CN106529726A - Method of performing classification and recommendation based on stock prediction trends - Google Patents
Method of performing classification and recommendation based on stock prediction trends Download PDFInfo
- Publication number
- CN106529726A CN106529726A CN201611008483.5A CN201611008483A CN106529726A CN 106529726 A CN106529726 A CN 106529726A CN 201611008483 A CN201611008483 A CN 201611008483A CN 106529726 A CN106529726 A CN 106529726A
- Authority
- CN
- China
- Prior art keywords
- stock
- tendency
- prediction
- classification
- kinds
- 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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Finance (AREA)
- Probability & Statistics with Applications (AREA)
- Accounting & Taxation (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Fuzzy Systems (AREA)
- Technology Law (AREA)
- Software Systems (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Physics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method of performing classification and recommendation based on stock prediction trends. The method comprises the steps that first, according to recent trends of to-be-predicted stocks, historical data of all the stocks is searched for and matched, three trends which are most similar are obtained, and three trend prediction results are generated based on the later trend of the similar trends; second, each trend is classified into five classifications including all the way up, down after up, up after down, all the way down and bad, and three prediction trends of each stock correspond to three classifications; third, the uniformity of the classifications is judged, when the classifications of the three trends are all different, it is determined that the stock trend is bad, and classification judgment is performed according to the majority principle of the remaining stocks; finally, first stocks with the maximum prediction trend rise range are selected from the stocks with identical classification tags to form a list for surge recommendation, and the stocks with the maximum prediction trend drop range form a list for plunge early-warning.
Description
Technical field
The present invention relates to stock certificate data digging technology field, is classified based on Prediction of Stock Index tendency more particularly, to one kind
Method with recommending.
Background technology
Combined influence of the stock price by many factors, with changing all the time, it is intricate the characteristics of, while with financial ring
Border is closely related, and data variation is extremely numerous and diverse, and change has extremely strong randomness, it is difficult to set up accurate mathematical model, therefore right
Stock is predicted the hot issue of always financial field research.
In terms of shares changing tendency prediction, technical Analysis method mainly has statistical method and artificial intelligence approach.Time sequence
Row analytic process is main a kind of mode in statistical method, wherein have the random walk model, autoregression of conventional model are moved
Dynamic averaging model, homogeneous nonstationary model and Markov-chain model etc..In recent years, artificial intelligence technology has obtained very big sending out
Application of the neutral net in Prediction of Stock Price has been carried out in exhibition, many scholars.
In terms of stock classification, the algorithm that can be used to classify has decision tree, Bayes's classification, neutral net, supporting vector
Machine etc..Although established various theories of algorithm can explain some phenomenons in stock market at present, practical application effect is simultaneously paid no attention to
Think.
The content of the invention
The invention discloses a kind of method classified based on Prediction of Stock Index tendency and recommended.Method is first with to be predicted
The recent tendency of stock, the historical data of the search all stocks of matching obtain three most like tendency sections, and based on it is similar walk
The later stage tendency of gesture section generates three kinds of forward prediction results；Secondly every kind of tendency is classified, after going up all the way, first rising
Fall, first fall rise afterwards, drop, five classes of bad determination all the way, three kinds of every stock prediction tendencies three kinds of classification of correspondence；Then to dividing
The concordance of class is judged that shares changing tendency is bad determination when the classification of three kinds of trend is all different, and remaining is according to majority
Principle carries out classification judgement；It is last that former maximum stocks of prediction tendency amount of increase are selected in the on all four stock of tag along sort
List of the ticket as recommendation of rising sharply, predicts the maximum list as early warning of slumping of tendency drop range.
The inventive method is a kind of recommendation method, and the foundation of recommendation is the prediction tendency based on its following a period of time, is pushed away
The stock recommended is in addition to requiring next to have larger amount of increase, while requiring with higher accuracy.
The step of the inventive method, is as follows：
（1）The tendency section similar to the recent tendency of stock to be predicted is searched in the historical data of all stocks, and based on similar
The later stage tendency of tendency section generates 3 kinds of prediction tendencies of stock；
（2）Every kind of prediction tendency of every stock is classified；
（3）3 kinds of prediction tendency classification to every stock carry out concordance statistics, determine stock final classification；
（4）Stock recommendation is carried out in the on all four stock of tag along sort.
Wherein, step (1) will mainly obtain 3 kinds of prediction tendencies of stock to be predicted.Specifically include following 5 sub-steps
Suddenly：
（1）Data prediction operation is carried out to all stocks.Calculate opening price, closing price, increasing of the highest price relative to closing price
Amount of decrease degree；
（2）Loading stock certificate data to be predicted and historical data to be matched.Stock certificate data to be predicted is the recent L day of trade
Amount of increase and amount of decrease data；Historical data to be matched is certain time point of every stock to come the front L transaction of current trading day
The data of day, and corresponding trade date；
（3）The recent tendency section of stock to be predicted and the historical data of every stock are moved into weighted registration by turn, per only
Stock only records a best match section；
（4）In the best match section of all stocks, the overall situation is compared, and obtains most 3 tendency sections of matching again；
（5）Tendency based on the similar tendency section later stage generates 3 kinds of prediction tendencies of stock to be predicted.
Wherein, every kind of prediction tendency of every stock is classified in step (2), is specifically divided into 5 classifications, its title
With correspondence code it is：" going up all the way "：" 1 ", " first rise and fall afterwards "：" 2 ", " first fall and rise afterwards "：" 3 ", " dropping all the way "：" 4 ", " no
It is good to determine "：“5”.The determination process of every kind of tendency classification is：
（1）The position corresponding to the highest of tendency, lowest price is obtained, remembers that position that maximum is located is HighPos, minima
The position at place is LowPos；
（2）The classification of prediction tendency is determined according to the various combination of HighPos and LowPos positions；
For 3 kinds of forward prediction results of every stock, 3 classification of relative this stock will be obtained.
Wherein, three kinds of prediction tendency classification to every stock of step (3) carry out concordance statistics, specially：For
3 kinds of prediction tendencies of every stock correspond to 3 classification, and the classification classified only has 4 kinds, therefore the combination of its class categories has
3 kinds of situations：3 classification is completely the same, unanimously another difference, 3 classification are all different for 2 classification.For this 3 kinds of situations,
3 all different stock final classifications of classification are judged as " bad determination " that other two is classified according to most Voting principles
The final classification of the stock will be defined as with identical classification.
Wherein, step (4) carries out stock recommendation in the on all four stock of tag along sort, specially：To recommending
It is divided into two kinds, one kind is " recommendation of rising sharply ", and another kind is " early warning of slumping ".With reference to 3 kinds of prediction tendencies originally to stock, by 3
Plant the amount of increase prediction that the maximum amount of increase in prediction tendency is averaged as the stock；3 kinds are predicted the corresponding Maximum Drawdown of tendency
The drop range prediction averaged as the stock；3 kinds of tendency tag along sorts of acquisition are completely the same and amount of increase predicts the former of maximum
The list of stock as recommendation of rising sharply, the list as early warning of slumping of the maximum former stocks of drop range prediction.
Description of the drawings
Fig. 1 is the flow chart that the present invention is classified based on Prediction of Stock Index tendency and recommended method.
Fig. 2 is the example of recommendation results list of rising sharply during stock is recommended.
Fig. 3 is the example of early warning the results list of slumping during stock is recommended.
Specific embodiment
Below in conjunction with the accompanying drawings and example, the present invention is described in detail.
The inventive method carries out the forward prediction of short-term first to stock, then according to prediction walk will definitely stock be divided into 5
Individual classification, i.e. " all the way go up ", " first rise and fall afterwards ", " first fall and rise afterwards ", " dropping all the way ", " bad determination " 5 classes, finally according to point
The consistent implementations of class carry out the recommendation of stock.
Assume that stock list is S, S=[S1, S2,…,Si,…,Sn], n is the quantity of stock in stock pond, such as in China
The quantity of city's stock or the quantity of listed stock of the U.S..
First, the short-term forecast of stock.
The step carries out the prediction of short-term to stock, so as to obtain the Short Term of stock.When Short Term includes one section
Between opening price, closing price, highest price, lowest price, amount of increase and amount of decrease etc..For every stock, it is assumed that stock to be predicted is Sm,m=
1 ..., the concrete prediction steps of n are as follows.
（1）Stock certificate data pretreatment.
Assume for every stock in stock list S has following data field：Opening price Open, closing price Close,
Highest price High, lowest price Low, amount of increase and amount of decrease Change etc., wherein closing price Close weigh price again for front.Data prediction mistake
Journey mainly calculates the increase and decrease amplitude relative to same day closing price Close, and adds newer field.Calculative field includes out
Disk valency Open, highest price High and lowest price Low, it is specific as follows：
Opening price increase and decrease amplitude StdOpen=100* (Open-Close)/Close;
Highest price increase and decrease amplitude StdHigh=100* (High-Close)/Close;
Lowest price increase and decrease amplitude StdLow=100* (Low-Close)/Close.
（2）The loading of stock certificate data.
The step mainly completes the set-up procedure of data, and the recent friendship of stock to be predicted is obtained from original issue stock data base
The historical trading day data of easy day data and stock to be matched.
2.1 obtain stock S to be predictedmThe nearly L day of trade amount of increase and amount of decrease data, formed an array, be designated as A,
A=[a1,a2,…,ai,…aL]
Wherein, aiRepresent the amount of increase and amount of decrease of the day of trade of nearly L-i.a0Represent the amount of increase and amount of decrease of current trading day.The value of L must at least be accorded with
The requirement of the short-term forecast day of trade is closed, such as>=20.
Every stock in 2.2 couples of S, since obtaining certain time point（Such as on January 1st, 2005）, arrive current trading day
The data of the front L day of trade, form another array, are designated as Bi,i=[1,n],
Bi=[bi1,bi2,…,bij,…bik]
Wherein, bijRepresent the amount of increase and amount of decrease of the j positions correspondence day of trade of i-th stock.Each BiLength k be not necessarily equal
, because there is the impact of the factors such as suspension in the middle of stock；B is recorded simultaneouslyijTrade date, be designated as another array Ci,i=[1,
n],
Ci=[ci1,ci2,…,cij,…cik]
Wherein, cijRepresent the j positions correspondence trade date of i-th stock.
（3）Mobile weighted registration between stock.
The step mainly completes stock to be predicted and certain matching process only between stock to be matched, and acquires and treat
The minima of matching stock matching and Corresponding matching date.Matching process is in A and BiLaunch, the length of A is L, BiLength be
K, K>=L, concrete matching process are as follows.
3.1 with step-length as 1, circulates from BiThe middle amount of increase and amount of decrease data for obtaining length L, are designated as B.
3.2 A and B is compared, and obtains matching value.The comparison procedure of A, B flexibly can be carried out, with reference to following several sides
Formula is compared,
A the amount of increase and amount of decrease of () A, B relevant positions directly subtracts each other and takes absolute value；
B () is weighted to position, higher the closer to current trading day weights；
C () considers the same tropism weighting of A, B amount of increase and amount of decrease；
D () is compared after A, B amount of increase and amount of decrease is normalized to [0,1] interval.
The minima of 3.3 matching values for recording all comparisons and its corresponding matching date, as A and BiMatching is most
Termination fruit, is designated as Pi, i=[1,n],
Pi=[Vi,Di]
Wherein, ViRepresent A and BiSmallest match value, DiRepresent the smallest match value corresponding matching date.
（4）Obtain global most like tendency section.
The P of 4.1 pairs of arrays PiNumerical value is ranked up from small to large, and minimum front 3 P of numerical value are obtained after sequenceiAnd its correspondence
Di, form new array Mt ,t=[1,T], Mt=[Pt,Dt].Here T=3.
4.2 obtain M respectivelytN number of day of trade (N is typically taken as 20-30) data after the matching date of middle correspondence stock,
The data of acquisition include amount of increase and amount of decrease Change, and each increase and decrease amplitude field increased newly in step 1, here as StdOpen,
StdHigh, StdLow etc..
4.3 closing prices Close (front multiple power) for obtaining stock current trading day to be predicted.
（5）Stock Prediction tendency result is generated.
Using current trading day closing price Close of stock to be predicted as " the previous day of trade " benchmark data, with reference to similar
Tendency data after the stock matching date, calculating are reduced into the tendency data of Stock Prediction to be predicted.Array MtMiddle correspondence
The tendency of each similar stock will all be reduced into a kind of short-term forecast result, detailed process is as follows.
Data Close of stock current trading day to be predicted are set to previous day of trade lastClose results by 5.1；Obtain
Tendency data after a certain similar stock matching date.
The prediction data of 5.2 reduction next days of trade, reduction process is：
Next day of trade closing price Close=lastClose+lastClose*Change/100;
Next day of trade opening price Open=Close+Close*StdOpen/100;
Next day of trade highest price High=Close+Close*StdHigh/100;
Next day of trade lowest price Low=Close+Close*StdLow/100.
Closing price Close of current trading day is set to previous day of trade lastClose results by 5.3.
5.4 repeat 5.2,5.3 steps, until restoring desired whole day of trade prediction data.
Above step will obtain predicting the outcome based on a certain similar shares changing tendency.3 most phases being the previously calculated
3 Short Terms that stock to be predicted is obtained are predicted the outcome R like stock.The R that predicts the outcome of every kind of tendencyt(t=1,23) by
Four arrays composition, i.e. Opens, Closes, Highs, Lows, represent respectively the daily opening quotation of anticipation trend, closing quotation, highest,
Lowest price.The length of array is all L, takes L=30 here, subsequently carries out tendency classification to stock to facilitate.
2nd, the prediction tendency of stock is classified.
The classification results of tendency are predefined, are divided into 5 classifications, and its title and correspondence code are：" going up all the way "：
" 1 ", " first rise and fall afterwards "：" 2 ", " first fall and rise afterwards "：" 3 ", " dropping all the way "：" 4 ", " bad determination "：“5”.For every stock
3 kinds of forward prediction results, 3 classification of relative this stock will be obtained.For a kind of prediction tendency Rt(t=1,23), its point
The determination process of class is as follows.
（1）Obtain the position corresponding to the highest of tendency, lowest price.
1.1 in RtHighs highest price arrays in by comparing its numerical value, obtain the position that maximum is located, be designated as
HighPos；Amount of increase Up of the maximum relative to stock current trading day closing price Close to be predicted is calculated simultaneouslyt。
1.2 in RtLows lowest price arrays in by comparing its numerical value, obtain the position that minima is located, be designated as
LowPos；While drop range Down of the calculated minimum relative to stock current trading day Close to be predictedt。
（2）Classified according to position grouping.
If 2.1 HighPos<LowPos, is downward trend, and the process of determining whether is：
If LowPos>15, HighPos<5, then classification is 4, is dropped all the way；
If LowPos>15, HighPos>=5, then classification is 2, first rises and falls afterwards；
If LowPos<=15, HighPos<5, then classification is 3, first falls and rises afterwards；
If LowPos<=15, HighPos>=5, then classification is 2, first rises and falls afterwards.
If 2.2 HighPos>=LowPos, is ascendant trend, and the process of determining whether is：
If HighPos>15, LowPos<5, then classification is 1, is gone up all the way；
If HighPos>15, LowPos>=5, then classification is 3, first falls and rises afterwards；
If HighPos<=15, LowPos<5, then classification is 2, first rises and falls afterwards；
If HighPos<=15, LowPos>=5, then classification is 3, first falls and rises afterwards.
For every kind of prediction tendency RtAs above classification is carried out all, it should be noted that every kind of tendency has clearly at present
Classification, temporarily also there is no the 5th class of " bad determination ".
3rd, stock concordance statistics and classification.
For 3 kinds of prediction tendencies of every stock correspond to 3 classification, the classification of classification only has 4 kinds at present.Class categories
Combination have 3 kinds of situations：3 classification is completely the same, unanimously another difference, 3 classification are all different for 2 classification.For this 3
3 all different stock final classifications of classification, the situation of kind, is judged as " bad determination " that other two classification is according to most throwings
Ticket principle will be defined as the final classification of the stock with identical classification.So far, have one by every stock of the step
Individual well-determined correspondence is classified.
4th, stock recommendation is carried out with reference to the amount of increase and amount of decrease of anticipation trend.
Recommendation is divided into two kinds, and one kind is " recommendation of rising sharply ", and another kind is " early warning of slumping ".By 3 kinds of prediction tendencies of stock
Corresponding 3 kinds of predictions maximum amount of increase UptThe amount of increase prediction averaged as the stock.By 3 kinds of prediction tendency correspondences of stock
3 kinds prediction Maximum Drawdowns DowntThe drop range prediction averaged as the stock.
The amount of increase of all stocks is predicted the outcome h before taking out after being sorted from big to small, while requiring these stocks
The label of 3 classification results of corresponding 3 kinds of trend is completely the same.Result of these stocks as " recommendation of rising sharply ".
The drop range of all stocks is predicted the outcome h before taking out after absolute value is sorted from big to small, while requiring this
The label of a little stock 3 classification results of corresponding 3 kinds of trend is completely the same.Result of these stocks as " early warning of slumping ".
In sum, the invention discloses a kind of method classified based on Prediction of Stock Index tendency and recommended.Method is pushed away
The foundation recommended is the prediction tendency based on stock a period of time in future, and it is complete that the stock of recommendation is taken from three kinds of tendency tag along sorts
Complete consistent stock, reliability are higher.
The inventive method is similarly applied to the data that security class has time serieses feature, such as fund, futures etc..Cause
This, although disclosing the specific embodiments and the drawings of the present invention for the purpose of illustration, its object is to help understand that the present invention's is interior
Hold and implement according to this, but it will be appreciated by those skilled in the art that：In the essence without departing from claim of the invention and appended
In god and scope, various replacements, to change and modifications all be impossible.Therefore, the present invention should not be limited to most preferred embodiment and
Accompanying drawing disclosure of that.Presently disclosed embodiment should be understood illustrative in all respects rather than which be claimed
Scope restriction.
Claims (6)
1. a kind of method classified based on Prediction of Stock Index tendency and recommended, it is characterised in that methods described includes following step
Suddenly：
（1）The tendency section similar to the recent tendency of stock to be predicted is searched in the historical data of all stocks, and based on similar
The later stage tendency of tendency section generates 3 kinds of prediction tendencies of stock；
（2）Every kind of prediction tendency of every stock is classified；
（3）3 kinds of prediction tendency classification to every stock carry out concordance statistics, determine stock final classification；
（4）Stock recommendation is carried out in the on all four stock of tag along sort.
2. the method classified based on Prediction of Stock Index tendency and recommended according to claim 1, it is characterised in that per only
The generating process of Prediction of Stock Index tendency, be by comparing with all stock global registrations after, by most like 3 tendencies for obtaining
The later stage tendency of section generates 3 kinds of prediction tendencies of stock to be predicted.
3. the method classified based on Prediction of Stock Index tendency and recommended according to claim 1, it is characterised in that to every
The classification of every kind of prediction tendency of stock, is the combination based on tendency highest price and lowest price position, will predict tendency
Be divided into all the way go up, first rise fall afterwards, first fall rise afterwards, drop four classes all the way.
4. the method classified based on Prediction of Stock Index tendency and recommended according to claim 1, it is characterised in that to stock
The final classification of ticket is carried out after concordance statistics by the classification to its 3 kinds prediction tendencies, and stock is divided into rise all the way, elder generation
Fall after rising, first fall rise afterwards, all the way drop, it is bad determination five classes.
5. the method classified based on Prediction of Stock Index tendency and recommended according to claim 1, it is characterised in that in stock
In the recommendation of ticket, selected in the on all four stock of tag along sort, while recommend to be divided into rise sharply recommending and big
Fall early warning, this causes the stock recommended to have higher credibility.
6. the concordance statistical classification process of stock according to claim 4, it is characterised in that stock has 3 predictions to walk
Gesture, and every kind of tendency has 4 kinds of possible classifications, this causes exactly only have 3 kinds of situations in concordance combination：3 classification complete
Cause, unanimously another difference, 3 classification are all different for 2 classification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611008483.5A CN106529726A (en) | 2016-11-16 | 2016-11-16 | Method of performing classification and recommendation based on stock prediction trends |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611008483.5A CN106529726A (en) | 2016-11-16 | 2016-11-16 | Method of performing classification and recommendation based on stock prediction trends |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106529726A true CN106529726A (en) | 2017-03-22 |
Family
ID=58352117
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611008483.5A Pending CN106529726A (en) | 2016-11-16 | 2016-11-16 | Method of performing classification and recommendation based on stock prediction trends |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106529726A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107480819A (en) * | 2017-08-09 | 2017-12-15 | 灯塔财经信息有限公司 | A kind of method and device of data analysis |
| CN108280763A (en) * | 2018-01-23 | 2018-07-13 | 北京网信云服信息科技有限公司 | Stock market data processing method and relevant device |
| CN108765146A (en) * | 2018-04-25 | 2018-11-06 | 武汉灯塔之光科技有限公司 | The method and apparatus that a kind of basis has tracing pattern selection specific curves stock |
| CN113902558A (en) * | 2021-10-12 | 2022-01-07 | 深圳格隆汇信息科技有限公司 | Information recommendation method and related product |
-
2016
- 2016-11-16 CN CN201611008483.5A patent/CN106529726A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107480819A (en) * | 2017-08-09 | 2017-12-15 | 灯塔财经信息有限公司 | A kind of method and device of data analysis |
| CN108280763A (en) * | 2018-01-23 | 2018-07-13 | 北京网信云服信息科技有限公司 | Stock market data processing method and relevant device |
| CN108765146A (en) * | 2018-04-25 | 2018-11-06 | 武汉灯塔之光科技有限公司 | The method and apparatus that a kind of basis has tracing pattern selection specific curves stock |
| CN113902558A (en) * | 2021-10-12 | 2022-01-07 | 深圳格隆汇信息科技有限公司 | Information recommendation method and related product |
| CN113902558B (en) * | 2021-10-12 | 2022-06-24 | 深圳格隆汇信息科技有限公司 | Information recommendation method and related product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109657805B (en) | Hyper-parameter determination method, device, electronic equipment and computer readable medium | |
| Cheng et al. | Fuzzy time series forecasting based on fuzzy logical relationships and similarity measures | |
| Eslami et al. | A data ensemble approach for real-time air quality forecasting using extremely randomized trees and deep neural networks | |
| CN106529726A (en) | Method of performing classification and recommendation based on stock prediction trends | |
| Faritha Banu et al. | Artificial intelligence based customer churn prediction model for business markets | |
| Guo et al. | Integration of semi-fuzzy SVDD and CC-Rule method for supplier selection | |
| CN110675029A (en) | Dynamic management and control method and device for commercial tenant, server and readable storage medium | |
| CN111210347A (en) | Transaction risk early warning method, device, equipment and storage medium | |
| CN113762579A (en) | Model training method and device, computer storage medium and equipment | |
| Özlem et al. | Predicting cash holdings using supervised machine learning algorithms | |
| Shu et al. | Perf-al: Performance prediction for configurable software through adversarial learning | |
| Loke et al. | Portfolio optimization problem: a taxonomic review of solution methodologies | |
| Wu et al. | Ensemble model of intelligent paradigms for stock market forecasting | |
| Anderies et al. | Telekom-net: The embedded bi-lstm and expert knowledge model for stock forecasting and suggestion | |
| Korovin et al. | The application of evolutionary algorithms in the artificial neural network training process for the oilfield equipment malfunctions’ forecasting | |
| Kanwal et al. | An attribute weight estimation using particle swarm optimization and machine learning approaches for customer churn prediction | |
| CN106780020A (en) | A kind of stock trend analysis and share-selecting method based on strong matching sequence | |
| Xiong et al. | L-RBF: A customer churn prediction model based on lasso+ RBF | |
| Song et al. | Maize seed appearance quality assessment based on improved Inception-ResNet | |
| WO2022162839A1 (en) | Learning device, learning method, and recording medium | |
| Chang et al. | Short-Term Stock Price-Trend Prediction Using Meta-Learning | |
| Shahvaroughi Farahani | Prediction of interest rate using artificial neural network and novel meta-heuristic algorithms | |
| Mandli et al. | Selection of most relevant features from high dimensional data using ig-ga hybrid approach | |
| Mostafa et al. | Estimate of stochastic model parameter of exchange rate using machine learning techniques | |
| Jain et al. | Steel faults diagnosis using predictive analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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: 20170322 |
|
| RJ01 | Rejection of invention patent application after publication |