CN108287819A - A method of realizing that financial and economic news is automatically associated to stock - Google Patents
A method of realizing that financial and economic news is automatically associated to stock Download PDFInfo
- Publication number
- CN108287819A CN108287819A CN201810030580.7A CN201810030580A CN108287819A CN 108287819 A CN108287819 A CN 108287819A CN 201810030580 A CN201810030580 A CN 201810030580A CN 108287819 A CN108287819 A CN 108287819A
- Authority
- CN
- China
- Prior art keywords
- financial
- economic news
- news information
- stock
- original
- 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention discloses a kind of method realized financial and economic news and be automatically associated to stock, including：By the way of labeled data and pending data, original financial and economic news information is handled；The original financial and economic news information is pre-processed, including the data mode that the original financial and economic news information data conversion is learnt for convenience of successive depths includes multi-C vector form etc.；By the way of multi-tag disaggregated model, the deep learning financial and economic news information after pretreatment；The good model of application training is handled new financial and economic news information, the relationship of auto-associating financial and economic news information and stock；To reach realization can in conjunction with participle mode, deep learning mode, realize high accuracy it is automatic by original financial and economic news information association to the associated stock of needs, facilitate user to obtain the full spectrum information of stock of interest.
Description
Technical field
The present invention relates to stock technical field more particularly to a kind of methods realized financial and economic news and be automatically associated to stock.
Background technology
Personal share news refers to the financial and economic news for being associated with specific stock.
Auto-associating refers to that the original news obtained from metadata provider is not associated and is not associated with to stock or fully institute
Have should associated stock or be associated with mistake stock, increase/correct the mistake of this incidence relation automatically with program strategy
Journey is known as auto-associating.
Existing financial and economic news is automatically associated to the scheme of stock, including：
Scheme one, manual association：
Increase/correct article by artificial judgment to be associated with stock, the disadvantage is that human cost is huge.
Scheme two, based on crucial word association：
Every stock presets fixed several keywords, such as stock name can be used as keyword, if article title or
There is corresponding keyword and is then considered as relevant in content, and the shortcomings that this programme two is:
1, keyword is manually safeguarded, often imperfect；
2, because only mechanical matching, error rate can be higher, such as easily by the news of " Yi Lianzhong (300096.SZ) " company
It is associated under " connection is many (06899.HK) ".
Invention content
The present invention provides a kind of method realized financial and economic news and be automatically associated to stock, of the existing technology to solve
Keyword is manually safeguarded, often imperfect, because of only mechanical matching, the higher technical problem of error rate.
In order to solve the above technical problems, the present invention provides a kind of method realized financial and economic news and be automatically associated to stock,
It is characterized in that, including：
By the way of labeled data and pending data, original financial and economic news information is handled；
The original financial and economic news information is pre-processed, including by the original financial and economic news information data conversion for convenience of after
The data mode of continuous deep learning includes multi-C vector form；
By the way of multi-tag disaggregated model, financial and economic news information after pretreatment described in deep learning；
The good model of application training is handled new financial and economic news information, auto-associating financial and economic news information and stock
The relationship of ticket.
Wherein, the pretreatment original financial and economic news information, including the original financial and economic news information data is turned
It includes multi-C vector form to be changed to the data mode for facilitating successive depths to learn, including：
For original financial and economic news information, html labels are removed：Original financial and economic news information is in general html documents,
Html labels all do not act on following model training and application, therefore need to remove；
For original financial and economic news information, participle and removal stop words：Facilitate making dictionary after participle, removal stop words with
Just subsequent arithmetic amount is reduced；
For original financial and economic news information, dictionary is made；
By the original financial and economic news information data conversion for convenience of the data mode of successive depths study include multidimensional to
Amount form.
Wherein, described to be directed to original financial and economic news information, html labels are removed, including：
Html labels are removed using the htmlstrip-charfilter modes of ES for original financial and economic news information.
Wherein, described to be directed to original financial and economic news information, it segments and removes stop words, including：
For original financial and economic news information, by the way of the ik segmenter of ES, participle and removal stop words.
Wherein, described to be directed to original financial and economic news information, dictionary is made, including：
For original financial and economic news information, coarseness screens keyword；
According to the keyword that the coarseness filters out, dictionary is generated.
Wherein, the keyword filtered out according to the coarseness generates dictionary, including：
According to the keyword that the coarseness filters out, the final vocabulary list sorting that will be filtered out, the serial number of each vocabulary
As the digital code of the vocabulary, the dictionary that vocabulary is mapped to digital code is generated.
Wherein, described by the way of multi-tag disaggregated model, financial and economic news letter after pretreatment described in deep learning
Breath, including：
By the way of multi-tag disaggregated model, training dataset is selected；
According to the selected training dataset, the financial and economic news information after pretreatment is carried out at PCA dimensionality reductions
Reason；
According to the financial and economic news information after PCA dimension-reduction treatment, by the way of deep learning disaggregated model, depth
Learn the financial and economic news information.
Wherein, the financial and economic news information according to described after PCA dimension-reduction treatment, using deep learning disaggregated model
Mode, financial and economic news information described in deep learning, including：
According to the financial and economic news information after PCA dimension-reduction treatment, by the way of deep learning disaggregated model, use
The mode of full Connection Neural Network, financial and economic news information described in deep learning.
Wherein, the good model of the application training is handled new financial and economic news information, auto-associating financial and economic news
The relationship of information and stock, including：
The good model of application training is handled new financial and economic news information, auto-associating financial and economic news information and stock
The relationship of ticket, when the degree of association is more than certain threshold value, and original data source does not indicate this incidence relation, then is associated with and closes in supplement
System.
Wherein, the good model of the application training is handled new financial and economic news information, auto-associating financial and economic news
The relationship of information and stock, including：
The good model of application training is handled new financial and economic news information, auto-associating financial and economic news information and stock
The relationship of ticket, when the degree of association is less than certain threshold value, and original data source designates corresponding incidence relation, then deletes this association and close
System.
The method provided by the invention realized financial and economic news and be automatically associated to stock, including：Using labeled data and wait locating
The mode for managing data, handles original financial and economic news information；The original financial and economic news information is pre-processed, including the original finance and economics is new
The data mode for hearing information data conversion for convenience of successive depths study includes multi-C vector form etc.；Using multi-tag classification mould
The mode of type, the deep learning financial and economic news information after pretreatment；The good model of application training believes new financial and economic news
Breath is handled, the relationship of auto-associating financial and economic news information and stock；It can be in conjunction with participle mode, depth to reach realization
Mode of learning, realize high accuracy it is automatic by original financial and economic news information association to associated stock is needed, facilitate user to obtain
Take the full spectrum information of stock of interest.
Description of the drawings
Fig. 1 is that the present invention realizes that financial and economic news is automatically associated to the flow diagram of the method for stock.
Specific implementation mode
With reference to the accompanying drawings and examples, the present invention is described in further detail.It is emphasized that following implement
Example is merely to illustrate the present invention, but is not defined to the scope of the present invention.Likewise, following embodiment is only the portion of the present invention
Point embodiment and not all embodiments, the institute that those of ordinary skill in the art are obtained without creative efforts
There are other embodiments, shall fall within the protection scope of the present invention.
The present invention provides a kind of method realized financial and economic news and be automatically associated to stock.
Fig. 1 is referred to, Fig. 1 is that the present invention realizes that financial and economic news is automatically associated to the flow diagram of the method for stock, this
The method that financial and economic news is automatically associated to the method for stock is realized in invention, including：
S101：By the way of labeled data and pending data, original financial and economic news information is handled.
S102：The original financial and economic news information is pre-processed, including for convenience by the original financial and economic news information data conversion
The data mode of successive depths study includes multi-C vector form etc..
S103：By the way of multi-tag disaggregated model, the deep learning financial and economic news information after pretreatment.
S104：The good model of application training handles new financial and economic news information, auto-associating financial and economic news information
With the relationship of stock.
Wherein, the original financial and economic news information is pre-processed, including for convenience by the original financial and economic news information data conversion
The data mode of successive depths study includes multi-C vector form etc., including：
For the original financial and economic news information, html labels are removed：Original financial and economic news information is in general html
Document, html labels all do not act on following model training and application, therefore need to remove；
For the original financial and economic news information, participle and removal stop words：Making dictionary, removal is facilitated to deactivate after participle
Word is to reduce subsequent arithmetic amount；
For the original financial and economic news information, dictionary is made；
The data mode that the original financial and economic news information data conversion is learnt for convenience of successive depths includes multi-C vector
Form etc..
Wherein, for the original financial and economic news information, html labels are removed, including：
Html marks are removed using the htmlstrip-charfilter modes of ES for the original financial and economic news information
Label.
Wherein, for the original financial and economic news information, participle and removal stop words, including：
For the original financial and economic news information, by the way of the ik segmenter of ES, participle and removal stop words.
Wherein, for the original financial and economic news information, dictionary is made, including：
For the original financial and economic news information, coarseness screens keyword；
According to the keyword that the coarseness filters out, dictionary is generated.
Wherein, the keyword filtered out according to the coarseness generates dictionary, including：
According to the keyword that the coarseness filters out, the serial number of the final vocabulary list sorting that will be filtered out, each vocabulary is made
For the digital code of the vocabulary, the dictionary that vocabulary is mapped to digital code is generated.
Wherein, by the way of multi-tag disaggregated model, the deep learning financial and economic news information after pretreatment, packet
It includes：
By the way of multi-tag disaggregated model, training dataset is selected；
According to the training dataset of the selection, PCA dimension-reduction treatment is carried out to the financial and economic news information after pretreatment；
According to the financial and economic news information after PCA dimension-reduction treatment, by the way of deep learning disaggregated model, depth
Practise the financial and economic news information.
Wherein, the financial and economic news information according to this after PCA dimension-reduction treatment, by the way of deep learning disaggregated model,
The deep learning financial and economic news information, including：
According to the financial and economic news information after PCA dimension-reduction treatment, by the way of deep learning disaggregated model, using complete
The mode of Connection Neural Network, the deep learning financial and economic news information.
Wherein, the good model of application training handles new financial and economic news information, auto-associating financial and economic news information
With the relationship of stock, including：
The good model of application training is handled new financial and economic news information, auto-associating financial and economic news information and stock
The relationship of ticket, when the degree of association is more than certain threshold value, and original data source does not indicate this incidence relation, then is associated with and closes in supplement
System.
The good model of application training is handled new financial and economic news information, auto-associating financial and economic news information and stock
The relationship of ticket further includes：
The good model of application training is handled new financial and economic news information, auto-associating financial and economic news information and stock
The relationship of ticket, when the degree of association is less than certain threshold value, and original data source designates corresponding incidence relation, then deletes this association and close
System.
Keyword is screened about the coarseness, including：
Although having been removed stop words, but still remaining a large amount of vocabulary need here to reduce follow-up machine learning operand
It wants coarseness to screen, apparent unrelated vocabulary is filtered, needs an evaluation measures herein, it is problem neck to assess each vocabulary
Domain relative words calculate vocabulary criticality with following formula herein：
The number that word frequency=certain vocabulary occurs in total sample news list；
Normalize word frequency=certain vocabulary word frequency/highest word frequency (word frequency of the vocabulary of highest word frequency)；
Doubtful criticality 1=normalization word frequency (note:The more high doubtful criticality of word frequency is higher)；
General geological coodinate system=article the sum comprising the vocabulary/sample news article sum；
Doubtful criticality 2=log (1/ general geological coodinate system) (note:The more high doubtful criticality of general geological coodinate system is lower)；
Criticality=λ1* doubtful key 1+ (1- λ1) the doubtful 2 (notes of criticality of *：λ1For weight, value 0.6 here).
Because criticality only weighs roughly the significance level of some word, inaccurately, therefore there is following principle here:
1, it need to individually be calculated according to labeled data, the related news sample of each stock.In order to avoid global calculation has ignored certain
The characteristic information of stock individually calculates and individually screens key vocabularies, finally the key vocabularies of each stock are incorporated as
Final dictionary；
2, absolute index should not be determined according to criticality screening, such as every stock retains top200 criticality vocabulary, operation
It can retain under ability enabled condition more, rather than retain the vocabulary that criticality is more than certain concrete numerical value.
The data mode that the original financial and economic news information data conversion is learnt for convenience of successive depths includes multi-C vector
Form etc..
It includes more to convert the original financial and economic news information data for convenience of data mode of successive depths study about this
Dimensional vector form etc., including：
There are dictionary, every financial and economic news to can be exchanged into vector, such as a piece of news article includes following vocabulary：
" 2 bight area 3 power-assisted * of * of Zhejiang *, 3 the Belt and Road * 5 rise * 1... " note:* X indicates that the vocabulary occurs X times；
Wherein " power-assisted rise " be not in dictionary, therefore does not consider, remaining 3 vocabulary serial number in dictionary is respectively：
Zhejiang：12 bight areas：13 the Belt and Road：20；
Then being converted to vector form is：
(0,0 ..., 2,3 ..., 5 ...) note：" 2 " therein are located at the 12nd dimension of vector, and the rest may be inferred by analogy.
About by the way of multi-tag disaggregated model, training dataset is selected, including：
The stock incidence relation that some outstanding data sources provide selects such data source extraction almost without mistake
A part is used as training dataset, reserves another part as validation data set and test data set.
About the training dataset according to the selection, which is carried out at PCA dimensionality reductions
Reason, including：
The vector dimension number got by pretreatment stage conversion is larger and very sparse, is very suitable for using tensorflow
The PCA dimension-reduction treatment of support can substantially reduce number of dimensions after processing, reduce neural network node number, to reduce operand,
Over-fitting situation can be mitigated.
It is deep by the way of deep learning disaggregated model about according to the financial and economic news information after PCA dimension-reduction treatment
Degree learns the financial and economic news information, including：
Basic deep learning disaggregated model can be applied mechanically, by the way of full Connection Neural Network, because of same piece wealth
More stocks may be associated with through news, therefore used here as multi-tag disaggregated model, loss function uses tensorflow branch
The sigmoid_cross_entropy_with_logits held.
The method provided by the invention realized financial and economic news and be automatically associated to stock, including：Using labeled data and wait locating
The mode for managing data, handles original financial and economic news information；The original financial and economic news information is pre-processed, including the original finance and economics is new
The data mode for hearing information data conversion for convenience of successive depths study includes multi-C vector form etc.；Using multi-tag classification mould
The mode of type, the deep learning financial and economic news information after pretreatment；The good model of application training believes new financial and economic news
Breath is handled, the relationship of auto-associating financial and economic news information and stock；It can be in conjunction with participle mode, depth to reach realization
Mode of learning, realize high accuracy it is automatic by original financial and economic news information association to associated stock is needed, facilitate user to obtain
Take the full spectrum information of stock of interest.
The method provided by the invention realized financial and economic news and be automatically associated to stock, can arrive financial and economic news information association
Under corresponding stock, there is great user to be worth, user is facilitated to obtain the full spectrum information of stock of interest.
The method provided by the invention realized financial and economic news and be automatically associated to stock, can segment elasticsearch
Technology is used for data prediction, reduces successive depths learning neural network number of nodes, operand is reduced, by the classification of deep learning
Model applies to financial and economic news information and the associated scene of corresponding stock.
It is provided by the invention to realize the financial and economic news method that is automatically associated to stock, it is not limited to do stock and is associated with this part thing,
It can be with：
One, automatic labeling：Label class only need to be predefined, is associated with news article with label with similar approach.
Two, automatic branch industry：Industry class only need to be predefined, is associated with news article with industry class with similar approach.
Three, other that the scene classified automatically is done to financial and economic news.
In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can
To realize by another way.For example, device embodiments described above are only schematical, for example, module or
The division of unit, only a kind of division of logic function, formula that in actual implementation, there may be another division manner, such as multiple units
Or component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, institute
Display or the mutual coupling, direct-coupling or communication connection discussed can be by some interfaces, device or unit
INDIRECT COUPLING or communication connection can be electrical, machinery or other forms.
The unit illustrated as separating component may or may not be physically separated, and be shown as unit
Component may or may not be physical unit, you can be located at a place, or may be distributed over multiple networks
On unit.Some or all of unit therein can be selected according to the actual needs to realize the mesh of present embodiment scheme
's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, also may be used
It, can also be during two or more units be integrated in one unit to be that each unit physically exists alone.It is above-mentioned integrated
The form that hardware had both may be used in unit is realized, can also be realized in the form of SFU software functional unit.
It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in a computer read/write memory medium.Based on this understanding, technical scheme of the present invention substantially or
Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products
Out, which is stored in a storage medium, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) or processor (processor) execute each implementation of the present invention
The all or part of step of methods.And storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. it is various
The medium of program code can be stored.
The foregoing is merely the section Examples of the present invention, are not intended to limit protection scope of the present invention, every utilization
Equivalent device or equivalent process transformation made by description of the invention and accompanying drawing content are applied directly or indirectly in other correlations
Technical field, be included within the scope of the present invention.
Claims (10)
1. a kind of method realized financial and economic news and be automatically associated to stock, which is characterized in that including：
By the way of labeled data and pending data, original financial and economic news information is handled；
The original financial and economic news information is pre-processed, including by the original financial and economic news information data conversion for convenience of follow-up deep
The data mode of degree study includes multi-C vector form；
By the way of multi-tag disaggregated model, financial and economic news information after pretreatment described in deep learning；
The good model of application training handles new financial and economic news information, auto-associating financial and economic news information and stock
Relationship.
2. realizing the method that financial and economic news is automatically associated to stock as described in claim 1, which is characterized in that the pretreatment
The original financial and economic news information includes the number by the original financial and economic news information data conversion for convenience of successive depths study
Include multi-C vector form according to form, including：
For original financial and economic news information, html labels are removed：Original financial and economic news information is in general html documents, html
Label does not all act on following model training and application, therefore need to remove；
For original financial and economic news information, participle and removal stop words：Facilitate making dictionary after participle, removes stop words to drop
Low subsequent arithmetic amount；
For original financial and economic news information, dictionary is made；
The data mode that the original financial and economic news information data conversion is learnt for convenience of successive depths includes multi-C vector shape
Formula.
3. realizing the method that financial and economic news is automatically associated to stock as claimed in claim 2, which is characterized in that described for original
Beginning financial and economic news information removes html labels, including：
Html labels are removed using the htmlstrip-charfilter modes of ES for original financial and economic news information.
4. realizing the method that financial and economic news is automatically associated to stock as claimed in claim 2, which is characterized in that described for original
Beginning financial and economic news information, participle and removal stop words, including：
For original financial and economic news information, by the way of the ik segmenter of ES, participle and removal stop words.
5. realizing the method that financial and economic news is automatically associated to stock as claimed in claim 2, which is characterized in that described for original
Beginning financial and economic news information makes dictionary, including：
For original financial and economic news information, coarseness screens keyword；
According to the keyword that the coarseness filters out, dictionary is generated.
6. realizing the method that financial and economic news is automatically associated to stock as claimed in claim 5, which is characterized in that described according to institute
The keyword that coarseness filters out is stated, dictionary is generated, including：
According to the keyword that the coarseness filters out, the final vocabulary list sorting that will be filtered out, the serial number conduct of each vocabulary
The digital code of the vocabulary generates the dictionary that vocabulary is mapped to digital code.
7. realizing the method that financial and economic news is automatically associated to stock as claimed in claim 5, which is characterized in that described using more
The mode of labeling model, financial and economic news information after pretreatment described in deep learning, including：
By the way of multi-tag disaggregated model, training dataset is selected；
According to the selected training dataset, PCA dimension-reduction treatment is carried out to the financial and economic news information after pretreatment；
According to the financial and economic news information after PCA dimension-reduction treatment, by the way of deep learning disaggregated model, deep learning
The financial and economic news information.
8. realizing the method that financial and economic news is automatically associated to stock as claimed in claim 7, which is characterized in that described according to institute
The financial and economic news information after PCA dimension-reduction treatment is stated, by the way of deep learning disaggregated model, finance and economics described in deep learning is new
Information is heard, including：
According to the financial and economic news information after PCA dimension-reduction treatment, by the way of deep learning disaggregated model, using connecting entirely
Connect the mode of neural network, financial and economic news information described in deep learning.
9. realizing the method that financial and economic news is automatically associated to stock as described in claim 1, which is characterized in that the application instruction
The model perfected is handled new financial and economic news information, the relationship of auto-associating financial and economic news information and stock, including：
The good model of application training handles new financial and economic news information, auto-associating financial and economic news information and stock
Relationship, when the degree of association is more than certain threshold value, and original data source does not indicate this incidence relation, then supplements upper incidence relation.
10. realizing the method that financial and economic news is automatically associated to stock as described in claim 1, which is characterized in that the application
Trained model is handled new financial and economic news information, the relationship of auto-associating financial and economic news information and stock, packet
It includes：
The good model of application training handles new financial and economic news information, auto-associating financial and economic news information and stock
Relationship, when the degree of association is less than certain threshold value, and original data source designates corresponding incidence relation, then deletes this incidence relation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810030580.7A CN108287819A (en) | 2018-01-12 | 2018-01-12 | A method of realizing that financial and economic news is automatically associated to stock |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810030580.7A CN108287819A (en) | 2018-01-12 | 2018-01-12 | A method of realizing that financial and economic news is automatically associated to stock |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108287819A true CN108287819A (en) | 2018-07-17 |
Family
ID=62835187
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810030580.7A Pending CN108287819A (en) | 2018-01-12 | 2018-01-12 | A method of realizing that financial and economic news is automatically associated to stock |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108287819A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110889024A (en) * | 2019-10-25 | 2020-03-17 | 武汉灯塔之光科技有限公司 | Method and device for calculating information-related stock |
| CN112287101A (en) * | 2019-07-23 | 2021-01-29 | 上海第一财经数据科技有限公司 | Information processing method and device and computer equipment |
| CN113722432A (en) * | 2021-08-26 | 2021-11-30 | 杭州隆埠科技有限公司 | Method and device for associating news with stocks |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110069936A1 (en) * | 2002-01-16 | 2011-03-24 | Microsoft Corporation | Data preparation for media browsing |
| CN103226554A (en) * | 2012-12-14 | 2013-07-31 | 西藏同信证券有限责任公司 | Automatic stock matching and classifying method and system based on news data |
| CN107463658A (en) * | 2017-07-31 | 2017-12-12 | 广州市香港科大霍英东研究院 | File classification method and device |
-
2018
- 2018-01-12 CN CN201810030580.7A patent/CN108287819A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110069936A1 (en) * | 2002-01-16 | 2011-03-24 | Microsoft Corporation | Data preparation for media browsing |
| CN103226554A (en) * | 2012-12-14 | 2013-07-31 | 西藏同信证券有限责任公司 | Automatic stock matching and classifying method and system based on news data |
| CN107463658A (en) * | 2017-07-31 | 2017-12-12 | 广州市香港科大霍英东研究院 | File classification method and device |
Non-Patent Citations (2)
| Title |
|---|
| 赵中堂著: "《基于智能移动终端的行为识别方法研究》", 30 April 2015, 电子科技大学出版社 * |
| 郑铁男等著: "《数字编辑运营实训教程》", 30 September 2017, 知识产权出版社 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112287101A (en) * | 2019-07-23 | 2021-01-29 | 上海第一财经数据科技有限公司 | Information processing method and device and computer equipment |
| CN112287101B (en) * | 2019-07-23 | 2024-04-16 | 上海应帆数字科技有限公司 | Information processing method, device and computer equipment |
| CN110889024A (en) * | 2019-10-25 | 2020-03-17 | 武汉灯塔之光科技有限公司 | Method and device for calculating information-related stock |
| CN113722432A (en) * | 2021-08-26 | 2021-11-30 | 杭州隆埠科技有限公司 | Method and device for associating news with stocks |
| CN113722432B (en) * | 2021-08-26 | 2024-01-09 | 杭州隆埠科技有限公司 | Method and device for associating news with stocks |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109190110B (en) | Named entity recognition model training method and system and electronic equipment | |
| CN108959431B (en) | Automatic label generation method, system, computer readable storage medium and equipment | |
| CN109685056B (en) | Method and device for acquiring document information | |
| US7788087B2 (en) | System for processing sentiment-bearing text | |
| CN111767716B (en) | Method and device for determining enterprise multi-level industry information and computer equipment | |
| US20060200341A1 (en) | Method and apparatus for processing sentiment-bearing text | |
| CN111177186B (en) | Single sentence intention recognition method, device and system based on question retrieval | |
| CN111159407A (en) | Method, apparatus, device and medium for training entity recognition and relation classification model | |
| CN106294466A (en) | Disaggregated model construction method, disaggregated model build equipment and sorting technique | |
| CN109766437A (en) | A kind of Text Clustering Method, text cluster device and terminal device | |
| CN109684476A (en) | A kind of file classification method, document sorting apparatus and terminal device | |
| CN104216876A (en) | Informative text filter method and system | |
| CN107844558A (en) | The determination method and relevant apparatus of a kind of classification information | |
| CN108287819A (en) | A method of realizing that financial and economic news is automatically associated to stock | |
| CN106934055B (en) | Semi-supervised webpage automatic classification method based on insufficient modal information | |
| CN108363700A (en) | The method for evaluating quality and device of headline | |
| CN110489753B (en) | Neural structure corresponding learning cross-domain emotion classification method for improving feature selection | |
| WO2020091618A1 (en) | System for identifying named entities with dynamic parameters | |
| CN116775897A (en) | Knowledge graph construction and query method and device, electronic equipment and storage medium | |
| CN110795930A (en) | Article title optimization method, system, medium and equipment | |
| CN116882414A (en) | Automatic comment generation method and related device based on large-scale language model | |
| CN112487186A (en) | Human-human conversation log analysis method, system, equipment and storage medium | |
| CN111951079A (en) | Credit rating method and device based on knowledge graph and electronic equipment | |
| CN111428034A (en) | Training method of classification model, and classification method and device of comment information | |
| CN108733702B (en) | Method, device, electronic equipment and medium for extracting upper and lower relation of user query |
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: 20180717 |
|
| RJ01 | Rejection of invention patent application after publication |