CN107992601A - Trend prediction analysis method, equipment and storage medium - Google Patents
Trend prediction analysis method, equipment and storage medium Download PDFInfo
- Publication number
- CN107992601A CN107992601A CN201711337647.3A CN201711337647A CN107992601A CN 107992601 A CN107992601 A CN 107992601A CN 201711337647 A CN201711337647 A CN 201711337647A CN 107992601 A CN107992601 A CN 107992601A
- Authority
- CN
- China
- Prior art keywords
- target
- attention rate
- curve
- user
- moos index
- 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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of trend prediction analysis method, equipment and storage medium, method includes：Access window is established, the window includes multiple target quotation informations and information；The access information of user is obtained, user's visit capacity in each target unit interval is counted according to user access information, the access information includes time point, corresponding access target access times and the corresponding user identifier of access behavior；Calculate attention rate and moos index of each target within the unit interval；Each target attention rate curve and moos index curve are generated, according to the attention rate curve and the tendency in moos index curve prediction target future.The trend of next stage is predicted by attention rate and moos index, the transformation flex point of trend can be obtained at the first time, avoid delaying for information.
Description
Technical field
This application involves data processing field, especially, is related to the analysis side on upward price trend prediction in financial field
Method, equipment and storage medium.
Background technology
With the development of internet industry, information technology is occupied an leading position, market direction of the securities market towards modernization
Development.Present Shanghai and Shenzhen listed company alreadys exceed thousands of families, however the income of equity investment to risk be often it is directly proportional, i.e.,
Investment return is higher, may risk risk is bigger.Therefore, the research of Stock Market Forecasting method have extremely important application value and
Theory significance.Always there are many conventional analytical techniques, it should say these traditional technical Analysis methods on stock analysis also
To achieve larger achievement, however, it is not difficult to find that these existing theoretical and methods be also there is it is very big the defects of
, they be invariably using static method, qualitative description it is more, quantitative description lacks, and many factors of influence stock market are cut
Split and carry out single analysis.Therefore, these limitations cause these methods cannot be effective in Stock Price Fluctuation changeable
The change of stock price is held exactly.Therefore need to explore the complexity and regularity of stock market fluctuations, and according to it
Regularity design is a series of easy to operate, the enough high forecasting softwares of precision, and investors avoid risk to be vast.
Stock market is Multivariable Nonlinear Dynamic Systems, at present academicly also without preferable modeling method, meanwhile, stock
Ticket market has a degree of uncertainty, therefore, those is attempted to carry out share price with the method for establishing accurate model pre-
The method of survey, its prediction effect necessarily will not be preferable；Non-linear spy is presented between each variable of share price itself and influence share price
Property, therefore it is required that there is the ability of powerful processing nonlinear problem.
Correlation in stock market between various factors is intricate, and primary-slave relation comes and go, and quantitative relation is difficult to
Extraction, therefore it is also extremely difficult to apply conventional Forecasting Methodology to make quantitative analysis to stock market；
Enchancement factor is many in stock market, and the influence to stock index, price is notable, and price fluctuation is violent, dry sound pitch, performance
Go out very strong non-linear, uncertain.Many methods are proposed to Stock Price Forecasting both at home and abroad.These methods are in real work
In have an important guiding effect, but there are still some places not fully up to expectations, as regression model extrapolation is poor, analogy coefficient
Method accuracy is poor, and neural computing amount is big, is also easy to produce over-fitting.
The content of the invention
To overcome drawbacks described above, the application provides a kind of trend prediction analysis method, equipment and storage medium, to solve
The problem of trend prediction is inaccurate.
To solve the above problems, disclosure is sent out a kind of trend prediction analysis method, suitable for being performed in computing device,
Comprise the following steps：
(1) access window is established, the window includes the quotation information and information of multiple targets；
(2) access information of user is obtained, the user counted according to user access information in each target unit interval accesses
Amount, the access information include time point, the corresponding access times for accessing target and the corresponding user identifier of access behavior；
(3) attention rate and moos index of each target within the unit interval are calculated；
(4) each target attention rate curve and moos index curve are generated, it is bent according to the attention rate curve and moos index
Line predicts the tendency in target future.
Further, user's visit capacity statistical method includes：
The user identifier is identified and filtered to correct to the target access times, when obtaining each target unit
Interior user's visit capacity.
Further, the computational methods of the attention rate include：
User visit capacity divided by whole target of each target within the unit interval are calculated within the unit interval
User's visit capacity obtain as a result, and using attention rate of the result as each target within the per unit time, it is described
Attention rate is used to be characterized in the interim preference of user in the corresponding unit interval to accessing target.
Further, the computational methods of the moos index include：
Definition is positive, neutral, the corresponding score value of passiveness three classes mood keyword；
Count positive, the neutral and passive three classes mood in the relevent information information that each target occurs within the unit interval
The totalling frequency of keyword and the score value of each mood, and after the score value of each mood is added up divided by add up the frequency and draw institute
State the moos index of target.
Further, the step (4) includes：
By attention rate connection generation attention rate curve of the single target in preset time section；
Moos index of the single target in preset time section is connected and carries out smoothly generating moos index smoothed curve；
If the attention rate of current one time is more than predetermined threshold value, and moos index is more than predetermined threshold value, described in prediction
The target next stage is bull market；
If the attention rate of current one time is more than predetermined threshold value, and moos index is less than predetermined threshold value, described in prediction
The target next stage is bearish market；
In moos index and the section that attention rate is predetermined threshold value, then the next stage for predicting the target is to check and regulate market.
Further, the step (4) includes：
By attention rate connection generation attention rate curve of the single target in preset time section, and to the attention rate curve
Carry out smooth generation attention rate smoothed curve；
Moos index of the single target in preset time section is connected and carries out smoothly generating moos index smoothed curve；
Predict that the target is following according to attention rate curve, attention rate smoothed curve and moos index smoothed curve to walk
Gesture, if current emotional index is more than predetermined threshold value, when attention rate curve and attention rate smoothed curve persistently rise, or currently
Attention rate curve is upward through attention rate smoothed curve, and it is bull market to predict the target next stage；
If current emotional index is less than predetermined threshold value, when attention rate curve and attention rate smoothed curve persistently rise, or
The current attention rate curve of person is upward through attention rate smoothed curve, and it is bearish market to predict the target next stage；
If moos index and attention rate in the section of predetermined threshold value, predict that the next stage of the target is to check and regulate row
Feelings.
Further, trend prediction analysis method of the present invention further includes information push step：
By the attention rate curve of the target in preset time section and/or attention rate smoothed curve and/or moos index curve
And/or moos index smoothed curve is pushed to user terminal；And/or
The prediction result of target next stage is pushed to user terminal.
Further, described information push step specifically includes：
The frequency that the sole user within the current one time accesses single target is counted, is accessed according to the sole user single
The frequency of one target determines the preference target of the user, and the preference target accesses frequency for the user in the current one time
Rate exceedes the target of predetermined threshold value；
By the attention rate curve and/or attention rate smoothed curve and/or moos index of the target in preset time section
Curve and/or the push of moos index smoothed curve are requested to user terminal；And/or
The prediction result of the target next stage is pushed or is requested to user terminal.
Or
Server receives the request of user, by the attention rate curve and/or attention rate of the request target in preset time section
Smoothed curve and/or moos index curve and/or moos index smoothed curve are pushed to user terminal；And/or
The prediction result of the request target next stage is pushed to user terminal.
The application also provides a kind of computing device, including：
One or more processors；
Memory；And
One or more programs, wherein one or more of program storages are in the memory and are configured as by institute
One or more processors execution is stated, one or more of programs include being used to perform any in trend prediction analysis method
Method.
The application also provides a kind of computer-readable storage medium, and the storage medium is stored with one or more programs, described
One or more programs include instruction, and described instruction is when executed by a computing apparatus so that the computing device trend is pre-
Survey the either method in analysis method.
Compared with prior art, the application has the following advantages：
1. predicting the trend of next stage by attention rate and moos index, can obtain trend at the first time turns flex point,
Avoid delaying for information；
2. introducing smoothing algorithm, curve is carried out can smoothly to reduce burr so that steady and continuous is distributed, can be more stable
One section of future prospects of prediction, and provide accurate probability.
Brief description of the drawings
Fig. 1 is the flow chart of one embodiment of the invention trend prediction analysis method.
Fig. 2 is another embodiment of the present invention trend prediction analysis method and flow chart.
Embodiment
The main purpose of the application is to provide a kind of trend prediction analysis method, by the user for obtaining access window target
Visit capacity generates user attention rate of each target within the unit interval, and the user's attention rate is the single target by the unit interval
User's visit capacity divided by whole target the result that draws of user's visit capacity；In information by statistical window target
Mood keyword calculates the moos index of corresponding target；Target attention rate curve and the moos index for generating preset time section are bent
Line, passes through the attention rate curve and the trend of moos index curve prediction next stage.
Target described in trend prediction analysis method of the present invention includes but not limited to the finance such as stock, fund, bond, gold
Product and relevant financial spin-off, lower single order is predicted by analyzing current target and user's attention rate of history and moos index
The price ups and downs trend of section, introduces smooth, has stability, one section of future prospects of prediction that can be more stable.
For ease of understanding the present embodiment, first to the trend prediction analysis method disclosed in the embodiment of the present invention into
Row is discussed in detail, and will be described in detail in the following examples by taking stock target as an example.
Embodiment one
The present embodiment provides a kind of trend prediction analysis method, this method is suitable for performing in the computing device.With reference to
Fig. 1, shows the present embodiment trend prediction analysis method flow diagram, comprises the following steps：
Step 101, access window is established, the window includes the quotation information and information of multiple targets.
User's access window is established, which includes the quotation information of branched stock and the relevant information in stock market is believed
Breath, user can be accessed by APP, can also be accessed by website links.
Step 102, the access information of user is obtained, the user in each target unit interval is counted according to user access information
Visit capacity, the access information include time point, the corresponding access times for accessing target and the corresponding user mark of access behavior
Know.
In a particular embodiment, every access information sends the time point of access behavior, corresponding access target including user
Access times and corresponding user identifier.The access information can be obtained from the server of the access content.User can
, can also be by clicking on eventually to open the target of the browser input network address access window by PC or mobile terminal
Target described in links and accesses in the Application Program Interface at end.
The time point that the access behavior is obtained in this step be in order to subsequently with sometime granularity is counted when,
Which unit interval can be belonged to carry out judgement according to the time point of the behavior of access, obtain unit interval target access times and
It is in order to for statistic of user accessing amount to access user identifier.In the present embodiment, the unit interval by user preset, can using week as
Unit, day is unit, in units of hour, to be divided into unit, in seconds, does not make special limit herein.User exists
To a certain target it is possible that repeatedly access behavior in unit interval, therefore it need to be identified and filter to accessing user identifier
Correct to the target access times, obtain user's visit capacity in each target unit interval.Know to accessing user identifier
Not and filter to correct to target access times, it is therefore an objective to avoid tiring out multiple access of the same user within the unit interval
Meter, in order to avoid influence the follow-up result of calculation of the target.For example, using day as a unit interval, there are 100 different users to visit in one day
Ask target A-share ticket, including access the quotation information of A-share ticket, wherein there are 10 users repeatedly to have accessed the target in the day, should
User's visit capacity on the day of target is calculated as 100.
Step 103, attention rate and moos index of each target within the unit interval are calculated.
In a particular embodiment, the attention rate reflects within corresponding period user to accessing the interim preference of target
Degree, attention rate is bigger, represents the target and is more held positive expectations to the target by investor's preference, a kind of situation, investor, separately
A kind of situation, investor hold downbeat mood to the target.Specifically, can be by calculating use of the single target within the unit interval
The user's visit capacity of family visit capacity divided by whole target within the unit interval obtain as a result, and using the result as the mark
The attention rate within the unit interval, will own in the unit interval in user's visit capacity of a certain target and the unit interval
Attention rate of the ratio of user's visit capacity of target as the target in the unit interval.
In a particular embodiment, access window further includes the corresponding information of target.It may be selected to preset mood pass
The database of keyword, the database include positive, neutral, passive three classes mood keyword, and definition is positive, neutrality, passive three classes
Mood keyword corresponds to 1 point, 0 point and -1 point of score value respectively.Screening and the mood keyword in the relevant information of target,
Determine that the target counts the product in the relevent information information that each target occurs within the unit interval according to mood keywords database
The score value F of pole, the totalling frequency N of neutrality and passive three classes mood keyword and each moodi(i=1,2,3), and by each mood
Score value added up after divided by keyword add up the frequency and draw the moos index F of the target, formula is as follows：
Moos index wherein more than 0 is characterized as active mood, and the moos index equal to 0 is characterized as neutral mood, less than 0
Moos index be characterized as negative feeling.
Step 104, each target attention rate curve and moos index curve are generated, is referred to according to the attention rate curve and mood
Number curve predicts the tendency in target future.
The attention rate generated in above-mentioned unit time is attached generation by single target in preset time section
Attention rate curve, and accordingly the moos index generated in above-mentioned unit time is connected to single target in preset time section
Deliver a child into moos index curve.Become in the present embodiment with attention rate curve and moos index curve currency prediction target future
Gesture.
Preferably, to make moos index more stable continuously distributed, the present embodiment introduces smoothing algorithm, to the time interval
Interior target mood smoothed curve carries out smooth generation moos index smoothed curve, and to reduce burr, which adopts
With but be not limited to rolling average (Moving average), linear fit (linear fit), curve matching (quadratic
Fit), SG filters (Savitzky-Golay) smoothing algorithm and carries out smoothly so that curve steady and continuous is distributed.
In the present embodiment, the market trend of next stage is predicted with the attention rate in the current one time and moos index.
Attention rate threshold value and moos index threshold value are preset,
If the attention rate of current one time is more than predetermined threshold value, and moos index is more than predetermined threshold value, described in prediction
The target next stage is bull market.If the attention rate of current one time is more than predetermined threshold value, and moos index is less than default
Threshold value, it is bearish market to predict the target next stage.In moos index and/or the section that attention rate is predetermined threshold value, then in advance
The next stage for surveying the target is to check and regulate market.
Specifically, the unit interval calculates in minutes, and the current one time is previous minute, such as previous point of target of setting
Clock attention rate is more than threshold value 0.05, and previous minute moos index is more than threshold value 0.02, and it is upper to predict the target next stage
Rise market；The setting previous minute attention rate of target is more than threshold value 0.05, and previous minute moos index is less than threshold value -0.02,
It is bearish market to predict the target next stage；When attention rate is in predetermined threshold value [0,0.05] section, or moos index is situated between
In in predetermined threshold value [- 0.02,0.02] section, then predicting next stage of the target to check and regulate market.
The present embodiment by the way that the attention rate of current one time and moos index to be used as to the trend prediction of target next stage,
The flex point and trend that can be predicted in time, the time delays for avoiding the equal line trend of analysis of history price from bringing.
Embodiment two
The present embodiment is step 104 with one difference of embodiment, bent according to the attention rate curve and moos index
Line predicts the tendency in target future.
The attention rate generated in above-mentioned unit time is attached generation by single target in preset time section
Attention rate curve, and smooth generation attention rate smoothed curve is carried out to curve, by mood of the single target in preset time section
Index connects and carries out smoothly generating moos index smoothed curve.With attention rate curve, attention rate smoothed curve in the present embodiment
Target future trend is predicted with moos index smoothed curve.
Preferably, to make curve more stable continuously distributed, the present embodiment introduces smoothing algorithm, in the time interval
Target attention rate curve and moos index curve carry out smooth generation respective smoothed curve, and to reduce burr, which calculates
Method uses but is not limited to rolling average (Moving average), linear fit (linear fit), curve matching
(quadratic fit), SG filtering (Savitzky-Golay) smoothing algorithm carry out smooth so that curve steady and continuous is distributed.
In the present embodiment, predicted down with the attention rate smoothed curve in preset time section and moos index smoothed curve
The market trend in stage.Attention rate threshold value and moos index threshold value are preset,
Predict that the target is following according to attention rate curve, attention rate smoothed curve and moos index smoothed curve to walk
Gesture, if current emotional index is more than predetermined threshold value, when attention rate curve and attention rate smoothed curve persistently rise, or currently
Attention rate curve is upward through attention rate smoothed curve, and it is bull market to predict the target next stage；If current emotional index
During less than predetermined threshold value, when attention rate curve and attention rate smoothed curve persistently rise, or current attention rate curve is worn upwards
Attention rate smoothed curve is crossed, it is bearish market to predict the target next stage；If moos index or attention rate are between predetermined threshold value
Section in, then predict next stage of the target to check and regulate market.
Moos index is more than threshold value 0.02 in the current one time, works as pass
Note writes music line and attention rate smoothed curve persistently rises, or current attention rate curve is upward through attention rate smoothed curve, in advance
It is bull market to survey the target next stage.
Moos index is less than threshold value -0.02 in the current one time, when attention rate curve and attention rate smoothed curve continue
Rise, or current attention rate curve is upward through attention rate smoothed curve, it is bearish market to predict the target next stage.
If in time interval moos index between predetermined threshold value section [- 0.02,0.02] or attention rate between predetermined threshold value
In the section in section [0,0.05], then the next stage for predicting the target is to check and regulate market.
Specifically, can also be according to attention rate curve, attention rate smoothed curve, moos index smoothed curve, referring concurrently to mark
The historical quotes in preset time section predict the market trend of the target next stage jointly, details are not described herein.
Embodiment three
The present embodiment is considered as the improvement of above-described embodiment, with reference to figure 2, shows the present embodiment trend prediction analysis method
Flow chart.To 204 identical with 101 to 104 steps of embodiment one or embodiment two, difference exists the present embodiment step 201
Step 205 is pushed in further including information.
Specifically, the attention rate curve of the target in preset time section and/or attention rate smoothed curve and/or mood are referred to
Number curve and/or moos index smoothed curve are pushed to user terminal；Or the prediction result of target next stage is pushed to user
Terminal；Or by the attention rate curve of the target in preset time section and/or attention rate smoothed curve and/or moos index curve
And/or the prediction result of moos index smoothed curve and target next stage are pushed to user terminal at the same time.
Preferably, the number that the sole user within the current one time accesses single target is counted, according to the single use
The number of the single target of family access determines the preference target of the user, and the preference target is the user in the current one time
Access times exceed the target of predetermined threshold value.Specifically, party A-subscriber have accessed target M in the previous day 5 times, more than default threshold value
4 times, then M is determined as to the preference target of user A, automatically by the attention rate curve of the target M in preset time section and/or concern
Degree smoothed curve and/or moos index curve and/or moos index smoothed curve are pushed to user terminal；Or by rank under target M
The prediction result of section is pushed to user terminal；It is or the attention rate curve of the target M in preset time section and/or attention rate is smooth
Curve and/or the prediction result of moos index curve and/or moos index smoothed curve and target M next stages are pushed at the same time
User terminal.
Alternatively, server receives the request of user, by the attention rate curve of the request target in preset time section and/or
Attention rate smoothed curve and/or moos index curve and/or moos index smoothed curve are pushed to user terminal；Or described it will ask
The prediction result of target next stage is asked to be pushed to user terminal；Or the attention rate curve by the request target in preset time section
And/or attention rate smoothed curve and/or moos index curve and/or moos index smoothed curve and the target next stage is pre-
Survey result and be pushed to user terminal at the same time.
For the push of user volume body customized information, the reference of investment decision can be made to user in time.
Example IV
The present embodiment also provides a kind of trend prediction analysis device, which is suitable for residing in computing device, including with
Lower module：
Access window establishes module, is adapted to set up access window, and the window includes the quotation information and money of multiple targets
Interrogate information.
Access information acquisition module, suitable for obtaining the access information of user, each target list is counted according to user access information
User's visit capacity in the time of position, the access information include the time point of access behavior, the corresponding access times for accessing target
With corresponding user identifier.
Computing module, suitable for calculating attention rate and moos index of each target within the unit interval.
Trend prediction module, suitable for generating each target attention rate curve and moos index curve, writes music according to the concern
Line and the tendency in moos index curve prediction target future.
Embodiment five
The present embodiment also provides a kind of computing device, including：
One or more processors；
Memory；And
One or more programs, wherein one or more of program storages are in the memory and are configured as by institute
One or more processors execution is stated, one or more of programs include being used to perform the either method in following method：
(1) access window is established, the window includes the quotation information and information of multiple targets；
(2) access information of user is obtained, the user counted according to user access information in each target unit interval accesses
Amount, the access information include time point, the corresponding access times for accessing target and the corresponding user identifier of access behavior；
(3) attention rate and moos index of each target within the unit interval are calculated；
(4) each target attention rate curve and moos index curve are generated, it is bent according to the attention rate curve and moos index
Line predicts the tendency in target future.
Embodiment six
The present embodiment also provides a kind of computer-readable storage medium, and the storage medium is stored with one or more programs, institute
Stating one or more programs includes instruction, and described instruction is when executed by a computing apparatus so that the computing device is as follows
Either method in method：
(1) access window is established, the window includes the quotation information and information of multiple targets；
(2) access information of user is obtained, the user counted according to user access information in each target unit interval accesses
Amount, the access information include time point, the corresponding access times for accessing target and the corresponding user identifier of access behavior；
(3) attention rate and moos index of each target within the unit interval are calculated；
(4) each target attention rate curve and moos index curve are generated, it is bent according to the attention rate curve and moos index
Line predicts the tendency in target future.
Method and apparatus described in above-described embodiment, can specifically be realized by computer chip or entity, or by with certain
The product of function realizes, wherein, a kind of typical equipment is computer.Specifically, computer can be personal computer, clothes
Business device, laptop computer, cell phone, camera phone, smart phone, personal digital assistant, media player, navigation are set
The group of any equipment in standby, electronic mail equipment, game console platform, tablet PC, wearable device or these equipment
Close.
It will be understood by those skilled in the art that the embodiment of the present invention can provide method, system or computer program product.
Therefore, the present invention can use the embodiment in terms of complete hardware embodiment, complete software embodiment or combination software and hardware
Form.Deposited moreover, the present invention can use one or more computers for wherein including computer usable program code can use
The shape for the computer program product that storage media is implemented on (including but not limited to magnetic disk storage, CD, ROM, optical memory etc.)
Formula.
The foregoing is merely the embodiment of the present invention, is not intended to limit the invention.To those skilled in the art,
The invention may be variously modified and varied.All any modifications made within spirit and principles of the present invention, equivalent substitution,
Improve etc., it should be included within scope of the presently claimed invention.
Claims (10)
1. a kind of trend prediction analysis method, suitable for being performed in computing device, comprises the following steps：
(1) access window is established, the window includes the quotation information and information of multiple targets；
(2) access information of user is obtained, user's visit capacity in each target unit interval, institute are counted according to user access information
Stating access information includes time point, the corresponding access times for accessing target and the corresponding user identifier of access behavior；
(3) attention rate and moos index of each target within the unit interval are calculated；
(4) each target attention rate curve and moos index curve are generated, it is pre- according to the attention rate curve and moos index curve
Following tendency of mark.
2. trend prediction analysis method as claimed in claim 1, it is characterised in that user described in the step (2) accesses
Amount statistical method includes：
The user identifier is identified and filtered to correct to the target access times, obtained in each target unit interval
User's visit capacity.
3. trend prediction analysis method as claimed in claim 1, it is characterised in that the calculating of attention rate in the step (3)
Method includes：
Calculate user visit capacity divided by whole target use during unit interval of each target within the unit interval
It is that family visit capacity obtains as a result, and using attention rate of the result as each target within the per unit time, the concern
Degree is used to be characterized in the interim preference of user in the corresponding unit interval to accessing target.
4. trend prediction analysis method as claimed in claim 1, it is characterised in that the meter of moos index in the step (3)
Calculation method includes：
Define the score value that positive, neutral, passive three classes mood keyword corresponds to 1 point, 0 point and -1 point respectively；
Positive, the neutral and passive three classes mood counted in the relevent information information that each target occurs within the unit interval is crucial
The totalling frequency N of the word and score value F of each moodi(i=1,2,3), and it is after the score value of each mood is added up divided by described
The totalling frequency of keyword draws the moos index F of the target, and formula is as follows：
<mrow>
<mi>F</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mn>3</mn>
</munderover>
<msub>
<mi>F</mi>
<mi>i</mi>
</msub>
<mo>/</mo>
<mi>N</mi>
</mrow>
5. trend prediction analysis method as claimed in claim 1, it is characterised in that the step (4) includes：
By attention rate connection generation attention rate curve of the single target in preset time section；
Moos index of the single target in preset time section is connected and carries out smoothly generating moos index smoothed curve；
If the attention rate of current one time is more than predetermined threshold value, and moos index is more than predetermined threshold value, predicts the target
Next stage is bull market；
If the attention rate of current one time is more than predetermined threshold value, and moos index is less than predetermined threshold value, predicts the target
Next stage is bearish market；
In moos index or the section that attention rate is predetermined threshold value, then the next stage for predicting the target is to check and regulate market.
6. trend prediction analysis method as claimed in claim 1, it is characterised in that the step (4) includes：
By attention rate connection generation attention rate curve of the single target in preset time section, and the attention rate curve is carried out
Smooth generation attention rate smoothed curve；
Moos index of the single target in preset time section is connected and carries out smoothly generating moos index smoothed curve；
The tendency in the target future is predicted according to attention rate curve, attention rate smoothed curve and moos index smoothed curve, if
When current emotional index is more than predetermined threshold value, when attention rate curve and attention rate smoothed curve persistently rise, or current concern
Line of writing music is upward through attention rate smoothed curve, and it is bull market to predict the target next stage；
If current emotional index is less than predetermined threshold value, when attention rate curve and attention rate smoothed curve persistently rise, or work as
Preceding attention rate curve is upward through attention rate smoothed curve, and it is bearish market to predict the target next stage；
If moos index or attention rate in the section of predetermined threshold value, predict that the next stage of the target is to check and regulate market.
7. trend prediction analysis method as claimed in claim 1, it is characterised in that after the step (4), further include information
Push step：
By the attention rate curve of the target in preset time section and/or attention rate smoothed curve and/or moos index curve and/or
Moos index smoothed curve is pushed to user terminal；And/or
The prediction result of target next stage is pushed to user terminal.
8. trend prediction analysis method as claimed in claim 7, it is characterised in that described information push step specifically includes：
The number that the sole user within the current one time accesses single target is counted, single mark is accessed according to the sole user
Number determine the preference target of the user, the preference target surpasses for user's access times in the current one time
Cross the target of predetermined threshold value；
By the attention rate curve of the target in preset time section and/or attention rate smoothed curve and/or moos index curve
And/or moos index smoothed curve is pushed to user terminal；And/or
The prediction result of the target next stage is pushed to user terminal.
Or
Server receives the request of user, will ask the attention rate curve of target and/or attention rate smooth in preset time section
Curve and/or moos index curve and/or moos index smoothed curve are pushed to user terminal；And/or
The prediction result of the request target next stage is pushed to user terminal.
9. a kind of computing device, including：
One or more processors；
Memory；And
One or more programs, wherein one or more of program storages are in the memory and are configured as by described one
A or multiple processors perform, and one or more of programs include being used to perform stating appointing in method according to claim 1-8
One method.
10. a kind of computer-readable storage medium, the storage medium is stored with one or more programs, one or more of programs
Including instruction, described instruction is when executed by a computing apparatus so that the computing device states method according to claim 1-8
In either method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711337647.3A CN107992601A (en) | 2017-12-14 | 2017-12-14 | Trend prediction analysis method, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711337647.3A CN107992601A (en) | 2017-12-14 | 2017-12-14 | Trend prediction analysis method, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107992601A true CN107992601A (en) | 2018-05-04 |
Family
ID=62038389
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711337647.3A Pending CN107992601A (en) | 2017-12-14 | 2017-12-14 | Trend prediction analysis method, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107992601A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109166041A (en) * | 2018-08-29 | 2019-01-08 | 北京京东金融科技控股有限公司 | Stock market's forward prediction method and system, computer system and readable storage medium storing program for executing |
| CN110968745A (en) * | 2019-11-13 | 2020-04-07 | 泰康保险集团股份有限公司 | Data processing method and device, electronic equipment and computer readable medium |
| CN111210070A (en) * | 2020-01-03 | 2020-05-29 | 恩亿科(北京)数据科技有限公司 | Data analysis method and device, electronic equipment and readable storage medium |
| CN113590925A (en) * | 2020-04-30 | 2021-11-02 | 中国移动通信集团北京有限公司 | User determination method, device, equipment and computer storage medium |
-
2017
- 2017-12-14 CN CN201711337647.3A patent/CN107992601A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109166041A (en) * | 2018-08-29 | 2019-01-08 | 北京京东金融科技控股有限公司 | Stock market's forward prediction method and system, computer system and readable storage medium storing program for executing |
| CN110968745A (en) * | 2019-11-13 | 2020-04-07 | 泰康保险集团股份有限公司 | Data processing method and device, electronic equipment and computer readable medium |
| CN111210070A (en) * | 2020-01-03 | 2020-05-29 | 恩亿科(北京)数据科技有限公司 | Data analysis method and device, electronic equipment and readable storage medium |
| CN113590925A (en) * | 2020-04-30 | 2021-11-02 | 中国移动通信集团北京有限公司 | User determination method, device, equipment and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Vargas et al. | Deep leaming for stock market prediction using technical indicators and financial news articles | |
| Wang et al. | Continuous‐time mean–variance portfolio selection: A reinforcement learning framework | |
| WO2018214935A1 (en) | Method for determining risk preference of user, information recommendation method, and apparatus | |
| Sun et al. | Predicting stock price returns using microblog sentiment for chinese stock market | |
| WO2018214933A1 (en) | Method and apparatus for determining level of risk of user, and computer device | |
| CN107992601A (en) | Trend prediction analysis method, equipment and storage medium | |
| Li et al. | Stock price prediction incorporating market style clustering | |
| CN108734587A (en) | The recommendation method and terminal device of financial product | |
| WO2014152353A1 (en) | Goal-based portfolio management system | |
| Madani et al. | Is gold a hedge or safe haven against oil and currency market movements? A revisit using multifractal approach | |
| CN104598539A (en) | Internet event hot degree calculation method and terminal | |
| Uhr et al. | Sentiment analysis in financial markets A framework to utilize the human ability of word association for analyzing stock market news reports | |
| CN111612581A (en) | Method, device and equipment for recommending articles and storage medium | |
| CN115375177A (en) | User value evaluation method and device, electronic equipment and storage medium | |
| CN116362823A (en) | Recommendation model training method, recommendation method and recommendation device for behavior sparse scene | |
| Khalil et al. | Is deep-learning and natural language processing transcending the financial forecasting? Investigation through lens of news analytic process | |
| Alkhodhairi et al. | Bitcoin candlestick prediction with deep neural networks based on real time data | |
| CN109472586A (en) | Strategy determines method and device, storage medium, electronic device | |
| WO2024021354A1 (en) | Model training method, price prediction method, terminal device and storage medium | |
| CN107305662A (en) | Recognize the method and device of violation account | |
| CN115827994A (en) | Data processing method, device, equipment and storage medium | |
| Ormerod et al. | Measuring financial sentiment to predict financial instability: A new approach based on text analysis | |
| Kremser et al. | How do large stakes influence bitcoin performance? Evidence from the Mt. Gox liquidation case | |
| CN108197729A (en) | Value investment analysis method, equipment and storage medium based on machine learning | |
| CN115048487A (en) | Artificial intelligence-based public opinion analysis method, device, computer equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180504 |
|
| WD01 | Invention patent application deemed withdrawn after publication |