making loans is paid, it, should by the human-computer interaction module the output phase if the feature of risk parameter is more than threshold value
Response masterplate, otherwise, complete on-line payment and make loans operation;
Order tracing management module, for tracking the refund situation of the order after making loans with management and control payment.
Preferably, machine learning algorithm of the air control approval module based on recurrence, classification, cluster, decision tree etc is built
The risk evaluation model is found, realizes the online examination & approval to user's order.
Preferably, order module 106 includes:
Order application module, for user application and data submit function, including but not limited to apply list make a report on
It submits, material upload and submission, authentication etc. and the relevant interactive operation of order application.
Evaluation services module, data that order application submits are sub-category to be assessed for passing through, and is commented including all kinds of assets
Estimate, such as:Vehicle, room etc..
Air control approval module, including uploading the identification of data validity to user, being assessed, user's assets to number of users
It identified according to anti-fraud, extract user credit data using credit investigation system, by above-mentioned data through credit evaluation model to order application
It is examined, payment instruction is exported by examining, then will not be not by reason feedback user by examining, user can be according to not leading to
Reason is crossed to be defended oneself or updates.
Assets assessment module includes classification assets:The dress that the guaranties such as house property, vehicle, share, bill are assessed
It puts, by the examination & approval amount of money that credit product is determined to the valuation of guaranty.
Identifying system module, by deep learning algorithm to picture and text numbers such as user's uplink data, identifying data, face camera shootings
According to being parsed, judge to upload the authenticity of data.
Collage-credit data library module, indication telecommunications databases system, the system is by accessing all kinds of collage-credit data storehouses (personal and enterprise
Industry) integration of the realization to user credit data.
Credit evaluation module is referred to the credit evaluation model established based on data mining technology and machine learning algorithm, collected
The user data arrived calculates export credit point value of evaluation by model, is completed with reference to credit investigation system database data to user credit
The accurate portrait of data.
By Feature Extraction Technology and the machine learning algorithms such as cluster, classification are combined, are formed same in financial base module
The characteristic of a kind of product forms sort product feature database.Also it is by tagsort per one kind product feature:Demand characteristic,
Credit feature, product feature and other features.
Module of making loans is paid, payment or payment instruction of making loans are sent to the user examined by order.
Order tracing management module feeds back the instruction request that order is sent for handling user, including:Consulting,
Complaint, suggestion etc. are to interaction demand derived from order.
Preferably, the finance data acquisition module 101, is specifically used for:
The finance in internet and the background server is gathered by manual entry, reptile instrument, online transmission mode
Data;The finance data includes:Financial common sense, product data, legal knowledge, risk knowledge, mechanism public sentiment:
Data cleansing, data integration, data conversion and hough transformation pretreatment are carried out to the finance data of acquisition;
Pretreated finance data is sent to the financial base module 102 and carries out classification storage.
It should be noted that this chat robots provided by the invention, using Encoder-Decoder frames, specifically
For, X refers to user's read statement, and Y refers to the answer statement of chat robots.
X=< x1, x2…xm>
Y=<y1, y2…yn>
It is meant that after robot receives user chat data X, is calculated by Encoder-Decoder frames, first
Semantic coding first is carried out to X by Encoder, semantic vector generation is carried out using LSTM models, forms intermediate semantic expressiveness C,
Decoder generates response Y according to intermediate semantic expressiveness C by LSTM models.
Dialogue robot is established using the model, general way is the language material number collected using information acquisition module
According to be used as training data, train corresponding neutral net Connecting quantity in the model with substantial amounts of such language material.
Three door Gate (input, forget, output) of LSTM models and a Cell units composition.Gate uses one
A sigmoid activation primitives, input and cell state would generally be converted using tanh activation primitives.The cell of LSTM can
It is defined with using following equation:
Gates:
it=g (Wxixt+Whiht-1+bi)
ft=g (Wxfxt+Whfht-1+bf)
ot=g (Wxoxt+Whoht-1+bo)
Incoming door itForget door ftOut gate ot
ft=σ (Wf·[ht-1, xt]+bf)
Input transformation:
it=σ (Wi·[ht-1, xt]+bi)
Ct is exported
H is exported
ot=σ (Wo[ht-1, xt]+bo)
ht=ot*tanh(Ct)
Sigmoid activation primitives:
Tanh activation primitives:
The decoded portion of Encoder-Decoder frames has used attention model (Attention), in target sentences
Each word learns the Automobile driving probabilistic information of word in its corresponding source statement.It is former when each word Yi is generated
First be all identical intermediate semantic expressiveness C can be substituted for basis be currently generated word and continually changing Ci.Generate target sentences
The process of word is following form:
y1=f1 (C1)
y2=f1 (C2, y1)
y3=f1 (C3, y1, y2)
The attention of attention model
α i, j be Attention models generation weights, si-1For Decoder hidden states, hjIt is exported for Encoder
A is that the core of Attention models is referred to as alignment models
eij=a (si-1, hj)
Context vector Ci is to hide sequence vector (h1 ..., hTx) to be added by weight, and Tx represents the length of input sentence,
α i, j are the weights of Attention models generation, at the i moment, combination are weighted to feature h, a series of h give birth to for Encoder
Into feature.Formula is as follows:
Si represents the hidden state at decoder i moment.si-1For Decoder hidden states, yi-1It is inputted for Decoder, ci
For Attention functions, calculation formula is
si=f (si-1, yi-1, ci)
yiFor the Decoder output vectors at i moment
yi=g (yi-1, si, ci)
It should be noted that above-mentioned LSTM models can also use other models, such as:CNN/RNN/BiRNN/GRU/LSTM/
The replacements such as Deep LSTM, all in scope of the present invention.
In addition, pass through the anti-method cheated of chat robots the present invention also provides a kind of:Utilize blacklist, user
Portrait model, anti-fraud language material and enquirement, feedback result, output order examination & approval strategy, user feedback result judge whether to meet
Fraud portrait.
Step 1:User draws a portrait and certain class product matching degree reaches overdue threshold values and generates Products Show list;
Step 2:The submission with application material is made a report in user's completion order application;
Step 3:User's portrait module completes the extraction to user characteristics;
Step 4:Order module extracts user characteristics;
Step 5:User data and anti-fraud model are analysed and compared, calculate matching degree;
Step 6:To the doubtful profile feedback of high matching degree user to chat robots;
Step 7:The anti-module of cheating of chat robots exports corresponding enquirement or supplement material to the doubtful feature of user to user
The requirement of material;
Step 8:User makes system feedback;
Step 9:Order module judges whether user meets the corresponding user of anti-fake system, and order approval results are made
Judge.
Referring to Fig. 2, the invention also provides a kind of chat robots system 1 towards financial service, including background service
Device 20 and the above-mentioned chat robots 10 towards financial service, wherein,
The background server 20 and 30 wireless connection of the chat robots 10 and user terminal, user terminal 30 is to institute
It states chat robots 10 and sends the financial business consultation information of user or online order application;The chat robots 10 with it is described
Background server 20 carries out information exchange, with the financial business consultation information of response user or the online order Shen of response user
Please;
Wherein, the user terminal 30 includes but not limited to:Desktop computer, tablet computer, laptop and intelligence
Mobile phone, intelligent wearable device, intelligent audio and video equipment, multimedia terminal, anthropomorphic robot etc.;The user terminal 30 passes through net
Page or user APP receive financial business consultation information input by user.
For the ease of understanding this chat robots and system towards financial service provided by the invention, several realities are enumerated
Under such as:
Example 1:
After user first logs into, robot salutatory is for example:" you are good, has anything that can help you”
User inputs text:" the problem of I wants to ask a little loans "
After user's input text is identified, exports and give user's portrait module;
User's module of drawing a portrait carries out feature extraction to user data, does feature mark to user characteristics, will treated number
Financial business processing module is given according to output;
The support of financial business processing module calculates user characteristics by classification and matching model, and user user is classified
Scene is seeked advice from for product, while is matched with existing product model, no product matching is judged, user data is exported to chat
Loan consultation module in robot;
The corresponding product consultation module of chat robots module is by the analysis to user data, by the interaction data of user
It is categorized as problem data, the responder module in product consultation module exports response according to user input data through interactive strategy model
For " good, out of question, you please put question to ", interaction data is transferred to human-computer interaction module by robot, so as to form the friendship for cycling
Mutual pattern.
Improvement shot and long term memory net on the basis of Recognition with Recurrent Neural Network (RNN) is then utilized in order to realize that the meaning of one's words of context understands
Network (LSTM) is realized.The RNN models of coder-decoder (Encoder-Decoder) are first with LSTM units come to inputting sequence
Row are learnt, and are encoded to the vector expression of regular length;Then represented simultaneously with some LSTM units to read this vector again
Each hidden unit that being decoded as the output sequence model includes has the cell of memory function, and composition includes:
(1) input node:The output and current input for receiving the concealed nodes of a upper moment point are used as input, so
Pass through the activation primitive of a tanh afterwards;
(2) input gate:For control input information, door input for a upper moment point concealed nodes output and
Current input, activation primitive sigmoid;
(3) internal state node:It inputs to be entered the currently inside shape of input and previous time point after a filtering
State node exports;
(4) door is forgotten:For controlling internal state information, the input of door is the output of the concealed nodes of a upper moment point
And current input, activation primitive sigmoid;
(5) out gate:For controlling output information, the input of door for the concealed nodes of a upper moment point output and
Current input, activation primitive sigmoid;
Attention mechanism is added in simultaneously, by LSTM encoders to output among list entries as a result, then training
Output sequence is therewith associated come the study to these making choice property of input and when model exports by one model.
Example 2:
User selects corresponding product according to self-demand and clicks on application button, and application instruction is transferred to order mould
Block;
Order module prompting user fills in request for product list;
User is prompted to upload or submit application materials after list application, data includes:The individual of hand-held identity document
It takes pictures, personal income proves, take in flowing water, bank card, marriage certificate, residence booklet, work prove etc..According to the difference of product, Shen
Please data include part must data and updates.If data does not upload successfully or the incomplete just selection of data is submitted, system
Instruction chat robots are then exported into prompting class interaction data to human-computer interaction module.
Data input air control approval module, air control approval module submit data combination wind according to user after subscriber data is submitted
It controls model and calculates user's default risk, while data is submitted to be identified with the presence or absence of false data to user;User is extracted to hand over
Mutual data and data whether there is the data for meeting anti-fraud model.If user data meets default in anti-fraud model
Threshold values then exports the user characteristics for meeting anti-fraud feature to chat robots, and chat robots are according to anti-fraud model pair
Problem base is answered to export interaction data to human-computer interaction module.
Example 3:
User's portrait mould user characteristics in the block includes:Demand characteristic, credit feature, product feature
Demand characteristic refers to the extraction to the keyword of demand, such as:Borrowing demand:Spend money in 3 days, by the end of December before;Price
Demand:Interest rate is low, no expense.
Credit feature refers to based on the relevant keyword of user credit, such as:It is length of service, monthly income, educational background, whether married
Deng.Product feature refers to and product function, condition, the relevant keyword of feature, such as:It is examined without mortgage, nine folding of interest rate, one day
Deng.
As shown from the above technical solution, this chat robots and system towards financial service provided by the invention lead to
Cross extraction user characteristics structure user portrait, by business classification scene Matching Model establish user behavior classification, realize with
The matching analysis of product selects suitable product type and pushes, and chat robots are according to the interaction mould trained through knowledge base
Type realizes interaction data output using coding-decoding frame combination shot and long term memory models (LSTM), solves context data
Continuity, robot be identified classification of the application class model to interaction data simultaneously, selection uses response or enquirement
Output policy allows robot more actively to excavate user demand in interaction.
The above description is merely a specific embodiment, but protection scope of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in change or replacement, should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Term " first ", " second " are only used for description purpose, and it is not intended that instruction or hint relative importance.Term " multiple " refers to
Two or more, unless otherwise restricted clearly.
Claims (10)
1. a kind of chat robots towards financial service carry out information exchange, which is characterized in that bag with a background server
It includes:Finance data acquisition module, financial base module, human-computer interaction module, financial business processing module, user's portrait module
And order module, wherein,
The finance data acquisition module, for gathering the finance data in internet and the background server, and is sent to
The finance base module carries out classification storage;
The human-computer interaction module, for receiving financial business consultation information input by user and carrying out response;The man-machine friendship
Mutual interface is equipped with Text Entry, phonetic entry button and multimedia file send key;
The financial business processing module, for carrying out business scenario classification according to financial business consultation information input by user,
And it is extracted according to sorted business scenario in the financial base module described in corresponding human-computer interaction strategy is pushed to
Human-computer interaction module, so that the human-computer interaction module carries out human-computer dialogue according to the human-computer interaction strategy;
User's portrait module, for building user's portrait according to human-computer dialogue contents extraction user characteristics;
The order module meets the financial product generation of user's portrait and orders for being picked out from the financial base module
It is single, and tracing management is carried out to the order after user's confirmation.
2. the chat robots according to claim 1 towards financial service, which is characterized in that the financial business processing
Module is specifically used for:
Semantic analysis is carried out to financial business consultation information input by user, to extract keyword;
According to the keyword, the business scenario that user seeks advice from is judged;
Corresponding business scenario has been searched whether in the financial base module;
If so, extracted from the financial base module the corresponding human-computer interaction strategy of the business scenario be pushed to it is described
Human-computer interaction module, so that the human-computer interaction module carries out human-computer dialogue according to the human-computer interaction strategy.
3. the chat robots according to claim 2 towards financial service, which is characterized in that described to input by user
Financial business consultation information carries out semantic analysis, to extract keyword, is specially:
After carrying out data prediction to financial business consultation information input by user, the image completed will be handled, voice data turns
Text data is melted into, to extract keyword;Wherein, the data prediction includes:Data parsing, data cleansing, data pick-up,
Data conversion, data loading.
4. the chat robots according to claim 3 towards financial service, which is characterized in that user's portrait mould
Block, for from the extracting data after the semantic analysis can reflect user demand, credit information user characteristic data with
Build user's portrait.
5. the chat robots according to claim 4 towards financial service, which is characterized in that the financial business processing
Module is additionally operable to for the user characteristic data of user portrait module output to be input to the financial product Matching Model to prestore
In, if matching result reaches preset value, the order module is controlled to be picked out from the financial base module and meets user
The financial product generation order of portrait.
6. the chat robots according to claim 5 towards financial service, which is characterized in that the financial business processing
Module builds financial product Matching Model using machine learning algorithm, and using in machine learning algorithm classification, cluster, pass
Algorithms of different is sought in connection rule, regression analysis, calculates the matching degree of user characteristic data and financial product Matching Model.
7. the chat robots according to claim 1 towards financial service, which is characterized in that the order module bag
It includes:
Order application module, for the order application for receiving the financial business application material of user's submission and making a report on;
Air control approval module, for auditing the financial business application material of user's submission, and according to the financial business Shen after examination & verification
Please material and order application carry out user credit assessment and payment loan risk assessment, and assessment result is input to the wind to prestore
In dangerous assessment models, the feature of risk parameter of user is obtained;
Module of making loans is paid, if the feature of risk parameter is more than threshold value, accordingly should by human-computer interaction module output
Masterplate is answered, otherwise, on-line payment is completed and makes loans operation;
Order tracing management module, for tracking the refund situation of the order after making loans with management and control payment.
8. the chat robots according to claim 7 towards financial service, which is characterized in that the air control approval module
Machine learning algorithm based on recurrence, classification, cluster, decision tree etc establishes the risk evaluation model, realizes and user is ordered
Single online examination & approval.
9. the chat robots according to claim 1 towards financial service, which is characterized in that the finance data acquisition
Module is specifically used for:
The finance data in internet and the background server is gathered by manual entry, reptile instrument, online transmission mode;
The finance data includes:Financial common sense, product data, legal knowledge, risk knowledge, mechanism public sentiment;
Data cleansing, data integration, data conversion and hough transformation pretreatment are carried out to the finance data of acquisition;
Pretreated finance data is sent to the financial base module and carries out classification storage.
10. a kind of chat robots system towards financial service, which is characterized in that including background server and claim 1
~9 any one of them towards financial service chat robots, wherein,
The background server and the chat robots and user terminal wireless connection, user terminal is to the chat robots
Send the financial business consultation information of user or online order application;The chat robots carry out letter with the background server
Breath interaction, with the online order application of the financial business consultation information of response user or response user;
Wherein, the user terminal includes:Desktop computer, tablet computer, laptop and smart mobile phone;The user is whole
End receives financial business consultation information input by user by webpage or user APP.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810030096.4A CN108121824A (en) | 2018-01-12 | 2018-01-12 | A kind of chat robots and system towards financial service |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810030096.4A CN108121824A (en) | 2018-01-12 | 2018-01-12 | A kind of chat robots and system towards financial service |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108121824A true CN108121824A (en) | 2018-06-05 |
Family
ID=62233957
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810030096.4A Pending CN108121824A (en) | 2018-01-12 | 2018-01-12 | A kind of chat robots and system towards financial service |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108121824A (en) |
Cited By (38)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108921695A (en) * | 2018-06-22 | 2018-11-30 | 四川华翼共享区块链科技有限公司 | A kind of civil aviaton's customer information management system based on block chain |
CN109189903A (en) * | 2018-08-08 | 2019-01-11 | 互通金融科技有限公司 | High contact manager robot system and its method |
CN109255697A (en) * | 2018-08-15 | 2019-01-22 | 普信恒业科技发展(北京)有限公司 | A kind of automatic credit estimation method and system based on artificial intelligence |
CN109344174A (en) * | 2018-09-13 | 2019-02-15 | 深圳易投云智能科技有限公司 | Financial analysis method and system |
CN109360052A (en) * | 2018-09-27 | 2019-02-19 | 北京亚联之星信息技术有限公司 | A kind of data classification based on machine learning algorithm, data processing method and equipment |
CN109408694A (en) * | 2018-09-25 | 2019-03-01 | 广东中标数据科技股份有限公司 | A kind of customs reaches a standard shipping bill analysis method, system and device |
CN109447002A (en) * | 2018-10-31 | 2019-03-08 | 广州慧睿思通信息科技有限公司 | Government affairs business material preliminary examination method, apparatus, equipment and storage medium |
CN109492712A (en) * | 2018-12-17 | 2019-03-19 | 上海应用技术大学 | The method for establishing internet finance air control model |
CN109492679A (en) * | 2018-10-24 | 2019-03-19 | 杭州电子科技大学 | Based on attention mechanism and the character recognition method for being coupled chronological classification loss |
CN109492863A (en) * | 2018-09-28 | 2019-03-19 | 平安普惠企业管理有限公司 | The automatic generation method and device of financial document |
CN109636237A (en) * | 2018-12-27 | 2019-04-16 | 上海硕恩网络科技股份有限公司 | A kind of financial risks appraisal procedure and system |
CN109711892A (en) * | 2018-12-28 | 2019-05-03 | 浙江百应科技有限公司 | The method for automatically generating client's label during Intelligent voice dialog |
CN109727091A (en) * | 2018-12-14 | 2019-05-07 | 平安科技(深圳)有限公司 | Products Show method, apparatus, medium and server based on dialogue robot |
CN109785140A (en) * | 2018-12-05 | 2019-05-21 | 上海拍拍贷金融信息服务有限公司 | A kind of financing investment planning method and device based on intelligent sound |
CN110009505A (en) * | 2019-04-09 | 2019-07-12 | 昆山古鳌电子机械有限公司 | A kind of financial product recommender system |
CN110019742A (en) * | 2018-06-19 | 2019-07-16 | 北京京东尚科信息技术有限公司 | Method and apparatus for handling information |
CN110019316A (en) * | 2018-07-06 | 2019-07-16 | 新疆玖富万卡信息技术有限公司 | Information interacting method and server |
CN110162641A (en) * | 2019-05-17 | 2019-08-23 | 五竹科技(天津)有限公司 | Marketing method, device and storage medium based on speech enabled |
CN110210902A (en) * | 2019-05-27 | 2019-09-06 | 北京金山安全软件有限公司 | Click through rate prediction method, device and equipment |
CN110473086A (en) * | 2019-08-14 | 2019-11-19 | 中国工商银行股份有限公司 | Treating method and apparatus for business |
CN110489623A (en) * | 2019-07-10 | 2019-11-22 | 本识科技(深圳)有限公司 | A kind of intelligent assistant's system and intelligent assistant robot based on user information interaction |
CN110581772A (en) * | 2019-09-06 | 2019-12-17 | 腾讯科技(深圳)有限公司 | Instant messaging message interaction method and device and computer readable storage medium |
CN110827372A (en) * | 2018-08-09 | 2020-02-21 | 普华云创科技(北京)有限公司 | Method, system and storage medium for constructing person label portrait based on block chain |
CN110827033A (en) * | 2019-10-11 | 2020-02-21 | 支付宝(杭州)信息技术有限公司 | Information processing method and device and electronic equipment |
CN111161068A (en) * | 2019-12-23 | 2020-05-15 | 支付宝(杭州)信息技术有限公司 | Transaction element checking method and device |
CN111191921A (en) * | 2019-12-27 | 2020-05-22 | 安徽科讯金服科技有限公司 | Small and micro enterprise operation condition wind control analysis system |
CN111353013A (en) * | 2018-12-05 | 2020-06-30 | 中兴通讯股份有限公司 | Method and system for realizing intelligent delivery and reception |
CN111353028A (en) * | 2020-02-20 | 2020-06-30 | 支付宝(杭州)信息技术有限公司 | Method and device for determining customer service call cluster |
CN111429270A (en) * | 2020-04-22 | 2020-07-17 | 广州东百信息科技有限公司 | Overseas credit card wind control model acquisition method, device, equipment and storage medium |
CN111914983A (en) * | 2019-05-07 | 2020-11-10 | 阿里巴巴集团控股有限公司 | Interaction method and device, sound box, electronic equipment and storage medium |
CN112463937A (en) * | 2020-11-02 | 2021-03-09 | 上海中船船舶设计技术国家工程研究中心有限公司 | Intelligent psychological service software system |
CN112506963A (en) * | 2020-11-23 | 2021-03-16 | 上海方立数码科技有限公司 | Multi-service-scene-oriented service robot problem matching method |
CN112734458A (en) * | 2020-12-25 | 2021-04-30 | 珠海大横琴科技发展有限公司 | Method and device for inviting business, electronic equipment and readable storage medium |
CN113313337A (en) * | 2020-10-10 | 2021-08-27 | 国网冀北电力有限公司物资分公司 | Intelligent wind control system and management method for material management |
CN114820186A (en) * | 2022-04-01 | 2022-07-29 | 北京卓视智通科技有限责任公司 | Service system and method based on block chain and chat robot |
CN116010465A (en) * | 2023-03-28 | 2023-04-25 | 杭州东岸网络信息服务有限公司 | Financial business management system and method |
CN116186543A (en) * | 2023-03-01 | 2023-05-30 | 深圳崎点数据有限公司 | Financial data processing system and method based on image recognition |
CN116883181A (en) * | 2023-09-07 | 2023-10-13 | 北京中关村科金技术有限公司 | Financial service pushing method based on user portrait, storage medium and server |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105975622A (en) * | 2016-05-28 | 2016-09-28 | 蔡宏铭 | Multi-role intelligent chatting method and system |
CN106991611A (en) * | 2017-03-27 | 2017-07-28 | 北京贝塔智投科技有限公司 | A kind of intelligence financing investment consultant's robot system and its method of work |
CN107330785A (en) * | 2017-07-10 | 2017-11-07 | 广州市触通软件科技股份有限公司 | A kind of petty load system and method based on the intelligent air control of big data |
-
2018
- 2018-01-12 CN CN201810030096.4A patent/CN108121824A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105975622A (en) * | 2016-05-28 | 2016-09-28 | 蔡宏铭 | Multi-role intelligent chatting method and system |
CN106991611A (en) * | 2017-03-27 | 2017-07-28 | 北京贝塔智投科技有限公司 | A kind of intelligence financing investment consultant's robot system and its method of work |
CN107330785A (en) * | 2017-07-10 | 2017-11-07 | 广州市触通软件科技股份有限公司 | A kind of petty load system and method based on the intelligent air control of big data |
Cited By (50)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110019742A (en) * | 2018-06-19 | 2019-07-16 | 北京京东尚科信息技术有限公司 | Method and apparatus for handling information |
CN110019742B (en) * | 2018-06-19 | 2024-05-17 | 北京京东尚科信息技术有限公司 | Method and device for processing information |
CN108921695A (en) * | 2018-06-22 | 2018-11-30 | 四川华翼共享区块链科技有限公司 | A kind of civil aviaton's customer information management system based on block chain |
CN110019316A (en) * | 2018-07-06 | 2019-07-16 | 新疆玖富万卡信息技术有限公司 | Information interacting method and server |
CN109189903A (en) * | 2018-08-08 | 2019-01-11 | 互通金融科技有限公司 | High contact manager robot system and its method |
CN110827372A (en) * | 2018-08-09 | 2020-02-21 | 普华云创科技(北京)有限公司 | Method, system and storage medium for constructing person label portrait based on block chain |
CN109255697A (en) * | 2018-08-15 | 2019-01-22 | 普信恒业科技发展(北京)有限公司 | A kind of automatic credit estimation method and system based on artificial intelligence |
CN109344174A (en) * | 2018-09-13 | 2019-02-15 | 深圳易投云智能科技有限公司 | Financial analysis method and system |
CN109408694A (en) * | 2018-09-25 | 2019-03-01 | 广东中标数据科技股份有限公司 | A kind of customs reaches a standard shipping bill analysis method, system and device |
CN109360052A (en) * | 2018-09-27 | 2019-02-19 | 北京亚联之星信息技术有限公司 | A kind of data classification based on machine learning algorithm, data processing method and equipment |
CN109492863A (en) * | 2018-09-28 | 2019-03-19 | 平安普惠企业管理有限公司 | The automatic generation method and device of financial document |
CN109492679A (en) * | 2018-10-24 | 2019-03-19 | 杭州电子科技大学 | Based on attention mechanism and the character recognition method for being coupled chronological classification loss |
CN109447002A (en) * | 2018-10-31 | 2019-03-08 | 广州慧睿思通信息科技有限公司 | Government affairs business material preliminary examination method, apparatus, equipment and storage medium |
CN111353013A (en) * | 2018-12-05 | 2020-06-30 | 中兴通讯股份有限公司 | Method and system for realizing intelligent delivery and reception |
CN109785140A (en) * | 2018-12-05 | 2019-05-21 | 上海拍拍贷金融信息服务有限公司 | A kind of financing investment planning method and device based on intelligent sound |
CN109727091A (en) * | 2018-12-14 | 2019-05-07 | 平安科技(深圳)有限公司 | Products Show method, apparatus, medium and server based on dialogue robot |
CN109727091B (en) * | 2018-12-14 | 2022-04-05 | 平安科技(深圳)有限公司 | Product recommendation method, device, medium and server based on conversation robot |
CN109492712A (en) * | 2018-12-17 | 2019-03-19 | 上海应用技术大学 | The method for establishing internet finance air control model |
CN109636237A (en) * | 2018-12-27 | 2019-04-16 | 上海硕恩网络科技股份有限公司 | A kind of financial risks appraisal procedure and system |
CN109711892A (en) * | 2018-12-28 | 2019-05-03 | 浙江百应科技有限公司 | The method for automatically generating client's label during Intelligent voice dialog |
CN110009505A (en) * | 2019-04-09 | 2019-07-12 | 昆山古鳌电子机械有限公司 | A kind of financial product recommender system |
CN111914983B (en) * | 2019-05-07 | 2023-10-24 | 阿里巴巴集团控股有限公司 | Interaction method and device, sound box, electronic equipment and storage medium |
CN111914983A (en) * | 2019-05-07 | 2020-11-10 | 阿里巴巴集团控股有限公司 | Interaction method and device, sound box, electronic equipment and storage medium |
CN110162641B (en) * | 2019-05-17 | 2022-01-04 | 五竹科技(北京)有限公司 | Marketing method and device based on audio interaction and storage medium |
CN110162641A (en) * | 2019-05-17 | 2019-08-23 | 五竹科技(天津)有限公司 | Marketing method, device and storage medium based on speech enabled |
CN110210902A (en) * | 2019-05-27 | 2019-09-06 | 北京金山安全软件有限公司 | Click through rate prediction method, device and equipment |
CN110489623A (en) * | 2019-07-10 | 2019-11-22 | 本识科技(深圳)有限公司 | A kind of intelligent assistant's system and intelligent assistant robot based on user information interaction |
CN110473086A (en) * | 2019-08-14 | 2019-11-19 | 中国工商银行股份有限公司 | Treating method and apparatus for business |
CN110581772A (en) * | 2019-09-06 | 2019-12-17 | 腾讯科技(深圳)有限公司 | Instant messaging message interaction method and device and computer readable storage medium |
CN110581772B (en) * | 2019-09-06 | 2020-10-13 | 腾讯科技(深圳)有限公司 | Instant messaging message interaction method and device and computer readable storage medium |
CN110827033A (en) * | 2019-10-11 | 2020-02-21 | 支付宝(杭州)信息技术有限公司 | Information processing method and device and electronic equipment |
CN111161068A (en) * | 2019-12-23 | 2020-05-15 | 支付宝(杭州)信息技术有限公司 | Transaction element checking method and device |
CN111161068B (en) * | 2019-12-23 | 2022-06-14 | 支付宝(杭州)信息技术有限公司 | Transaction element checking method and device |
CN111191921A (en) * | 2019-12-27 | 2020-05-22 | 安徽科讯金服科技有限公司 | Small and micro enterprise operation condition wind control analysis system |
CN111353028A (en) * | 2020-02-20 | 2020-06-30 | 支付宝(杭州)信息技术有限公司 | Method and device for determining customer service call cluster |
CN111353028B (en) * | 2020-02-20 | 2023-04-18 | 支付宝(杭州)信息技术有限公司 | Method and device for determining customer service call cluster |
CN111429270A (en) * | 2020-04-22 | 2020-07-17 | 广州东百信息科技有限公司 | Overseas credit card wind control model acquisition method, device, equipment and storage medium |
CN113313337A (en) * | 2020-10-10 | 2021-08-27 | 国网冀北电力有限公司物资分公司 | Intelligent wind control system and management method for material management |
CN112463937B (en) * | 2020-11-02 | 2022-10-11 | 上海中船船舶设计技术国家工程研究中心有限公司 | Intelligent psychological service software system |
CN112463937A (en) * | 2020-11-02 | 2021-03-09 | 上海中船船舶设计技术国家工程研究中心有限公司 | Intelligent psychological service software system |
CN112506963B (en) * | 2020-11-23 | 2022-09-09 | 上海方立数码科技有限公司 | Multi-service-scene-oriented service robot problem matching method |
CN112506963A (en) * | 2020-11-23 | 2021-03-16 | 上海方立数码科技有限公司 | Multi-service-scene-oriented service robot problem matching method |
CN112734458A (en) * | 2020-12-25 | 2021-04-30 | 珠海大横琴科技发展有限公司 | Method and device for inviting business, electronic equipment and readable storage medium |
CN114820186A (en) * | 2022-04-01 | 2022-07-29 | 北京卓视智通科技有限责任公司 | Service system and method based on block chain and chat robot |
CN114820186B (en) * | 2022-04-01 | 2024-03-15 | 北京卓视智通科技有限责任公司 | Service system and method based on block chain and chat robot |
CN116186543A (en) * | 2023-03-01 | 2023-05-30 | 深圳崎点数据有限公司 | Financial data processing system and method based on image recognition |
CN116186543B (en) * | 2023-03-01 | 2023-08-22 | 深圳崎点数据有限公司 | Financial data processing system and method based on image recognition |
CN116010465A (en) * | 2023-03-28 | 2023-04-25 | 杭州东岸网络信息服务有限公司 | Financial business management system and method |
CN116883181A (en) * | 2023-09-07 | 2023-10-13 | 北京中关村科金技术有限公司 | Financial service pushing method based on user portrait, storage medium and server |
CN116883181B (en) * | 2023-09-07 | 2023-12-05 | 北京中关村科金技术有限公司 | Financial service pushing method based on user portrait, storage medium and server |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108121824A (en) | A kind of chat robots and system towards financial service | |
Souma et al. | Enhanced news sentiment analysis using deep learning methods | |
CN109767318A (en) | Loan product recommended method, device, equipment and storage medium | |
CN110162700A (en) | The training method of information recommendation and model, device, equipment and storage medium | |
CN112418520B (en) | Credit card transaction risk prediction method based on federal learning | |
CN109410036A (en) | A kind of fraud detection model training method and device and fraud detection method and device | |
Chang et al. | Measuring the information content of financial news | |
Coyne et al. | Forecasting stock prices using social media analysis | |
CN109214914A (en) | A kind of loan information checking method and device based on communication open platform | |
Jiang et al. | Cost-sensitive parallel learning framework for insurance intelligence operation | |
CN109919006A (en) | Expression detection method, device, electronic equipment and storage medium | |
CN111914553B (en) | Financial information negative main body judging method based on machine learning | |
CN107944851A (en) | The management method of group's fund account, apparatus and system | |
CN111104975A (en) | Credit assessment model based on breadth learning | |
Yfantis et al. | Exploring the implementation of artificial intelligence in the public sector: welcome to the clerkless public offices. applications in education | |
Yuping et al. | New methods of customer segmentation and individual credit evaluation based on machine learning | |
Sutkatti et al. | Stock market forecasting techniques: A survey | |
Cerny et al. | Data mining and neural networks from a commercial perspective | |
CN109829593A (en) | The credit rating of target object determines method, apparatus, storage medium and electronic device | |
CN113449103B (en) | Bank transaction running water classification method and system integrating label and text interaction mechanism | |
CN117034955A (en) | Telephone traffic text intention recognition method and device | |
Prasanth et al. | Intelligent Loan Eligibility and Approval System based on Random Forest Algorithm using Machine Learning | |
CN115907968A (en) | Wind control rejection inference method and device based on pedestrian credit | |
CN115687754A (en) | Active network information mining method based on intelligent conversation | |
Zhao et al. | Innovative mechanism of rural finance: Risk assessment methods and impact factors of agricultural loans based on personal emotion and artificial intelligence |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180605 |
|
RJ01 | Rejection of invention patent application after publication |