CN107481143A - A kind of intelligent stock commending system and implementation method - Google Patents
A kind of intelligent stock commending system and implementation method Download PDFInfo
- Publication number
- CN107481143A CN107481143A CN201710633479.6A CN201710633479A CN107481143A CN 107481143 A CN107481143 A CN 107481143A CN 201710633479 A CN201710633479 A CN 201710633479A CN 107481143 A CN107481143 A CN 107481143A
- Authority
- CN
- China
- Prior art keywords
- stock
- unit
- user
- label
- intelligent
- 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
- 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
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of intelligent stock commending system and implementation method, system to include：Distributed Calculation unit, stock tag library, user characteristics unit, intelligent control unit, the stock tag library, to establish the stock similarity gene based on label, the user characteristics unit, to excavate characterization factor based on user's self-selected stock, the intelligent control unit, to according to characterization factor and stock similarity gene, individual character recommendation is carried out for user, the Distributed Calculation unit, to perform and/or distribute the calculating task in above-mentioned stock tag library, above-mentioned user characteristics unit, above-mentioned intelligent control unit.Intelligent recommendation to stock can be realized by the present invention, the stock similitude recommended is higher to have reference value very much.In addition, the present invention can adjust weights according to user to the operational feedback of self-selected stock, the adjustment that stock is recommended is realized, the hobby for making commending system to be more close to the users, improves and recommends the degree of correlation.
Description
Technical field
The present invention relates to computer software fields, more particularly to a kind of intelligent stock commending system and implementation method.
Background technology
Nowadays due to the popularization of internet, the personalized recommendation in many fields is referred to again and again always.Current pushes away
Recommend method and contact user interest and related article by following several ways mostly.Such as in certain methods be by using
The article that family was liked, the article similar to the article that he liked can be recommended to user, here it is the algorithm based on article
(item-based)；For example, in certain methods it is again by the other users similar with user interest：That can be recommended to user
The article that a little other users similar with their hobbies are liked, here it is the algorithm based on user (user-based)；Again
For example in certain methods be to contact user and article by some features (feature), to user those can be recommended to have
The article for the feature that user likes.It can be seen that the final purpose of commending system is all consistent：The letter that recommended user is most interested in
Breath, while help user to save the time.
Some traditional Stock investment analysis methods mainly include, 1) Fundamental Analysis, be using traditional economy theory as
Basis, using enterprise value as main study subject, by determining internal value and influenceing the macroscopic view of stock price to pass through
Ji situation, industry development prospect, enterprise management condition etc. carry out analyzing in detail, probably to calculate the long-term investment valency of listed company
Value and the margin of safety, and compared with current stock price, form corresponding suggestion for investment.2) technical Analysis method, be with
Based on traditional security theory, using stock price as main study subject, using predicting Stock Price fluctuation tendency as main purpose,
Start with from the history chart of the change of stock price, the method summation analyzed stock market fluctuations rule.Technical Analysis thinks city
Field behavior, which is contained, digests all, and volatility can be with quantitative analysis and prediction, such as Dow Theory, Wave Theory, river grace theory.
3) EVOLUTION ANALYSIS method, it is based on evolution security theory, using the life movement attribute of fluctuation of stock market as main research pair
As starting with from the metabolic of stock market, going after profit or gain property, adaptability, plasticity, irritability, variability and rhythmicity etc., to market
Mobile state follow-up study is entered in fluctuation direction with space, and chance and the method summation of risk assessment are provided for stock exchange decision-making.
Then the general user of in the market is in the stock market data in face of magnanimity, as a consequence it is hardly possible to weight is refined in these information
Want information to be operated, rational assets accrued revenue can not be more obtained in stock market.Method relatively conventional at present is generally wrapped
Include：Recommended based on stock, by monitoring with collecting the multinomial key index such as real-time transaction data, public sentiment and Domestic News, with reference to
Consumer's risk preference, investment reference is provided the user, user can be selected stocks by platform progress intelligence, intelligence examines stock and intelligence is multiple
Disk.For example it is that to carry out, stock is similar to be pushed away based on correlative factors such as funds flow seniority among brothers and sisters+amount of increase seniority among brothers and sisters+industries that some stocks, which are recommended,
Recommend, but have a disadvantage in that：Because dimension is less, similitude is simultaneously inadequate；Secondly the stock reference value recommended is little, is only
Main force's funds flow and amount of increase ranking on the same day, but reference value to user and little.In such as Chinese patent application
CN201480014360.5, ascendant trend intensity of the instruction on corresponding stock on chart stock screen is to be used to show chart
The signal of the time buying of stock, when ascendant trend intensity is 1, stock works as ascendant trend with the recommended purchase of 1/3 ratio
When intensity is 2, stock is with the recommended purchase of 2/3 ratio, and when ascendant trend intensity is 3, and stock is very
(totalquantity) purchase is recommended, ascendant trend intensity can be used to provide a user the recommendation of time buying.Shortcoming exists
In the interest for being not bound with user is recommended, and is more inclined to technology and is selected stocks.Such as Chinese patent application again
CN201611016123.X, a kind of stock recommendation method counted based on last-period forecast amount of increase and amount of decrease and its consistency, 1) all
Stock in search for and obtain more similar tendency sections；2) ballot statistics is carried out with the later stage tendency of similar shares changing tendency section；
3) the recent recommendation of stock and combined recommendation.Shortcoming is：User's stock interested can not be positioned.
The content of the invention
The technical problem to be solved in the present invention is to provide a kind of carries out individual character recommendation with reference to similarity and user preferences
Intelligent stock commending system.
Solve above-mentioned technical problem, the invention provides a kind of intelligent stock commending system, including：Distributed Calculation list
Member, stock tag library, user characteristics unit, intelligent control unit,
The stock tag library, to establish the stock similarity gene based on label,
The user characteristics unit, to excavate characterization factor based on user's self-selected stock,
The intelligent control unit, according to characterization factor and stock similarity gene, to carry out individual character for user and push away
Recommend,
The Distributed Calculation unit, to perform and/or distribute above-mentioned stock tag library, above-mentioned user characteristics unit,
Calculating task in above-mentioned intelligent control unit.
Further, the stock tag library includes：Basic tag unit, concepts tab unit, stock nature label list
First, technical tag unit,
The basic tag unit, the base attribute of stock is quantified as into basic label,
The concepts tab unit, the concept of stock is quantified as into concepts tab,
The stock nature label unit, the characteristic of stock of stock is quantified as into characteristic of stock label,
The technical tag unit, the technological side of stock is quantified as into technical label.
Further, the stock tag library includes：Basic tag unit, correlation tag unit, stock nature label
Unit, technical tag unit,
The basic tag unit, the base attribute of stock is quantified as into basic label,
The correlation tag unit, the correlation of stock is quantified as into concepts tab,
The stock nature label unit, the characteristic of stock of stock is quantified as into characteristic of stock label,
The technical tag unit, the technological side of stock is quantified as into technical label.
Further, the user characteristics unit includes：Operation flowing water collector unit, recent speculation in stocks mode collect list in the recent period
Member,
The operation flowing water collector unit in the recent period, to collect the characteristic of user's water operation in certain period of time
According to,
The speculation in stocks mode collector unit in the recent period, the characteristic for mode of speculating in shares collecting user in certain period of time
According to.
Further, the user characteristics unit includes：User data polymerized unit and user data excavate unit,
The user data polymerized unit, it is accustomed to entirety to be speculated in shares according to the user of definition, obtains user's speculation in stocks custom
Form class,
The user data excavates unit, to analyze speculation in stocks preference of the user in certain period of time.
Further, system also includes：Focus unit, to as access stock market's current stock related focus API.
Further, system also includes：Scene dynamics adjustment unit, to different Stock-operations, to be adjusted according to user
Stock state scene corresponding to section.
Further, the intelligent control unit includes：Weight unit, score unit,
The weight unit, to carry out weight distribution to the stock similarity gene in the stock tag library,
The scoring unit, to carry out comprehensive weight scoring based on weight unit.
Based on above-mentioned, present invention also offers a kind of intelligent stock to recommend method, comprises the following steps：
Stock similarity gene of the foundation based on label/be based on user's self-selected stock excavation characterization factor,
The self-selected stock of user's selection is obtained,
According to characterization factor and stock similarity gene, individual character recommendation is carried out for user.
Further, using slicing algorithm, message queue or based on Hadoop Distributed Architecture in the above method
Beneficial effects of the present invention：
A kind of intelligent stock commending system in the present invention is due to including establishing the stock similarity based on label
The stock tag library of gene, can establish between all stocks the related network to quantize.Due to including to based on user from
Select stocks and excavate the user characteristics unit of characterization factor and due to including to according to characterization factor and stock similarity gene, for
User carries out the intelligent control unit of individual character recommendation, the self-selected stock that can be added according to user, finds most relevant property in a network
Stock recommended, and the weight of stock, the foundation as recommendation are adjusted according to current focus and scene dynamics.Due to bag
Include the Distributed Calculation unit for performing and/or distributing calculating task, it is possible to increase the operating efficiency of commending system, when reducing processing
Between, improve handling capacity.
In addition, using the method in the present invention, weights, such as frequency can be adjusted to the operational feedback of self-selected stock according to user
It is numerous to check, delete etc., the adjustment that stock is recommended is realized, the hobby for making commending system to be more close to the users, improves and recommends
The degree of correlation, make more personalized and intelligent commending system.
Brief description of the drawings
Fig. 1 is the system structure diagram in one embodiment of the invention；
Fig. 2 is the stock tag library structural representation in Fig. 1；
Fig. 3 is another structural representation of stock tag library in Fig. 1；
Fig. 4 is the user characteristics cellular construction schematic diagram in Fig. 1；
Fig. 5 is the user characteristics cellular construction journey schematic diagram in Fig. 1；
Fig. 6 is the system structure diagram in one embodiment of the present invention；
Fig. 7 is the system structure diagram in another preferred embodiment of the present invention；
Fig. 8 is the intelligent control cellular construction schematic diagram in Fig. 1；
Fig. 9 is the method flow schematic diagram in one embodiment of the invention；
Figure 10 is the method flow diagram in the present invention.
Embodiment
The principle of the disclosure is described referring now to some example embodiments.It is appreciated that these embodiments are merely for saying
It is bright and help it will be understood by those skilled in the art that with the purpose of the embodiment disclosure and describe, rather than suggest the model to the disclosure
Any restrictions enclosed.Content of this disclosure described here can in a manner of described below outside various modes implement.
As described herein, term " comprising " and its various variants are construed as open-ended term, it means that " bag
Include but be not limited to ".Term "based" is construed as " being based at least partially on ".Term " one embodiment " it is understood that
For " at least one embodiment ".Term " another embodiment " is construed as " at least one other embodiment ".
Those skilled in the art can understand, in the application, basic side：Refer to the achievement of stock with subject matter for personal share
Congenial value have influence.
Platter surface：Referring to that the tendency of deep bid has for the congenial value of personal share influences.
Market plane：Referring to that market focus has for the congenial value of personal share influences.
Technological side：Refer to that price trend has influence for the congenial value of personal share in itself.
Characteristic of stock face：Referring to that personal share volatility rate has for the congenial value of personal share influences.
The stock storehouse of self-selected stock, i.e. oneself selection.The stock that oneself is had an optimistic view of is added to oneself selected self-selected stock by user
In stock row, the used time can see multiple stocks, more convenient, and a mouse click right button can return to interface, can watch shares changing tendency.
Position in storehouse：Refer to that the market value for the stock that user holds accounts for the ratio of your total fund.
Open a position, hold position, close a position：The overall process of futures exchange, which may be summarized to be, opens a position, holds position, closing a position or physical delivery.
Open a position and also call off storehouse, refer to that dealer newly buys in or newly sold a number of futures contract.
Close a position, refer to write off original futures contract by equal, the in opposite direction futures exchange of a stroke count amount, with this
Futures exchange is tied by factory, releases the obligation for the progress physical delivery that expires.
Hold position, refer to the contract do not closed a position still after opening a position, be open interest or uncovered position position, also cry and hold
Storehouse.Dealer can select two ways to finish futures contract after opening a position：Select a good opportunity and close a position, otherwise retain to last trade
Day simultaneously carries out physical delivery.
In Hadoop Distributed Architecture is the basic platform framework that mass data calculates storage.
Calculate the MapReduce for referring to Distributed Calculation.
Storage refers to HDFS, has HBase, Hive based on this upper strata, for doing data storage.
Platform refers to, multiple users can be given to use, it is only necessary to write service logic according to corresponding interface, then
Program is published on platform in the form of wrapping, platform is allocated scheduling calculating etc..
Hadoop Distributed Architecture, including：Task division (message queue), subtask (service logic of processing unit),
As a result the steps such as (the result of calculation storage of processing unit), result of calculation (the customer consumption custom Data Data of corresponding storage) are merged
Suddenly.
Fig. 1 is the system structure diagram in one embodiment of the invention, and a kind of intelligent stock in the present embodiment is recommended
System, including：Distributed Calculation unit 4, stock tag library 1, user characteristics unit 2, intelligent control unit 3, the stock mark
Storehouse 1 is signed, to establish the stock similarity gene based on label, the user characteristics unit 2, to be dug based on user's self-selected stock
Dig characterization factor, the intelligent control unit 3, according to characterization factor and stock similarity gene, to be carried out for user individual
Property recommend, the Distributed Calculation unit 4, to perform and/or distribute above-mentioned stock tag library 1, above-mentioned user characteristics unit
2nd, the calculating task in above-mentioned intelligent control unit 3.Mainly undertaken in whole commending system by the Distributed Calculation unit 4
Calculating task in stock tag library 1, user characteristics unit 2, intelligent control unit 3, so that the generation between stock is closed
Connection, and the unique speculation in stocks of each user can be accumulated according to the long-term behavior of user and likes gene.In the basis of stock tag library 1
Different dimension is established based on the label of " thing ", quantifies the similarity between stock, is included but is not limited to, basic side, greatly
The labels such as card, market plane, technological side, characteristic of stock face, the similarity established by label between stock.In the user characteristics list
Member 2 mainly carries out that user characteristics is excavated and user behavior is collected according to the self-selected stock of user, and user characteristics, which excavates, to be included but not
It is limited to, investment (speculation in stocks) preference of user is gone out according to user behavior analysis.User behavior, which is collected, to be included but is not limited to, and collects user
Operating habit, collection user to self-selected stock is to the operation flowing water of the propensity to investment, collection user of self-selected stock to self-selected stock, collection
The speculation in stocks mode of user to self-selected stock etc..In the intelligent control unit 3 according in the stock tag library 1 based on " thing "
The investment preference operation of user, selects weight and scoring highest stock in the stock label of body and the user characteristics unit 2
Recommended to client.Due to including that to establish the stock tag library of the stock similarity gene based on label, can establish
The related network to be quantized between all stocks.Due to including to the user characteristics based on user's self-selected stock excavation characterization factor
Single 2 yuan and due to including to according to characterization factor and stock similarity gene, the intelligence that individual character recommendation is carried out for user is adjusted
Unit 3 is controlled, the self-selected stock that can be added according to user, the stock for finding most relevant property in a network recommended, and according to working as
The weight of preceding focus and scene dynamics adjustment stock, the foundation as recommendation.Due to including performing and/or distributing calculating task
Distributed Calculation unit 4, it is possible to increase the operating efficiency of commending system, reduce processing time, improve handling capacity.
As preferred in the present embodiment, Distributed Calculation unit 4 uses：Slicing algorithm, message queue are based on
Hadoop Distributed Architecture.
As preferred in the present embodiment, refer to Fig. 2 is the stock tag library structural representation in Fig. 1, the stock
Tag library 1 includes：Basic tag unit 11, concepts tab unit 12, stock nature label unit 13, technical tag unit
14, the basic tag unit 11, the base attribute of stock is quantified as into basic label, the concepts tab unit 12,
The concept of stock is quantified as into concepts tab, the stock nature label unit 13, the characteristic of stock of stock to be quantified as
Characteristic of stock label, the technical tag unit 14, the technological side of stock is quantified as into technical label.For example interrogated with University of Science and Technology
Exemplified by flying then,
It may include in the basic tag unit 11, { deep bid, net profit>400000000 etc. }, the basic side is included but not
It is limited to：Earnings per share, growth, p/e ratio.
It may include in the concepts tab unit 12, { the artificial intelligence degree of association 33%, internet+degree of association 30%, nature
Language Processing degree of association 70% etc. },
It may include in the stock nature label unit 13, { characteristic of stock：70 points, leading stock etc. }
The technical tag unit 14 may include, { bottom away from etc. }.Technological side includes but is not limited to：Comprehensive various fingers
Mark, such as equal line, MACD, RSI, support level, pressure position etc..
As preferred in the present embodiment, Fig. 3 is another structural representation of stock tag library in Fig. 1, the stock mark
Label storehouse 1 includes：Basic tag unit 11, correlation tag unit 15, stock nature label unit 13, technical tag unit
14, the basic tag unit 11, the base attribute of stock is quantified as into basic label, the correlation tag unit
15, the correlation of stock is quantified as into concepts tab, the stock nature label unit 13, to by the characteristic of stock amount of stock
Turn to characteristic of stock label, the technical tag unit 14, the technological side of stock is quantified as into technical label.Such as with great
Exemplified by news science and technology then,
It may include in the basic tag unit 11, { shallow bid, net profit>20000000 }, the basic side is included but not
It is limited to：Earnings per share, growth, p/e ratio.
It may include in the correlation tag unit 15, and artificial intelligence 30%, industry 4.0 | 20% },
It may include in the stock nature label unit 13, { characteristic of stock：65 points }
The technical tag unit 14 may include, { MACD gold forks, upper 10 average daily lines of standing }.Technological side includes but unlimited
In：Comprehensive various indexs, such as equal line, MACD, RSI, support level, pressure position etc..
In certain embodiments, the basic tag unit 11, includes but is not limited to, daily market rumor collects, be newest
Personal share good news, newest personal share bad news, company's major event disclosure, focus Calendar of Petroleum Economic Events comment, share out bonus dispatching prompting,
Selected investment personal share list.
In certain embodiments, technical tag unit 14, include but is not limited to, the technical indicator of reflection Jie's change, walk
Gesture form and the combination of K lines etc..Due to thinking that the market behavior includes all information, then for macroscopical face, policy etc. because
Element can be ignored, and thinking price change has rule and history repeat itself.
As preferred in the present embodiment, Fig. 4 is the user characteristics cellular construction schematic diagram in Fig. 1, the user characteristics
Unit 2 includes：Operation flowing water collector unit 21, in the recent period speculation in stocks mode collector unit 22 in the recent period, the flowing water of operation in the recent period are collected
Unit 21, to collect the characteristic of user's water operation in certain period of time, the speculation in stocks mode collector unit in the recent period
22, the characteristic for mode of speculating in shares collecting user in certain period of time.It is described to refer to 1-3 days in certain period of time；3-5
My god；In the relatively short time interval of 5-9 days.The collector unit of operation flowing water in the recent period 21 can be by monitoring mouse pointer
Event, the operation flowing water of user is obtained, included but is not limited to, logged in, delete, checking, searching for, recording, collection etc..Used
Family speculation in stocks mode includes but is not limited to, and short operation, the operation of long line, characteristic of stock is strong, market value is high etc..
As preferred in the present embodiment, Fig. 5 is the user characteristics cellular construction journey schematic diagram in Fig. 1, and the user is special
Sign unit 2 includes：User data polymerized unit 23 and user data excavate unit 24, the user data polymerized unit 23, use
It is accustomed to entirety to be speculated in shares according to the user of definition, obtains user's speculation in stocks custom composition class, the user data excavation unit 24, use
To analyze speculation in stocks preference of the user in certain period of time.
Aggregation in the user data polymerized unit 23 usually as data mining the first step, first to whole client
Do and assemble, by custom partitioning in respective aggregation, then to each different aggregation, answer a question, possible effect is more preferable.
The processing procedure that the user data is excavated in unit 24 includes but is not limited to,
24.1 classification, select the training set for having divided class from data first, and maintenance data excavates on the training set
The technology of classification, establishes disaggregated model, classifies for the data of no classification.
24.2 estimation, estimation it is similar with classification, difference is, classified description be discrete variable output, and
Valuation handles the output of successive value；The classification of classification is to determine number, and the amount of valuation is uncertain.
24.3 prediction, by classification or valuation work, by classification or valuation draw model, the model be used for pair
The prophesy of known variables.
24.4 correlations are grouped or correlation rule, determine which thing will occur together.For example user is in purchase A-share
Meanwhile often buy B stocks, i.e. A=>B (correlation rule)；Again for example, user, every a period of time, can buy after B stocks are bought
A-share (sequence analysis).
24.1 clusters, it is that record is grouped, similar is recorded in an aggregation.The difference of cluster and classification is aggregation
Independent of the class pre-defined, it is not necessary to training set.
24.1 descriptions and visualization, are the representations to data mining results.
In certain embodiments, the user data excavates unit 24 and uses C4.5 categorised decision tree algorithms.
In certain embodiments, the user data excavates unit 24 and uses K-means clustering algorithms.
In certain embodiments, the user data excavates method of the unit 24 using the study of SVM supervised.
In certain embodiments, it is frequent using Apriori Mining Boolean Association Rules to excavate unit 24 for the user data
Item set algorithm.
In certain embodiments, the user data excavates unit 24 and uses EM greatest hope value-based algorithms.
In certain embodiments, the user data excavates unit 24 and uses pagerank algorithms.
In certain embodiments, the user data excavates unit 24 and uses Adaboost iterative algorithms.
Refer to Fig. 6 is the system structure diagram in one embodiment of the present invention, and system includes：Distributed Calculation list
Member 4, stock tag library 1, user characteristics unit 2, intelligent control unit 3, the stock tag library 1, label are based on to establish
Stock similarity gene, the user characteristics unit 2, to based on user's self-selected stock excavate characterization factor, it is described intelligence adjust
Unit 3 is controlled, according to characterization factor and stock similarity gene, individual character recommendation, the Distributed Calculation are carried out for user
Unit 4, to perform and/or distribute in above-mentioned stock tag library 1, above-mentioned user characteristics unit 2, above-mentioned intelligent control unit 3
Calculating task.As preferred in the present embodiment, in addition to：Focus unit, to as access stock market's current stock institute phase
Close the API of focus.Popular concept/focus in focus unit just can give the injection of related stock to speculate value, can lift phase
Close the rising space of stock.When certain stock belongs to some market focus, its share price will investee pursue.
The API of the related focus includes, there is provided the column information belonging to self-selected stock.
The API of the related focus includes, there is provided the big personal share information of amount of increase.
The API of the related focus includes, there is provided column amount of increase information.
The API of the related focus includes, and related various message, such as some columns are pierced by message face
Swash, can there is action suddenly, or be several personal shares among column because message is drawn high, to drive other personal shares, more concerns
Message has reference value.
The API of the related focus includes, and pays close attention to various conferencing informations, for example a certain period will have internet meeting, then
The stock of some internet columns can be recommended
The API of the related focus includes, the exchange hand of personal share.
Fig. 7 is the system structure diagram in another preferred embodiment of the present invention；System includes：Distributed Calculation unit 4,
Stock tag library 1, user characteristics unit 2, intelligent control unit 3, the stock tag library 1, to establish the stock based on label
Ticket similarity gene, the user characteristics unit 2, to excavate characterization factor, the intelligent control list based on user's self-selected stock
Member 3, according to characterization factor and stock similarity gene, individual character recommendation, the Distributed Calculation unit are carried out for user
4, to perform and/or distribute the meter in above-mentioned stock tag library 1, above-mentioned user characteristics unit 2, above-mentioned intelligent control unit 3
Calculation task.As preferred in the present embodiment, in addition to：Scene dynamics adjustment unit 6, to according to user to different stock
Operation, stock state scene corresponding to regulation.Stock state scene includes but is not limited to, and position in storehouse, opens a position, holds position, closing a position.Such as
User's first is delithted with stock A, and adds and hold position, and in face of the surging of recent artificial intelligence plate, then system can push away to user's first
Recommend equally be deep bid stock B.
As preferred in the present embodiment, refer to Fig. 8 is the intelligent control cellular construction schematic diagram in Fig. 1, the intelligence
Single 3 yuan, which can be regulated and controled, includes：Weight unit 31, score unit 32, the weight unit 31, to in the stock tag library
Stock similarity gene carries out weight distribution, the scoring unit 32, to carry out comprehensive weight scoring based on weight unit.Power
Weight unit 31 refers to carry out different degrees of assignment according to stock gene.The unit 32 that scores is given a mark according to assigned result.
Such as in weight unit 31, in stock label, limit-up gene：0.9 | dividend gene：0.8 | main force's gene：0.8|
Market value：0.7 | other
Then, comprehensive grading is in scoring unit 32：0.78.
Again for example, in weight unit 31, user A, preference leading stock：0.8 | preference mid-game stock：0.8 | preference artificial intelligence
Concept：0.88 etc. adds stock A simultaneously：+5；Enter personal share details+2 daily；Hold position：+10；Etc.
Then, comprehensive grading is in scoring unit 32：Comprehensive grading：0.76.
Refer to Fig. 9 is the method flow schematic diagram in one embodiment of the invention, a kind of intelligent stock in the present embodiment
Recommendation method, comprises the following steps：
Stock similarity gene of the step S1 foundation based on label/be based on user's self-selected stock excavation characterization factor,
Step S2 obtains the self-selected stock of user's selection,
Step S3 carries out individual character recommendation according to characterization factor and stock similarity gene for user.
Preferably, it is using slicing algorithm, message queue or distributed based on Hadoop in above-mentioned steps S1- steps S3
Framework.
In certain embodiments, above-mentioned steps S1 also includes：The base attribute of stock is quantified as basic label,
The concept of stock is quantified as concepts tab,
The characteristic of stock of stock is quantified as characteristic of stock label,
The technological side of stock is quantified as technical label.
In certain embodiments, above-mentioned steps S1 also includes：The base attribute of stock is quantified as basic label,
The correlation of stock is quantified as concepts tab,
The characteristic of stock of stock is quantified as characteristic of stock label,
The technological side of stock is quantified as technical label.
In certain embodiments, above-mentioned steps S1 also includes：
The characteristic of user's water operation in certain period of time is collected,
User is collected to speculate in shares in certain period of time the characteristic of mode.
In certain embodiments, above-mentioned steps S1 also includes：
Being speculated in shares according to the user of definition, it is overall to be accustomed to, and obtains user's speculation in stocks custom composition class,
Analyze speculation in stocks preference of the user in certain period of time.
In certain embodiments, the above method also includes：Increase as access stock market's current stock related focus the
Tripartite API.
In certain embodiments, the above method also includes：According to user to different Stock-operations, stock corresponding to regulation
State scene.
In certain embodiments, the step S3 also includes：
Weight distribution is carried out to the stock similarity gene in the stock tag library,
Comprehensive weight scoring is carried out based on weight unit.
Refer to Figure 10 is the method flow diagram in the present invention, including step is：
First, distributed computing framework is built, undertakes whole commending system user stock hobby, behavioural characteristic and preference,
The calculating task of the basic datas such as stock label so that the generation association between stock, and can be accumulated according to the long-term behavior of user
The unique speculation in stocks hobby gene of tired each user.
Secondly, stock tag library is established, i.e. every stock contains the composition of speculation in stocks gene, for example University of Science and Technology's news fly, limit-up base
Cause：0.9 | dividend gene：0.8 | main force's gene：0.8 | market value：0.7 | other；Comprehensive grading is：0.78.
Then, progress user behavior collection and feature mining, the user oriented custom that uses and speculate in shares, recent with user
It is foundation to propagandize flowing water and speculation in stocks mode, likes feature using polymerization and data digging method to extract the speculation in stocks of user in short-term,
Such as user's first, preference leading stock：0.8 | preference mid-game stock：0.8 | preference artificial intelligence concept：0.88 etc. adds stock simultaneously
Ticket A：+5；Enter personal share details+2 daily；Hold position：+10；Etc..Comprehensive grading：0.76
Human beings, the hobby feature of user is excavated, such as, user A likes short operation, preference leading stock, Gao Shi
The strong University of Science and Technology's news of value, characteristic of stock fly；
Finally, according to the rating matrix of user, in addition to the stock that user adds self-selected stock, selection in proposed algorithm
Fraction highest recommends user.
In user terminal, if user's first is delithted with University of Science and Technology, news fly, and add and hold position, meanwhile, in face of recent artificial intelligence plate
Block it is surging, system can to user's first recommend equally be deep bid stock A.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.Above-mentioned
In embodiment, software that multiple steps or method can be performed in memory and by suitable instruction execution system with storage
Or firmware is realized.If, and in another embodiment, can be with well known in the art for example, realized with hardware
Any one of row technology or their combination are realized：With the logic gates for realizing logic function to data-signal
Discrete logic, have suitable combinational logic gate circuit application specific integrated circuit, programmable gate array (PGA), scene
Programmable gate array (FPGA) etc..
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or the spy for combining the embodiment or example description
Point is contained at least one embodiment or example of the present invention.In this manual, to the schematic representation of above-mentioned term not
Necessarily refer to identical embodiment or example.Moreover, specific features, structure, material or the feature of description can be any
One or more embodiments or example in combine in an appropriate manner.
In general, the various embodiments of the disclosure can be with hardware or special circuit, software, logic or its any combination
Implement.Some aspects can be implemented with hardware, and some other aspect can be with firmware or software implementation, and the firmware or software can
With by controller, microprocessor or other computing devices.Although the various aspects of the disclosure be shown and described as block diagram,
Flow chart is represented using some other drawing, but it is understood that frame described herein, equipment, system, techniques or methods can
With in a non limiting manner with hardware, software, firmware, special circuit or logic, common hardware or controller or other calculating
Equipment or some combinations are implemented.
Although in addition, operation is described with particular order, this is understood not to require this generic operation with shown suitable
Sequence is performed or performed with generic sequence, or requires that all shown operations are performed to realize expected result.In some feelings
Under shape, multitask or parallel processing can be favourable.Similarly, begged for although the details of some specific implementations is superincumbent
By comprising but these are not necessarily to be construed as any restrictions to the scope of the present disclosure, but the description of feature is only pin in
To specific embodiment.Some features described in some embodiments of separation can also be held in combination in single embodiment
OK.Mutually oppose, the various features described in single embodiment can also in various embodiments be implemented separately or to appoint
The mode of what suitable sub-portfolio is implemented.
Claims (10)
- A kind of 1. intelligent stock commending system, it is characterised in that including：Distributed Calculation unit, stock tag library, user characteristics Unit, intelligent control unit,The stock tag library, to establish the stock similarity gene based on label,The user characteristics unit, to excavate characterization factor based on user's self-selected stock,The intelligent control unit, according to characterization factor and stock similarity gene, individual character recommendation is carried out for user,The Distributed Calculation unit, to perform and/or distribute above-mentioned stock tag library, above-mentioned user characteristics unit, above-mentioned Calculating task in intelligent control unit.
- 2. intelligent stock commending system according to claim 1, it is characterised in that the stock tag library includes：Substantially Tag unit, concepts tab unit, stock nature label unit, technical tag unit,The basic tag unit, the base attribute of stock is quantified as into basic label,The concepts tab unit, the concept of stock is quantified as into concepts tab,The stock nature label unit, the characteristic of stock of stock is quantified as into characteristic of stock label,The technical tag unit, the technological side of stock is quantified as into technical label.
- 3. intelligent stock commending system according to claim 1, it is characterised in that the stock tag library includes：Substantially Tag unit, correlation tag unit, stock nature label unit, technical tag unit,The basic tag unit, the base attribute of stock is quantified as into basic label,The correlation tag unit, the correlation of stock is quantified as into concepts tab,The stock nature label unit, the characteristic of stock of stock is quantified as into characteristic of stock label,The technical tag unit, the technological side of stock is quantified as into technical label.
- 4. intelligent stock commending system according to claim 1, it is characterised in that the user characteristics unit includes：Closely Phase operation flowing water collector unit, in the recent period speculation in stocks mode collector unit,The operation flowing water collector unit in the recent period, to collect the characteristic of user's water operation in certain period of time,The speculation in stocks mode collector unit in the recent period, the characteristic for mode of speculating in shares collecting user in certain period of time.
- 5. the intelligent stock commending system according to claim 1 or 4, it is characterised in that the user characteristics unit includes： User data polymerized unit and user data excavate unit,The user data polymerized unit, it is accustomed to entirety to be speculated in shares according to the user of definition, obtains user's speculation in stocks custom composition Class,The user data excavates unit, to analyze speculation in stocks preference of the user in certain period of time.
- 6. intelligent stock commending system according to claim 1, it is characterised in that also include：Focus unit, to conduct Access stock market's current stock related focus API.
- 7. intelligent stock commending system according to claim 1, it is characterised in that also include：Scene dynamics adjustment unit, To according to user to different Stock-operations, stock state scene corresponding to regulation.
- 8. intelligent stock commending system according to claim 1, it is characterised in that the intelligent control unit includes：Power Weight unit, score unit,The weight unit, to carry out weight distribution to the stock similarity gene in the stock tag library,The scoring unit, to carry out comprehensive weight scoring based on weight unit.
- 9. a kind of intelligent stock recommends method, it is characterised in that comprises the following steps：Stock similarity gene of the foundation based on label/be based on user's self-selected stock excavation characterization factor,The self-selected stock of user's selection is obtained,According to characterization factor and stock similarity gene, individual character recommendation is carried out for user.
- 10. intelligent stock according to claim 9 recommends method, it is characterised in that in the above method using slicing algorithm, Message queue or based on Hadoop Distributed Architecture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710633479.6A CN107481143A (en) | 2017-07-28 | 2017-07-28 | A kind of intelligent stock commending system and implementation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710633479.6A CN107481143A (en) | 2017-07-28 | 2017-07-28 | A kind of intelligent stock commending system and implementation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107481143A true CN107481143A (en) | 2017-12-15 |
Family
ID=60597763
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710633479.6A Pending CN107481143A (en) | 2017-07-28 | 2017-07-28 | A kind of intelligent stock commending system and implementation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107481143A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108510393A (en) * | 2018-02-12 | 2018-09-07 | 岭尚（上海）科技发展有限公司 | Intelligence based on big data, which is thrown, cares for system and method |
| CN109584061A (en) * | 2018-10-17 | 2019-04-05 | 深圳壹账通智能科技有限公司 | Information generating method, device, computer equipment and storage medium |
| CN109685656A (en) * | 2018-12-21 | 2019-04-26 | 天津知柿信息科技有限公司 | A kind of stock market's tendency intelligent Forecasting and system |
| CN110209944A (en) * | 2019-06-10 | 2019-09-06 | 上海时廊人工智能科技有限公司 | A kind of stock analysis teacher recommended method, device, computer equipment and storage medium |
| CN111222993A (en) * | 2020-01-03 | 2020-06-02 | 中国工商银行股份有限公司 | Fund recommendation method and device |
| CN111611484A (en) * | 2020-05-13 | 2020-09-01 | 湖南福米信息科技有限责任公司 | Stock recommendation method and system based on article attribute identification |
| CN113034278A (en) * | 2021-03-10 | 2021-06-25 | 深圳前海嘉可能信息科技有限公司 | Stock software-based 3-buying alarm method |
| TWI747421B (en) * | 2020-08-05 | 2021-11-21 | 元大證券投資信託股份有限公司 | Investment portfolio selection system |
| CN115577093A (en) * | 2022-05-26 | 2023-01-06 | 深圳格隆汇信息科技有限公司 | AI analysis method and system of financial information |
| WO2023284531A1 (en) * | 2021-07-13 | 2023-01-19 | 富途网络科技(深圳)有限公司 | Target recommendation method and apparatus, and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105488662A (en) * | 2016-01-07 | 2016-04-13 | 北京歌利沃夫企业管理有限公司 | Bi-directional recommendation-based online recruitment system |
| CN105989540A (en) * | 2015-02-10 | 2016-10-05 | 上海华颂软件科技有限公司 | Stock selecting method and system |
| CN106982150A (en) * | 2017-03-27 | 2017-07-25 | 重庆邮电大学 | A kind of mobile Internet user behavior analysis method based on Hadoop |
-
2017
- 2017-07-28 CN CN201710633479.6A patent/CN107481143A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105989540A (en) * | 2015-02-10 | 2016-10-05 | 上海华颂软件科技有限公司 | Stock selecting method and system |
| CN105488662A (en) * | 2016-01-07 | 2016-04-13 | 北京歌利沃夫企业管理有限公司 | Bi-directional recommendation-based online recruitment system |
| CN106982150A (en) * | 2017-03-27 | 2017-07-25 | 重庆邮电大学 | A kind of mobile Internet user behavior analysis method based on Hadoop |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108510393A (en) * | 2018-02-12 | 2018-09-07 | 岭尚（上海）科技发展有限公司 | Intelligence based on big data, which is thrown, cares for system and method |
| CN109584061A (en) * | 2018-10-17 | 2019-04-05 | 深圳壹账通智能科技有限公司 | Information generating method, device, computer equipment and storage medium |
| CN109685656A (en) * | 2018-12-21 | 2019-04-26 | 天津知柿信息科技有限公司 | A kind of stock market's tendency intelligent Forecasting and system |
| CN110209944A (en) * | 2019-06-10 | 2019-09-06 | 上海时廊人工智能科技有限公司 | A kind of stock analysis teacher recommended method, device, computer equipment and storage medium |
| CN111222993A (en) * | 2020-01-03 | 2020-06-02 | 中国工商银行股份有限公司 | Fund recommendation method and device |
| CN111611484A (en) * | 2020-05-13 | 2020-09-01 | 湖南福米信息科技有限责任公司 | Stock recommendation method and system based on article attribute identification |
| CN111611484B (en) * | 2020-05-13 | 2023-08-11 | 湖南微步信息科技有限责任公司 | Stock recommendation method and system based on article attribute identification |
| TWI747421B (en) * | 2020-08-05 | 2021-11-21 | 元大證券投資信託股份有限公司 | Investment portfolio selection system |
| CN113034278A (en) * | 2021-03-10 | 2021-06-25 | 深圳前海嘉可能信息科技有限公司 | Stock software-based 3-buying alarm method |
| WO2023284531A1 (en) * | 2021-07-13 | 2023-01-19 | 富途网络科技(深圳)有限公司 | Target recommendation method and apparatus, and storage medium |
| CN115577093A (en) * | 2022-05-26 | 2023-01-06 | 深圳格隆汇信息科技有限公司 | AI analysis method and system of financial information |
| CN115577093B (en) * | 2022-05-26 | 2023-05-16 | 深圳格隆汇信息科技有限公司 | AI analysis method and system for financial information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107481143A (en) | A kind of intelligent stock commending system and implementation method | |
| Han et al. | Category role aided market segmentation approach to convenience store chain category management | |
| Chen | Multiple criteria ABC inventory classification using two virtual items | |
| Chen et al. | Design and implementation of bank CRM system based on decision tree algorithm | |
| Oztekin | Creating a marketing strategy in healthcare industry: a holistic data analytic approach | |
| Farajian et al. | Mining the banking customer behavior using clustering and association rules methods | |
| Hsu | An integrated portfolio optimisation procedure based on data envelopment analysis, artificial bee colony algorithm and genetic programming | |
| Wilson | Urban and regional dynamics from the global to the local: Hierarchies,‘DNA’, and ‘genetic’planning | |
| Vaganov et al. | Forecasting purchase categories with transition graphs using financial and social data | |
| Abidar et al. | Customer segmentation with machine learning: New strategy for targeted actions | |
| Chen et al. | Implementation of penalized survival models in churn prediction of vehicle insurance | |
| Bhaskar et al. | A fuzzy mathematical programming approach for cross-sell optimization in retail banking | |
| Dostál et al. | Business applications of fuzzy logic | |
| Zou et al. | Precise Marketing of E‐Commerce Products Based on KNN Algorithm | |
| Li | Analysis of e-commerce customers' shopping behavior based on data mining and machine learning | |
| CN109426891A (en) | It is a kind of high to send the forecasting system and method for turning electronic banking product | |
| Bhaduri et al. | Advanced Business Analytics | |
| Ang et al. | Employee turnover prediction by machine learning techniques | |
| Mahjoub et al. | A hybrid model for customer credit scoring in stock brokerages using data mining approach | |
| Werb et al. | Predictive modeling in marketing | |
| Jabir et al. | Big data analytics for strategic and operational decisions | |
| Yuan et al. | Research on mobile internet user behavior preference based on support vector machine | |
| Chen et al. | GameLifeVis: visual analysis of behavior evolutions in multiplayer online games | |
| Bruckhaus | Collective intelligence in marketing | |
| Kornilov | Assessing organizational efficiency under macroeconomic uncertainty in decision support systems: Ensemble methods in machine learning with two-stage nonparametric efficiency models |
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: 20171215 |
|
| RJ01 | Rejection of invention patent application after publication |