CN108876604A - Stock market's Risk Forecast Method, device, computer equipment and storage medium - Google Patents
Stock market's Risk Forecast Method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN108876604A CN108876604A CN201810514823.4A CN201810514823A CN108876604A CN 108876604 A CN108876604 A CN 108876604A CN 201810514823 A CN201810514823 A CN 201810514823A CN 108876604 A CN108876604 A CN 108876604A
- Authority
- CN
- China
- Prior art keywords
- word
- same day
- speech
- probability
- text data
- 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)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Machine Translation (AREA)
Abstract
The embodiment of the present application discloses a kind of stock market's Risk Forecast Method, device, computer equipment and storage medium.This method includes:Crawl the text data of same day designated time period;The text data of the same day designated time period is segmented, the word segmentation result of same day text is obtained;The word segmentation result of the same day text is input to the Recognition with Recurrent Neural Network for being used to predict stock market's risk that training obtains in advance, obtains same day stock advance-decline prediction result.This method realizes basic data of the massive information on making full use of internet as stock market's risk profile, carries out quick and intelligent Forecasting to stock market's risk according to Recognition with Recurrent Neural Network model.
Description
Technical field
This application involves risk profile technical fields more particularly to a kind of stock market's Risk Forecast Method, device, computer to set
Standby and storage medium.
Background technique
The amount of increase and amount of decrease of stock is predicted, can just be divided after typically carrying out specialty analysis based on the professional index such as K line chart
Analysis is as a result, this results in the acquisition of above-mentioned analysis result very high to stock invester's profession degree requirement.Moreover, with the hair of information technology
It opens up, includes the information of magnanimity in internet, wherein further including macroeconomy news, government not only comprising information such as stock market's transaction
The information such as economic policy, stock invester forum, stock invester's discussion bar.But these information are not utilized effectively to carry out stock market's risk
Prediction.
Summary of the invention
This application provides a kind of stock market's Risk Forecast Method, device, computer equipment and storage mediums, it is intended to solve existing
There is the amount of increase and amount of decrease of stock in technology to predict, being generally based on after the professional index such as K line chart carries out specialty analysis can just be analyzed
As a result, the problem of causing prediction process complicated and cannot achieve intelligent Forecasting.
In a first aspect, this application provides a kind of stock market's Risk Forecast Methods comprising:
Crawl the text data of same day designated time period;
The text data of the same day designated time period is segmented, the word segmentation result of same day text is obtained;
The word segmentation result of the same day text is input to the circulation mind for being used to predict stock market's risk that training in advance obtains
Through network, same day stock advance-decline prediction result is obtained.
Second aspect, this application provides a kind of stock market's risk profile devices comprising:
Same day text crawls unit, for crawling the text data of same day designated time period;
Same day text participle unit obtains the same day for segmenting the text data of the same day designated time period
The word segmentation result of text;
Same day Stock Market Forecasting unit trains what is obtained to be used in advance for the word segmentation result of the same day text to be input to
The Recognition with Recurrent Neural Network for predicting stock market's risk, obtains same day stock advance-decline prediction result.
The third aspect, the application provide a kind of computer equipment again, including memory, processor and are stored in described deposit
On reservoir and the computer program that can run on the processor, the processor realize this when executing the computer program
The described in any item stock market's Risk Forecast Methods provided are provided.
Fourth aspect, present invention also provides a kind of storage mediums, wherein the storage medium is stored with computer program,
The computer program includes program instruction, and described program instruction makes the processor execute the application when being executed by a processor
The described in any item stock market's Risk Forecast Methods provided.
The application provides a kind of stock market's Risk Forecast Method, device, computer equipment and storage medium.This method is by climbing
Take the text data of same day designated time period;The text data of the same day designated time period is segmented, obtains working as Japanese
This word segmentation result;The word segmentation result of the same day text is input to trains what is obtained to be used to predict following for stock market's risk in advance
Ring neural network obtains same day stock advance-decline prediction result.This method realizes the massive information conduct on making full use of internet
The basic data of stock market's risk profile carries out quick and intelligent Forecasting to stock market's risk according to Recognition with Recurrent Neural Network model.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in embodiment description
Attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is some embodiments of the present application, general for this field
For logical technical staff, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of schematic flow diagram of stock market's Risk Forecast Method provided by the embodiments of the present application;
Fig. 2 is a kind of sub-process schematic diagram of stock market's Risk Forecast Method provided by the embodiments of the present application;
Fig. 3 is a kind of another sub-process schematic diagram of stock market's Risk Forecast Method provided by the embodiments of the present application;
Fig. 4 is a kind of another schematic flow diagram of stock market's Risk Forecast Method provided by the embodiments of the present application;
Fig. 5 is a kind of schematic block diagram of stock market's risk profile device provided by the embodiments of the present application;
Fig. 6 is a kind of subelement schematic block diagram of stock market's risk profile device provided by the embodiments of the present application;
Fig. 7 is a kind of another subelement schematic block diagram of stock market's risk profile device provided by the embodiments of the present application;
Fig. 8 is a kind of another schematic block diagram of stock market's risk profile device provided by the embodiments of the present application;
Fig. 9 is a kind of schematic block diagram of computer equipment provided by the embodiments of the present application.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiment is some embodiments of the present application, instead of all the embodiments.Based on this Shen
Please in embodiment, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall in the protection scope of this application.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, the presence or addition of element, component and/or its set.
It is also understood that mesh of the term used in this present specification merely for the sake of description specific embodiment
And be not intended to limit the application.As present specification and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in present specification and the appended claims is
Refer to any combination and all possible combinations of one or more of associated item listed, and including these combinations.
Referring to Fig. 1, Fig. 1 is a kind of schematic flow diagram of stock market's Risk Forecast Method provided by the embodiments of the present application.It should
Method is applied in the terminals such as desktop computer, laptop computer, tablet computer.As shown in Figure 1, the method comprising the steps of S101~
S103。
S101, the text data for crawling same day designated time period.
In the present embodiment, if need to predict same day stock advance-decline prediction as a result, if can crawl text data in real time, based on
The text data crawled carries out preliminary advance versus decline width prediction, but crawls text data without departing from same day designated time period in real time
Interior text data (such as 9 are set by same day designated time period:00-15:00).Increasingly with the text data crawled
It is more, it can constantly adjust same day stock advance-decline prediction result.
S102, the text data of the same day designated time period is segmented, obtains the word segmentation result of same day text.
It in the present embodiment, is by based on probability when being segmented to the text data of the same day designated time period
The segmenting method of statistical model segments text data.For example, enabling C=C1C2...Cm, C is Chinese character string to be slit, is enabled
W=W1W2...Wn, W are cuttings as a result, Wa, Wb ... .Wk are all possible cutting schemes of C.So, it is united based on probability
The segmentation model of meter is to find purpose word string W, so that W meets:P (W | C)=MAX (P (Wa | C), P (Wb | C) ... P
(Wk | C)) participle model, the word string W i.e. estimated probability that above-mentioned participle model obtains is the word string of maximum.
By segmenting text data, accounting for for the word of various parts of speech in text data can be more accurately counted
Than more accurately to predict stock market's risk.
S103, the word segmentation result of the same day text is input to and trains what is obtained to be used to predict following for stock market's risk in advance
Ring neural network obtains same day stock advance-decline prediction result.
In the present embodiment, the word segmentation result of the same day text is input to trains what is obtained to be used to predict stock market in advance
The Recognition with Recurrent Neural Network of risk, can make full use of internet on basic data of the massive information as stock market's risk profile,
Quick and intelligent Forecasting is carried out to stock market's risk according to Recognition with Recurrent Neural Network model.For example, in the text data of same day acquisition
It is segmented including 100000, wherein positive participle has 40000, neutrality participle 10000, negative sense segments 50000, then exports
Same day stock advance-decline prediction result is -1% (indicating that same day share price drop range is 1%).
In one embodiment, as shown in figure 4, further including before step S101:
S1001, from multiple predesignated addresses URL text data, and stock corresponding with text data are crawled by date
Ups and downs data.
In the present embodiment, it can use reptile instrument and crawl related data by date, the data crawled mainly include using
The data such as the article of family comment, national policy, financial and economic news and economist, and obtain the corresponding stock of text data simultaneously and rise
Fall data.And due to setting the address URL, therefore there can be crawling from some representative websites with stock for directive property
The relevant national policy of ticket (expenses of taxation as reduced stock exchange), financial and economic news (such as company A and B company reach cooperation intention),
The article (such as AA expert is about the current analysis of financial conditions of company A or the prediction for situation of getting a profit) of economist, user comment
(comment of such as AAA online friend about company A stock), and crawl advance versus decline data (such as company A same day corresponding with text data
Amount of increase be 2.2%), by crawling above-mentioned text data and advance versus decline data corresponding with text data, can be used as instructing
Practice data to train neural network model.By reptile instrument, the targeted automatic acquisition of mass data is realized, is improved
The efficiency that basic data obtains.
It is crawled 2017 12 for example, can use reptile instrument from from financial web site, stock website and some investment websites
Text data of this month on -2017 years on the 1st December 31 of the moon, and advance versus decline data corresponding with text data, that is, distinguish
Crawl on December 1st, 2017 text data and same day advance versus decline data corresponding with text data, crawl December 2 in 2017
The text data of day and same day advance versus decline data ... ... corresponding with text data, crawl the text on December 31st, 2017
Data and same day advance versus decline data corresponding with text data.
If judging, the same day stops business, i.e., the corresponding same day advance versus decline data of text data that are crawled of the same day are sky, then will
The text data that the same day is crawled is added to the text data of next day, similarly by the textual data of upper two day if next day still stops business
According to the data basis being added in the text data of next day as model training.
S1002, text data progress word segmentation processing is obtained into word segmentation result, and each participle in word segmentation result is carried out
Part-of-speech tagging.
In the present embodiment, text data is segmented by the segmenting method based on probability statistics model, passes through base
In the word string that the word string that the segmenting method of probability statistics model obtains i.e. estimated probability is maximum.
In one embodiment, as shown in Fig. 2, the step S1002 includes:
S10021, candidate word is taken out from text data by sequence from left to right;
S10022, probability value corresponding with each candidate word is inquired in pre-stored dictionary, and record each candidate
The left adjacent word of word;
S10023, the cumulative probability for obtaining each candidate word is calculated, and obtains the corresponding multiple left adjacent words of each candidate word
Respective cumulative probability, if in the cumulative probability that there are cumulative probabilities in multiple left adjacent words of each candidate word for multiple left adjacent words
The left adjacent word of maximum value, using the left adjacent word of maximum value in cumulative probability as best left adjacent word corresponding with candidate word;
S10024, from the terminal word of text data be starting point, be sequentially output from right to left it is corresponding with each candidate word most
Good left adjacent word, obtains word segmentation result.
I.e. to a substring S to be segmented, according to sequence from left to right take out whole candidate word w1, w2 ..., wi ...,
wn;The probability value P (wi) of each candidate word is found in dictionary, and the left adjacent word of the whole for recording each candidate word;It calculates each
The cumulative probability of candidate word, while comparing the best left adjacent word for obtaining each candidate word;If current word wn is the tail of word string S
Word, and cumulative probability P (wn) is maximum, then wn is exactly the terminal word of S;It, successively will be each according to sequence from right to left since wn
The best left adjacent word output of word, the i.e. word segmentation result of S.By segmenting text data, text can be more accurately counted
The accounting of the word of various parts of speech in data, more accurately to predict stock market's risk.
In one embodiment, as shown in figure 3, the step S1002 further includes:
S10025, each participle in word segmentation result is inputted to the screening model for being used for part-of-speech tagging constructed in advance, obtained
To part of speech probability corresponding with each participle;
If S10026, segmenting corresponding part of speech probability and being greater than or equal to pre-set the in the probability of positive part of speech classification
Participle is labeled as positive part of speech by one probability;
If S10027, segmenting corresponding part of speech probability and being greater than or equal to pre-set the in the probability of neutral part of speech classification
Participle is labeled as neutral part of speech by two probability;
If S10028, segmenting corresponding part of speech probability and being greater than or equal to pre-set the in the probability of negative sense part of speech classification
Participle is labeled as negative sense part of speech by three probability.
In the present embodiment, screening mould constructed by model training is carried out to historical data according to NB Algorithm
Type;The screening model is for judging that input data is positive part of speech, neutral part of speech or negative sense part of speech.
In one embodiment, when the screening model for part-of-speech tagging of building, by multiple points included in training set
Input of the word as screening model, and using the corresponding part of speech of each participle as the output of screening model, it is trained and is sieved
Modeling type.
The model of the NB Algorithm of use is as follows:
Wherein,
NckIndicate c in training setkThe number of class document, N indicate total number of documents in training set;TjkIndicate lexical item tjIn classification
The number occurred in ck, V are the lexical item sets of all categories.Classifier by above-mentioned screening model as the part of speech of participle,
It can judge that input data is positive part of speech, neutral part of speech or negative sense part of speech.
For example, by the model for being input to NB Algorithm of each participle, when the data appear in positive part of speech
The probability of classification, which is greater than or equal to 50% (i.e. the first probability is set as 50%), can then be considered as the data positive part of speech;Work as participle
Corresponding part of speech probability is greater than or equal to 50% (i.e. the second probability is set as 50%) in the probability of neutral part of speech classification, will segment
It is labeled as neutral part of speech;It is greater than or equal to 50% (i.e. third when segmenting probability of the corresponding part of speech probability in negative sense part of speech classification
50%) probability is set as, participle is labeled as negative sense part of speech.
S1003, the word segmentation result after part-of-speech tagging will be carried out as the input of Recognition with Recurrent Neural Network, and by corresponding stock
Output of the ups and downs data as Recognition with Recurrent Neural Network is trained to obtain Recognition with Recurrent Neural Network model.
In the present embodiment, used Recognition with Recurrent Neural Network model is:
Ot=g (VSt)
St=f (UXt+St-1);
Wherein, XtIt is the value of Recognition with Recurrent Neural Network input layer, St, St-1It is the value of Recognition with Recurrent Neural Network hidden layer, OtIt is to follow
The value of ring neural network output layer, U are first weight matrix of the input layer to hidden layer, and V is hidden layer to the second of output layer
Weight matrix, g () are nonlinear activation primitive (such as tanh), and f () is softmax function.
By the way that input of the word segmentation result after part-of-speech tagging as Recognition with Recurrent Neural Network will be carried out, and corresponding stock is risen
Fall output of the data as Recognition with Recurrent Neural Network, can train obtain the first weight matrix, the second weight matrix and, circulation mind
Through network model, the model as subsequent advance versus decline width Trend judgement is obtained in this way.
Such as include 100000 participles in the text data of acquisition on January 1st, 2018, wherein positive participle has 50000
A, neutrality participle 10000, negative sense segments 40000, and the advance versus decline data on the same day (indicate that same day share price amount of increase is for 5%
5%) first training, can will be carried out at this time obtains first circulation neural network model;
The text data for obtaining on January 2nd, 2018 again later by above-mentioned identical analysis and part-of-speech tagging, and is brought into
First circulation neural network model is trained, the second circulation neural network model after obtaining correction weight matrix;By defeated
Enter a large amount of training data, constantly repeat the above process, until obtaining accurate Recognition with Recurrent Neural Network model, training process
The training data being based on is more, then Recognition with Recurrent Neural Network model is more accurate to the prediction of stock market's risk.
In one embodiment, as shown in figure 4, further including after the step S103:
S104, the text data for crawling another designated time period, and it is input to Recognition with Recurrent Neural Network model, obtain next day stock
Ticket advance-decline forecasting result.
In the present embodiment, if need to predict next day stock advance-decline prediction as a result, if can crawl text data in real time, based on
The text data crawled carries out preliminary advance versus decline width prediction, but crawls text data without departing from another designated time period in real time
Interior text data (such as the same day 15 is set by another designated time period:00- next day 9:00).With the textual data crawled
According to more and more, next day stock advance-decline prediction result can be constantly adjusted.Namely the period locating for the text data that ought be crawled is
The 9 of the same day:00-15:00, then what text data was input to that Recognition with Recurrent Neural Network model obtains is same day stock advance-decline prediction knot
Fruit;The period locating for the text data crawled is the 15 of the same day:00- next day 900, then text data is input to circulation nerve net
What network model obtained is next day stock advance-decline prediction result.
As it can be seen that this method realizes basic data of the massive information on making full use of internet as stock market's risk profile,
Quick and intelligent Forecasting is carried out to stock market's risk according to Recognition with Recurrent Neural Network model.
The embodiment of the present application also provides a kind of stock market's risk profile device, and stock market's risk profile device is aforementioned for executing
Any embodiment of stock market's Risk Forecast Method.Specifically, referring to Fig. 5, Fig. 5 is a kind of stock market provided by the embodiments of the present application
The schematic block diagram of risk profile device.Stock market's risk profile device 100 can be configured at desktop computer, tablet computer, portable
Computer, etc. in terminals.
As shown in figure 5, stock market's risk profile device 100 includes that same day text crawls unit 101, same day text participle unit
102 and same day Stock Market Forecasting unit 103.
Same day text crawls unit 101, for crawling the text data of same day designated time period.
In the present embodiment, if need to predict same day stock advance-decline prediction as a result, if can crawl text data in real time, based on
The text data crawled carries out preliminary advance versus decline width prediction, but crawls text data without departing from same day designated time period in real time
Interior text data (such as 9 are set by same day designated time period:00-15:00).Increasingly with the text data crawled
It is more, it can constantly adjust same day stock advance-decline prediction result.
Same day text participle unit 102 is worked as segmenting the text data of the same day designated time period
The word segmentation result of day text.
It in the present embodiment, is by based on probability when being segmented to the text data of the same day designated time period
The segmenting method of statistical model segments text data.For example, enabling C=C1C2...Cm, C is Chinese character string to be slit, is enabled
W=W1W2...Wn, W are cuttings as a result, Wa, Wb ... .Wk are all possible cutting schemes of C.So, it is united based on probability
The segmentation model of meter is to find purpose word string W, so that W meets:P (W | C)=MAX (P (Wa | C), P (Wb | C) ... P
(Wk | C)) participle model, the word string W i.e. estimated probability that above-mentioned participle model obtains is the word string of maximum.
By segmenting text data, accounting for for the word of various parts of speech in text data can be more accurately counted
Than more accurately to predict stock market's risk.
Same day Stock Market Forecasting unit 103, for the word segmentation result of the same day text to be input to what training in advance obtained
For predicting the Recognition with Recurrent Neural Network of stock market's risk, same day stock advance-decline prediction result is obtained.
In the present embodiment, the word segmentation result of the same day text is input to trains what is obtained to be used to predict stock market in advance
The Recognition with Recurrent Neural Network of risk, can make full use of internet on basic data of the massive information as stock market's risk profile,
Quick and intelligent Forecasting is carried out to stock market's risk according to Recognition with Recurrent Neural Network model.For example, in the text data of same day acquisition
It is segmented including 100000, wherein positive participle has 40000, neutrality participle 10000, negative sense segments 50000, then exports
Same day stock advance-decline prediction result is -1% (indicating that same day share price drop range is 1%).
In one embodiment, as shown in figure 8, stock market's risk profile device 100 further includes:
Training data crawls unit 1001, for crawling text data from multiple predesignated addresses URL by date, and with
The corresponding advance versus decline data of text data.
In the present embodiment, it can use reptile instrument and crawl related data by date, the data crawled mainly include using
The data such as the article of family comment, national policy, financial and economic news and economist, and obtain the corresponding stock of text data simultaneously and rise
Fall data.And due to setting the address URL, therefore there can be crawling from some representative websites with stock for directive property
The relevant national policy of ticket (expenses of taxation as reduced stock exchange), financial and economic news (such as company A and B company reach cooperation intention),
The article (such as AA expert is about the current analysis of financial conditions of company A or the prediction for situation of getting a profit) of economist, user comment
(comment of such as AAA online friend about company A stock), and crawl advance versus decline data (such as company A same day corresponding with text data
Amount of increase be 2.2%), by crawling above-mentioned text data and advance versus decline data corresponding with text data, can be used as instructing
Practice data to train neural network model.By reptile instrument, the targeted automatic acquisition of mass data is realized, is improved
The efficiency that basic data obtains.
It is crawled 2017 12 for example, can use reptile instrument from from financial web site, stock website and some investment websites
Text data of this month on -2017 years on the 1st December 31 of the moon, and advance versus decline data corresponding with text data, that is, distinguish
Crawl on December 1st, 2017 text data and same day advance versus decline data corresponding with text data, crawl December 2 in 2017
The text data of day and same day advance versus decline data ... ... corresponding with text data, crawl the text on December 31st, 2017
Data and same day advance versus decline data corresponding with text data.
If judging, the same day stops business, i.e., the corresponding same day advance versus decline data of text data that are crawled of the same day are sky, then will
The text data that the same day is crawled is added to the text data of next day, similarly by the textual data of upper two day if next day still stops business
According to the data basis being added in the text data of next day as model training.
Participle mark unit 1002, for text data progress word segmentation processing to be obtained word segmentation result, and to word segmentation result
In each participle carry out part-of-speech tagging.
In the present embodiment, text data is segmented by the segmenting method based on probability statistics model, passes through base
In the word string that the word string that the segmenting method of probability statistics model obtains i.e. estimated probability is maximum.
In one embodiment, as shown in fig. 6, participle mark unit 102 includes:
Candidate word selection unit 10021, for taking out candidate word from text data by sequence from left to right;
Initial left adjacent word acquiring unit 10022, it is corresponding with each candidate word for being inquired in pre-stored dictionary
Probability value, and record the left adjacent word of each candidate word;
Best left adjacent word selection unit 10023, for calculating the cumulative probability for obtaining each candidate word, and obtains each time
The corresponding multiple left adjacent respective cumulative probabilities of word of word are selected, if there are cumulative probability being more in multiple left adjacent words of each candidate word
The left adjacent word of maximum value in the cumulative probability of a left adjacent word, using the left adjacent word of maximum value in cumulative probability as corresponding with candidate word
Best left adjacent word;
Word segmentation result output unit 10024, for from the terminal word of text data be starting point, be sequentially output from right to left with
The corresponding best left adjacent word of each candidate word, obtains word segmentation result.
I.e. to a substring S to be segmented, according to sequence from left to right take out whole candidate word w1, w2 ..., wi ...,
wn;The probability value P (wi) of each candidate word is found in dictionary, and the left adjacent word of the whole for recording each candidate word;It calculates each
The cumulative probability of candidate word, while comparing the best left adjacent word for obtaining each candidate word;If current word wn is the tail of word string S
Word, and cumulative probability P (wn) is maximum, then wn is exactly the terminal word of S;It, successively will be each according to sequence from right to left since wn
The best left adjacent word output of word, the i.e. word segmentation result of S.By segmenting text data, text can be more accurately counted
The accounting of the word of various parts of speech in data, more accurately to predict stock market's risk.
In one embodiment, as shown in fig. 7, participle mark unit 102 further includes:
Part of speech probability acquiring unit 10025 is used for word for construct each participle input in word segmentation result in advance
Property mark screening model, obtain part of speech probability corresponding with each participle;
Positive part-of-speech tagging unit 10026, if big in the probability of positive part of speech classification for segmenting corresponding part of speech probability
In or equal to pre-set first probability, participle is labeled as positive part of speech;
Neutral part-of-speech tagging unit 10027, if big in the probability of neutral part of speech classification for segmenting corresponding part of speech probability
In or equal to pre-set second probability, participle is labeled as neutral part of speech;
Negative sense part-of-speech tagging unit 10028, if big in the probability of negative sense part of speech classification for segmenting corresponding part of speech probability
In or equal to pre-set third probability, participle is labeled as negative sense part of speech.
In the present embodiment, screening mould constructed by model training is carried out to historical data according to NB Algorithm
Type;The screening model is for judging that input data is positive part of speech, neutral part of speech or negative sense part of speech.
In one embodiment, when the screening model for part-of-speech tagging constructed in part of speech probability acquiring unit 10025,
Using multiple participles included in training set as the input of screening model, and using the corresponding part of speech of each participle as screening mould
The output of type is trained to obtain screening model.
The model of the NB Algorithm of use is as follows:
Wherein,
NckIndicate c in training setkThe number of class document, N indicate total number of documents in training set;TjkIndicate lexical item tjIn classification
The number occurred in ck, V are the lexical item sets of all categories.Classifier by above-mentioned screening model as the part of speech of participle,
It can judge that input data is positive part of speech, neutral part of speech or negative sense part of speech.
For example, by the model for being input to NB Algorithm of each participle, when the data appear in positive part of speech
The probability of classification, which is greater than or equal to 50% (i.e. the first probability is set as 50%), can then be considered as the data positive part of speech;Work as participle
Corresponding part of speech probability is greater than or equal to 50% (i.e. the second probability is set as 50%) in the probability of neutral part of speech classification, will segment
It is labeled as neutral part of speech;It is greater than or equal to 50% (i.e. third when segmenting probability of the corresponding part of speech probability in negative sense part of speech classification
50%) probability is set as, participle is labeled as negative sense part of speech.
Model training unit 1003, for the word segmentation result after part-of-speech tagging will to be carried out as the defeated of Recognition with Recurrent Neural Network
Enter, and using corresponding advance versus decline data as the output of Recognition with Recurrent Neural Network, is trained to obtain Recognition with Recurrent Neural Network model.
In the present embodiment, used Recognition with Recurrent Neural Network model is:
Ot=g (VSt)
St=f (UXt+St-1);
Wherein, XtIt is the value of Recognition with Recurrent Neural Network input layer, St, St-1It is the value of Recognition with Recurrent Neural Network hidden layer, OtIt is to follow
The value of ring neural network output layer, U are first weight matrix of the input layer to hidden layer, and V is hidden layer to the second of output layer
Weight matrix, g () are nonlinear activation primitive (such as tanh), and f () is softmax function.
By the way that input of the word segmentation result after part-of-speech tagging as Recognition with Recurrent Neural Network will be carried out, and corresponding stock is risen
Fall output of the data as Recognition with Recurrent Neural Network, can train obtain the first weight matrix, the second weight matrix and, circulation mind
Through network model, the model as subsequent advance versus decline width Trend judgement is obtained in this way.
Such as include 100000 participles in the text data of acquisition on January 1st, 2018, wherein positive participle has 50000
A, neutrality participle 10000, negative sense segments 40000, and the advance versus decline data on the same day (indicate that same day share price amount of increase is for 5%
5%) first training, can will be carried out at this time obtains first circulation neural network model;
The text data for obtaining on January 2nd, 2018 again later by above-mentioned identical analysis and part-of-speech tagging, and is brought into
First circulation neural network model is trained, the second circulation neural network model after obtaining correction weight matrix;By defeated
Enter a large amount of training data, constantly repeat the above process, until obtaining accurate Recognition with Recurrent Neural Network model, training process
The training data being based on is more, then Recognition with Recurrent Neural Network model is more accurate to the prediction of stock market's risk.
In one embodiment, as shown in figure 8, stock market's risk profile device 100 further includes:
Next day Stock Market Forecasting unit 104 for crawling the text data of another designated time period, and is input to circulation nerve
Network model obtains next day stock advance-decline prediction result.
In the present embodiment, if need to predict next day stock advance-decline prediction as a result, if can crawl text data in real time, based on
The text data crawled carries out preliminary advance versus decline width prediction, but crawls text data without departing from another designated time period in real time
Interior text data (such as the same day 15 is set by another designated time period:00- next day 9:00).With the textual data crawled
According to more and more, next day stock advance-decline prediction result can be constantly adjusted.Namely the period locating for the text data that ought be crawled is
The 9 of the same day:00-15:00, then what text data was input to that Recognition with Recurrent Neural Network model obtains is same day stock advance-decline prediction knot
Fruit;The period locating for the text data crawled is the 15 of the same day:00- next day 900, then text data is input to circulation nerve net
What network model obtained is next day stock advance-decline prediction result.
As it can be seen that the device realizes basic data of the massive information on making full use of internet as stock market's risk profile,
Quick and intelligent Forecasting is carried out to stock market's risk according to Recognition with Recurrent Neural Network model.
Above-mentioned stock market's risk profile device can be implemented as a kind of form of computer program, which can be
It is run in computer equipment as shown in Figure 9.
Referring to Fig. 9, Fig. 9 is a kind of schematic block diagram of computer equipment provided by the embodiments of the present application.The computer
500 equipment of equipment can be terminal.The terminal can be tablet computer, laptop, desktop computer, personal digital assistant etc.
Electronic equipment.
Refering to Fig. 9, which includes processor 502, memory and the net connected by system bus 501
Network interface 505, wherein memory may include non-volatile memory medium 503 and built-in storage 504.
The non-volatile memory medium 503 can storage program area 5031 and computer program 5032.The computer program
5032 include program instruction, which is performed, and processor 502 may make to execute a kind of stock market's Risk Forecast Method.
The processor 502 supports the operation of entire computer equipment 500 for providing calculating and control ability.
The built-in storage 504 provides environment for the operation of the computer program 5032 in non-volatile memory medium 503, should
When computer program 5032 is executed by processor 502, processor 502 may make to execute a kind of stock market's Risk Forecast Method.
The network interface 505 such as sends the task dispatching of distribution for carrying out network communication.Those skilled in the art can manage
It solves, structure shown in Fig. 9, only the block diagram of part-structure relevant to application scheme, is not constituted to the application side
The restriction for the computer equipment 500 that case is applied thereon, specific computer equipment 500 may include more than as shown in the figure
Or less component, perhaps combine certain components or with different component layouts.
Wherein, the processor 502 is for running computer program 5032 stored in memory, to realize following function
Energy:Crawl the text data of same day designated time period;The text data of the same day designated time period is segmented, is worked as
The word segmentation result of day text;The word segmentation result of the same day text is input to trains what is obtained to be used to predict stock market's risk in advance
Recognition with Recurrent Neural Network, obtain same day stock advance-decline prediction result.
In one embodiment, processor 502 also performs the following operations:It is crawled by date from multiple predesignated addresses URL
Text data, and advance versus decline data corresponding with text data;Text data progress word segmentation processing is obtained into word segmentation result, and
Part-of-speech tagging is carried out to each participle in word segmentation result;Using the word segmentation result after progress part-of-speech tagging as Recognition with Recurrent Neural Network
Input be trained to obtain Recognition with Recurrent Neural Network and using corresponding advance versus decline data as the output of Recognition with Recurrent Neural Network
Model.
In one embodiment, processor 502 also performs the following operations:It is taken out from text data by sequence from left to right
Candidate word;Probability value corresponding with each candidate word is inquired in pre-stored dictionary, and records the left neighbour of each candidate word
Word;The cumulative probability for obtaining each candidate word is calculated, and it is general to obtain the corresponding multiple left adjacent respective accumulations of word of each candidate word
Rate, if there are the left neighbours of maximum value in the cumulative probability that cumulative probability is multiple left adjacent words in multiple left adjacent words of each candidate word
Word, using the left adjacent word of maximum value in cumulative probability as best left adjacent word corresponding with candidate word;From the terminal word of text data
For starting point, it is sequentially output best left adjacent word corresponding with each candidate word from right to left, obtains word segmentation result.
In one embodiment, processor 502 also performs the following operations:Each participle in word segmentation result is inputted into preparatory structure
The screening model for part-of-speech tagging built obtains part of speech probability corresponding with each participle;If segmenting corresponding part of speech probability
It is greater than or equal to pre-set first probability in the probability of positive part of speech classification, participle is labeled as positive part of speech;If participle
Corresponding part of speech probability is greater than or equal to pre-set second probability in the probability of neutral part of speech classification, and participle is labeled as
Property part of speech;If the probability for segmenting corresponding part of speech probability in negative sense part of speech classification is greater than or equal to pre-set third probability,
Participle is labeled as negative sense part of speech.
In one embodiment, processor 502 also performs the following operations:Using multiple participles included in training set as sieve
The input of modeling type, and using the corresponding part of speech of each participle as the output of screening model, it is trained to obtain screening model.
In one embodiment, processor 502 also performs the following operations:The Recognition with Recurrent Neural Network model is
Ot=g (VSt)
St=f (UXt+St-1);
Wherein, XtIt is the value of Recognition with Recurrent Neural Network input layer, St, St-1It is the value of Recognition with Recurrent Neural Network hidden layer, OtIt is to follow
The value of ring neural network output layer, U are first weight matrix of the input layer to hidden layer, and V is hidden layer to the second of output layer
Weight matrix, g () are nonlinear activation primitive, and f () is softmax function.
In one embodiment, processor 502 also performs the following operations:The text data of another designated time period is crawled, and
It is input to Recognition with Recurrent Neural Network model, obtains next day stock advance-decline prediction result.
It will be understood by those skilled in the art that the embodiment of computer equipment shown in Fig. 9 is not constituted to computer
The restriction of equipment specific composition, in other embodiments, computer equipment may include components more more or fewer than diagram, or
Person combines certain components or different component layouts.For example, in some embodiments, computer equipment can only include depositing
Reservoir and processor, in such embodiments, the structure and function of memory and processor are consistent with embodiment illustrated in fig. 9,
Details are not described herein.
It should be appreciated that in the embodiment of the present application, processor 502 can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic
Device, discrete gate or transistor logic, discrete hardware components etc..Wherein, general processor can be microprocessor or
Person's processor is also possible to any conventional processor etc..
A kind of storage medium is provided in another embodiment of the application.The storage medium can be computer-readable storage
Medium.The storage medium is stored with computer program, and wherein computer program includes program instruction.The program instruction is by processor
It is realized when execution:Crawl the text data of same day designated time period;The text data of the same day designated time period is divided
Word obtains the word segmentation result of same day text;By the word segmentation result of the same day text be input in advance training obtain be used for it is pre-
The Recognition with Recurrent Neural Network for surveying stock market's risk, obtains same day stock advance-decline prediction result.
In one embodiment, realization when which is executed by processor:By date from multiple predesignated addresses URL
Crawl text data, and advance versus decline data corresponding with text data;Text data progress word segmentation processing is obtained into participle knot
Fruit, and part-of-speech tagging is carried out to each participle in word segmentation result;Using the word segmentation result after progress part-of-speech tagging as circulation mind
Input through network, and using corresponding advance versus decline data as the output of Recognition with Recurrent Neural Network, it is trained to obtain circulation mind
Through network model.
In one embodiment, realization when which is executed by processor:By sequence from left to right from text data
Middle taking-up candidate word;Probability value corresponding with each candidate word is inquired in pre-stored dictionary, and records each candidate word
Left adjacent word;The cumulative probability for obtaining each candidate word is calculated, and it is respective to obtain the corresponding multiple left adjacent words of each candidate word
Cumulative probability, if there are maximum values in the cumulative probability that cumulative probability is multiple left adjacent words in multiple left adjacent words of each candidate word
Left adjacent word, using the left adjacent word of maximum value in cumulative probability as best left adjacent word corresponding with candidate word;From text data
Terminal word is starting point, is sequentially output best left adjacent word corresponding with each candidate word from right to left, obtains word segmentation result.
In one embodiment, realization when which is executed by processor:By each participle input in word segmentation result
The screening model for part-of-speech tagging constructed in advance obtains part of speech probability corresponding with each participle;If segmenting corresponding word
Property probability positive part of speech classification probability be greater than or equal to pre-set first probability, participle is labeled as positive part of speech;
If the probability for segmenting corresponding part of speech probability in neutral part of speech classification is greater than or equal to pre-set second probability, participle is marked
Note is neutral part of speech;If the probability for segmenting corresponding part of speech probability in negative sense part of speech classification is greater than or equal to pre-set third
Participle is labeled as negative sense part of speech by probability.
In one embodiment, realization when which is executed by processor:By multiple participles included in training set
As the input of screening model, and using the corresponding part of speech of each participle as the output of screening model, it is trained and is screened
Model.
In one embodiment, realization when which is executed by processor:The Recognition with Recurrent Neural Network model is
Ot=g (VSt)
St=f (UXt+St-1);
Wherein, XtIt is the value of Recognition with Recurrent Neural Network input layer, St, St-1It is the value of Recognition with Recurrent Neural Network hidden layer, OtIt is to follow
The value of ring neural network output layer, U are first weight matrix of the input layer to hidden layer, and V is hidden layer to the second of output layer
Weight matrix, g () are nonlinear activation primitive, and f () is softmax function.
In one embodiment, realization when which is executed by processor:Crawl the textual data of another designated time period
According to, and it is input to Recognition with Recurrent Neural Network model, obtain next day stock advance-decline prediction result.
The storage medium can be the internal storage unit of aforementioned device, such as the hard disk or memory of equipment.It is described to deposit
Storage media is also possible to the plug-in type hard disk being equipped on the External memory equipment of the equipment, such as the equipment, intelligent storage
Block (Smart Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card) etc..
Further, the storage medium can also both including the equipment internal storage unit and also including External memory equipment.
It is apparent to those skilled in the art that for convenience of description and succinctly, foregoing description is set
The specific work process of standby, device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
Those of ordinary skill in the art may be aware that unit described in conjunction with the examples disclosed in the embodiments of the present disclosure and algorithm
Step can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and software
Interchangeability generally describes each exemplary composition and step according to function in the above description.These functions are studied carefully
Unexpectedly the specific application and design constraint depending on technical solution are implemented in hardware or software.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In several embodiments provided herein, it should be understood that disclosed unit and method, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, can also will have identical function
The unit set of energy can be combined or can be integrated into another system at a unit, such as multiple units or components, or
Some features can be ignored or not executed.In addition, shown or discussed mutual coupling or direct-coupling or communication link
Connect can be through some interfaces, the indirect coupling or communication connection of device or unit, be also possible to electricity, it is mechanical or other
Form connection.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.Some or all of unit therein can be selected to realize the embodiment of the present invention according to the actual needs
Purpose.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, is also possible to two or more units and is integrated in one unit.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in one storage medium.Based on this understanding, technical solution of the present invention is substantially in other words to existing
The all or part of part or the technical solution that technology contributes can be embodied in the form of software products, should
Computer software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be
Personal computer, server or network equipment etc.) execute all or part of step of each embodiment the method for the present invention
Suddenly.And storage medium above-mentioned includes:USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), magnetic disk or
The various media that can store program code such as person's CD.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (10)
1. a kind of stock market's Risk Forecast Method, which is characterized in that including:
Crawl the text data of same day designated time period;
The text data of the same day designated time period is segmented, the word segmentation result of same day text is obtained;
The word segmentation result of the same day text is input to the circulation nerve net for being used to predict stock market's risk that training obtains in advance
Network obtains same day stock advance-decline prediction result.
2. stock market's Risk Forecast Method according to claim 1, which is characterized in that the same day designated time period that crawls
Before text data, further include:
Text data, and advance versus decline data corresponding with text data are crawled from multiple predesignated addresses URL by date;
Text data progress word segmentation processing is obtained into word segmentation result, and part-of-speech tagging is carried out to each participle in word segmentation result;
Using the word segmentation result after progress part-of-speech tagging as the input of Recognition with Recurrent Neural Network, and corresponding advance versus decline data are made
For the output of Recognition with Recurrent Neural Network, it is trained to obtain Recognition with Recurrent Neural Network model.
3. stock market's Risk Forecast Method according to claim 2, which is characterized in that described to carry out text data at participle
Reason obtains word segmentation result, including:
Candidate word is taken out from text data by sequence from left to right;
Probability value corresponding with each candidate word is inquired in pre-stored dictionary, and records the left adjacent word of each candidate word;
The cumulative probability for obtaining each candidate word is calculated, and it is general to obtain the corresponding multiple left adjacent respective accumulations of word of each candidate word
Rate, if there are the left neighbours of maximum value in the cumulative probability that cumulative probability is multiple left adjacent words in multiple left adjacent words of each candidate word
Word, using the left adjacent word of maximum value in cumulative probability as best left adjacent word corresponding with candidate word;
It is starting point from the terminal word of text data, is sequentially output best left adjacent word corresponding with each candidate word from right to left, obtains
To word segmentation result.
4. stock market's Risk Forecast Method according to claim 3, which is characterized in that each point in word segmentation result
Word carries out part-of-speech tagging, including:
Each participle in word segmentation result is inputted to the screening model for being used for part-of-speech tagging constructed in advance, is obtained and each participle
Corresponding part of speech probability;
If the probability for segmenting corresponding part of speech probability in positive part of speech classification is greater than or equal to pre-set first probability, will divide
Word is labeled as positive part of speech;
If the probability for segmenting corresponding part of speech probability in neutral part of speech classification is greater than or equal to pre-set second probability, will divide
Word is labeled as neutral part of speech;
If the probability for segmenting corresponding part of speech probability in negative sense part of speech classification is greater than or equal to pre-set third probability, will divide
Word is labeled as negative sense part of speech.
5. stock market's Risk Forecast Method according to claim 4, which is characterized in that each point by word segmentation result
The screening model for part-of-speech tagging that word input constructs in advance also wraps before obtaining part of speech probability corresponding with each participle
It includes:
Using multiple participles included in training set as the input of screening model, and using the corresponding part of speech of each participle as sieve
The output of modeling type is trained to obtain screening model.
6. stock market's Risk Forecast Method according to claim 1, which is characterized in that the Recognition with Recurrent Neural Network model is:
Ot=g (VSt)
St=f (UXt+St-1);
Wherein, XtIt is the value of Recognition with Recurrent Neural Network input layer, St, St-1It is the value of Recognition with Recurrent Neural Network hidden layer, OtIt is circulation mind
Value through network output layer, U are first weight matrix of the input layer to hidden layer, and V is second weight of the hidden layer to output layer
Matrix, g () are nonlinear activation primitive, and f () is softmax function.
7. stock market's Risk Forecast Method according to claim 1, which is characterized in that the participle by the same day text
As a result it is input to the Recognition with Recurrent Neural Network for being used to predict stock market's risk that training in advance obtains, obtains same day stock advance-decline prediction knot
After fruit, further include:
The text data of another designated time period is crawled, and is input to Recognition with Recurrent Neural Network model, it is pre- to obtain next day advance versus decline
Survey result.
8. a kind of stock market's risk profile device, which is characterized in that including:
Same day text crawls unit, for crawling the text data of same day designated time period;
Same day text participle unit obtains same day text for segmenting the text data of the same day designated time period
Word segmentation result;
Same day Stock Market Forecasting unit trains what is obtained to be used to predict in advance for the word segmentation result of the same day text to be input to
The Recognition with Recurrent Neural Network of stock market's risk obtains same day stock advance-decline prediction result.
9. a kind of computer equipment, including memory, processor and it is stored on the memory and can be on the processor
The computer program of operation, which is characterized in that the processor is realized when executing the computer program as in claim 1-7
Described in any item stock market's Risk Forecast Methods.
10. a kind of storage medium, which is characterized in that the storage medium is stored with computer program, the computer program packet
Program instruction is included, described program instruction executes the processor such as any one of claim 1-7 institute
The stock market's Risk Forecast Method stated.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810514823.4A CN108876604A (en) | 2018-05-25 | 2018-05-25 | Stock market's Risk Forecast Method, device, computer equipment and storage medium |
PCT/CN2018/101036 WO2019223133A1 (en) | 2018-05-25 | 2018-08-17 | Method for forecasting stock market risk, device, computer apparatus, and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810514823.4A CN108876604A (en) | 2018-05-25 | 2018-05-25 | Stock market's Risk Forecast Method, device, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108876604A true CN108876604A (en) | 2018-11-23 |
Family
ID=64334180
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810514823.4A Pending CN108876604A (en) | 2018-05-25 | 2018-05-25 | Stock market's Risk Forecast Method, device, computer equipment and storage medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN108876604A (en) |
WO (1) | WO2019223133A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110009128A (en) * | 2019-01-28 | 2019-07-12 | 平安科技(深圳)有限公司 | Industry public opinion index prediction technique, device, computer equipment and storage medium |
CN110059851A (en) * | 2019-03-08 | 2019-07-26 | 平安科技(深圳)有限公司 | The method, apparatus and computer equipment of prediction data variation based on deep learning |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105956770A (en) * | 2016-05-03 | 2016-09-21 | 中国科学院大学 | Stock market risk prediction platform and text excavation method thereof |
CN106384166A (en) * | 2016-09-12 | 2017-02-08 | 中山大学 | Deep learning stock market prediction method combined with financial news |
CN107679680A (en) * | 2017-11-21 | 2018-02-09 | 北京锐安科技有限公司 | A kind of financial forward prediction method, apparatus, equipment and storage medium |
CN108022016A (en) * | 2017-12-08 | 2018-05-11 | 宏谷信息科技(珠海)有限公司 | A kind of Prediction of Stock Price method and system based on artificial intelligence |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103985055A (en) * | 2014-05-30 | 2014-08-13 | 西安交通大学 | Stock market investment decision-making method based on network analysis and multi-model fusion |
CN106952161A (en) * | 2017-03-31 | 2017-07-14 | 洪志令 | A kind of recent forward prediction method of stock based on shot and long term memory depth learning network |
CN107832897A (en) * | 2017-11-30 | 2018-03-23 | 浙江工业大学 | A kind of Stock Price Forecasting method based on deep learning |
-
2018
- 2018-05-25 CN CN201810514823.4A patent/CN108876604A/en active Pending
- 2018-08-17 WO PCT/CN2018/101036 patent/WO2019223133A1/en active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105956770A (en) * | 2016-05-03 | 2016-09-21 | 中国科学院大学 | Stock market risk prediction platform and text excavation method thereof |
CN106384166A (en) * | 2016-09-12 | 2017-02-08 | 中山大学 | Deep learning stock market prediction method combined with financial news |
CN107679680A (en) * | 2017-11-21 | 2018-02-09 | 北京锐安科技有限公司 | A kind of financial forward prediction method, apparatus, equipment and storage medium |
CN108022016A (en) * | 2017-12-08 | 2018-05-11 | 宏谷信息科技(珠海)有限公司 | A kind of Prediction of Stock Price method and system based on artificial intelligence |
Non-Patent Citations (4)
Title |
---|
@HANBINGTAO: "零基础入门深度学习(5)-循环神经网络", 《HTTPS://ZYBULUO.COM/HANBINGTAO/NOTE/541458》 * |
丁洁: "基于最大概率分词算法的中文分词方法研究", 《科技信息》 * |
张泽亚 等: "股票预测:一种基于新闻特征抽取和循环神经网络的方法", 《HTTPS://ARXIV.ORG/FTP/ARXIV/PAPERS/1707/1707.07585.PDF》 * |
戴成骏: "基于文本情感分析的股价预测研究与实现", 《中国优秀硕士学位论文全文数据库 经济与管理科学辑》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110009128A (en) * | 2019-01-28 | 2019-07-12 | 平安科技(深圳)有限公司 | Industry public opinion index prediction technique, device, computer equipment and storage medium |
CN110059851A (en) * | 2019-03-08 | 2019-07-26 | 平安科技(深圳)有限公司 | The method, apparatus and computer equipment of prediction data variation based on deep learning |
Also Published As
Publication number | Publication date |
---|---|
WO2019223133A1 (en) | 2019-11-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111506801B (en) | Sequencing method and device for application App neutron application | |
CN110516910A (en) | Declaration form core based on big data protects model training method and core protects methods of risk assessment | |
CN112148987A (en) | Message pushing method based on target object activity and related equipment | |
CN106875206A (en) | Acquisition of information, assessment, questionnaire method, device and server | |
CN110135942A (en) | Products Show method, apparatus and computer readable storage medium | |
CN109345130A (en) | Method, apparatus, computer equipment and the storage medium of Market Site Selection | |
CN110852785B (en) | User grading method, device and computer readable storage medium | |
CN108876604A (en) | Stock market's Risk Forecast Method, device, computer equipment and storage medium | |
US20180012239A1 (en) | Data Pre-Processing and Searching Systems | |
CN109615280A (en) | Employee's data processing method, device, computer equipment and storage medium | |
CN109614599A (en) | Report form generation method, device, computer equipment and storage medium | |
CN112184304A (en) | Method, system, server and storage medium for assisting decision | |
CN107679680A (en) | A kind of financial forward prediction method, apparatus, equipment and storage medium | |
CN104516938A (en) | Electronic computing device and personalized information recommendation method thereof | |
CN108776846A (en) | Recommend method, apparatus, computer equipment and storage medium | |
CN108509492A (en) | Big data processing based on real estate industry and system | |
CN115545712A (en) | Fraud prediction method, device, equipment and storage medium for transaction behaviors | |
CN109492863A (en) | The automatic generation method and device of financial document | |
CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
CN111126629A (en) | Model generation method, system, device and medium for identifying brushing behavior | |
WO2020111376A1 (en) | Machine learning-based default prediction device and method | |
CN105335886A (en) | Method and device for processing financial data | |
CN109146166A (en) | A kind of personal share based on the marking of investor's content of the discussions slumps prediction model | |
CN112634062B (en) | Hadoop-based data processing method, device, equipment and storage medium | |
CN106354822A (en) | Method and device for acquiring target user |
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: 20181123 |
|
RJ01 | Rejection of invention patent application after publication |