CN113435204A - Stock price fluctuation prediction method based on news information - Google Patents
Stock price fluctuation prediction method based on news information Download PDFInfo
- Publication number
- CN113435204A CN113435204A CN202110143641.2A CN202110143641A CN113435204A CN 113435204 A CN113435204 A CN 113435204A CN 202110143641 A CN202110143641 A CN 202110143641A CN 113435204 A CN113435204 A CN 113435204A
- Authority
- CN
- China
- Prior art keywords
- stock
- information
- news
- word
- prediction
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Economics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a stock price fluctuation prediction method based on news information, and relates to the fields of financial science and technology and computer artificial intelligence. S1, converting news content information into Word vectors, and converting Word information after Word segmentation into Word vectors for representation through Word2vec by training the Word vectors; s2, extracting the emotional characteristics of the news content through the LSTM; s3, stock technical index feature extraction is carried out by using an LSTM neural network; s4, model prediction: and inputting the generated emotional characteristic value and the technical index data characteristics obtained by training into a BP neural network for prediction and generating a transaction instruction according to a prediction result. The invention combines news information with technical indexes, utilizes a neural network model to predict the stock price fluctuation, utilizes a text analysis method to extract emotional characteristics from the forecast and combines the emotional characteristics with technical analysis, improves the accuracy of stock price direction forecast and gives correct guidance to investors.
Description
Technical Field
The invention belongs to the fields of financial science and technology and computer artificial intelligence, and particularly relates to a stock price fluctuation prediction method based on news information.
Background
The prediction of stock prices is of great significance in the business and financial fields. Stock market forecasting has received a great deal of attention in both the business and academic communities. Fama proposed an effective Market Hypothesis (effective Market hypthesis) in 1965 that he considered the stock Market to be an "effective information" Market, with stock prices adequately reflecting events that have occurred and the impact on stock prices of those events that have not occurred but are expected to occur in the Market. This assumption provides the basis for future stock price forecasting efforts.
However, predicting the price of a stock remains difficult because the price of a stock is affected by a number of factors. For a single stock, in addition to the macro factors such as the national currency policy and the business landscape, the micro factors such as the related events of the stock marketing company also have an influence on the stock price. Therefore, in addition to price information of stocks themselves, many related studies use news information related to stocks as an important basis for predicting stock prices.
As can be seen from the complexity of the stock market, the technical index of the stock market is only taken into consideration for prediction, and the obtained result is not ideal even if the prediction is carried out by using a machine learning or deep learning method. The invention integrates the emotional analysis characteristics of news information to predict the rise and fall of the stock price on the premise of considering the technical index of the stock. The invention adopts a recurrent neural network method to extract the emotional polarity characteristics of news information, takes the emotional analysis result and technical indexes as characteristic values of stock price direction prediction, and adopts a BP neural network method to predict.
Disclosure of Invention
The invention provides a stock price fluctuation prediction method based on news information, which solves the problems.
In order to solve the technical problems, the invention is realized by the following technical scheme:
the invention relates to a stock price rise and fall forecasting method based on news information, which comprises the following steps:
s1, converting news content information into word vectors: performing word segmentation on the preprocessed text data by using a crust word segmentation tool and a self-built emotion dictionary; training Word vectors by using a Word2vec tool, and converting Word information after Word segmentation into Word vector representation through Word2 vec;
s2, extracting emotional features of news content: sequentially inputting the word vectors into the LSTM according to the sequence in the text, converting the variable-length input into the fixed-length output by the LSTM, and taking the output of the LSTM at the last moment as the emotional characteristic;
s3, stock technical index feature extraction: aiming at the input technical index data, an LSTM neural network is adopted to capture the time series characteristics of stock technical indexes, and vectors formed by the technical indexes are used as the input of the cyclic neural network at each moment;
s4, model prediction: and inputting the generated emotional characteristic value and the technical index data characteristics obtained by training into a BP neural network for prediction, finally outputting a prediction result, and generating an exchange instruction according to the prediction result.
Further, the step S1 specifically includes:
s11, acquiring historical news public opinion information and market historical technical index data;
s12, preprocessing the text data, and performing word segmentation on the preprocessed text data by using a crust word segmentation tool and combining with a self-built emotion dictionary to obtain word information;
s13, training Word information by using a Word2vec tool, and converting the Word information after Word segmentation into Word vector representation by using Word2 vec.
Further, the step S3 specifically includes:
s31, cleaning the technical indexes, including checking data consistency, processing invalid values and missing values, and preprocessing the cleaned data;
s32, capturing the time series characteristics of the preprocessed technical index data by using LSTM;
and S33, extracting the last training result as the training characteristic of the prediction model.
Further, the step S4 specifically includes:
s41, standardizing the emotion analysis characteristics and the technical index characteristics, and eliminating the dimensional relationship between stock transaction data and emotion data;
s42, establishing a prediction model for future stock price fluctuation based on the BP neural network;
s43, acquiring real-time stock market news information text and technical index data, extracting news content emotional characteristics and stock technical index characteristics, and performing standardization processing;
s44, inputting the normalized emotional characteristics and the normalized technical index characteristics into a prediction model, and outputting a prediction result;
s45, building a stock pool based on the prediction result;
s46, predicting the stocks in the stock pool again to obtain trading signals;
s47, constructing a trading instruction according to the position holding condition of the stock;
and S48, executing the transaction instruction.
Further, when the historical news public opinion information and the market historical technology index data are specifically acquired in the step S11, the information of news websites and forums with high attention is captured by python; news or subject postings which cannot express the emotion of the investor are removed through writing cleaning rules, wherein the cleaning rules comprise four related types of pictures without characters, links without characters, careless random symbols and real disk combinations automatically generated by a system.
Further, the technical index data in step S3 includes stock index, closing price, opening price, fluctuation range, volume of transaction, index smooth iso-mean line MACD, random index KDJ, relative intensity index RSI, and change rate index ROC of large disc.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention combines news information and technical indexes, and utilizes a neural network model to predict the stock price fluctuation; the marginal benefit of the traditional method for predicting stock price trend, namely technical analysis and basic analysis, is gradually reduced, news public sentiment is used as effective investment information, the emotional characteristics can be extracted from the effective investment information by using a text analysis method to serve as incremental information, and the incremental information is combined with the technical analysis, so that the accuracy of stock price direction prediction can be improved to a great extent, and the investor is guided positively.
2. The invention selects the recurrent neural network method to carry out sentiment analysis on the collected news content text data, because the recurrent neural network is suitable for large data volume, has the characteristics of memorability and parameter sharing, and is a recurrent neural network which takes sequence data as input and all recurrent units are connected in a chain way to form a closed loop when recursion is carried out in the evolution direction of the sequence. Meanwhile, natural language data is typical sequence data, the cyclic neural network has inherent advantages in processing the sequence data, and the LSTM can extract more complex semantic feature information and has good expression on the emotion analysis task.
Of course, it is not necessary for any product in which the invention is practiced to achieve all of the above-described advantages at the same time.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of a method for predicting stock price fluctuation based on news information according to the present invention;
FIG. 2 is a flowchart illustrating steps of a process according to an embodiment of the present invention based on the process shown in FIG. 1;
FIG. 3 is a schematic flow chart of the emotional feature extraction of news content based on the method of the present invention;
FIG. 4 is a schematic diagram of the principle of the LSTM neural network of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1 to 4, the method for predicting stock price fluctuation based on news information according to the present invention includes the steps of:
s1, converting news content information into word vectors: performing word segmentation on the preprocessed text data by using a crust word segmentation tool and a self-built emotion dictionary; training Word vectors by using a Word2vec tool, and converting Word information after Word segmentation into Word vector representation through Word2 vec;
s2, extracting emotional features of news content: sequentially inputting the word vectors into the LSTM according to the sequence in the text, converting the variable-length input into the fixed-length output by the LSTM, and taking the output of the LSTM at the last moment as the emotional characteristic;
s3, stock technical index feature extraction: aiming at the input technical index data, an LSTM neural network is adopted to capture the time series characteristics of stock technical indexes, and vectors formed by the technical indexes are used as the input of the cyclic neural network at each moment;
s4, model prediction: and inputting the generated emotional characteristic value and the technical index data characteristics obtained by training into a BP neural network for prediction, finally outputting a prediction result, and generating an exchange instruction according to the prediction result.
Wherein, step S1 specifically includes:
s11, acquiring historical news public opinion information and market historical technical index data;
s12, preprocessing the text data, and performing word segmentation on the preprocessed text data by using a crust word segmentation tool and combining with a self-built emotion dictionary to obtain word information;
s13, training Word information by using a Word2vec tool, and converting the Word information after Word segmentation into Word vector representation by using Word2 vec.
Wherein, step S3 specifically includes:
s31, cleaning the technical indexes, including checking data consistency, processing invalid values and missing values, and preprocessing the cleaned data;
s32, capturing the time series characteristics of the preprocessed technical index data by using LSTM;
and S33, extracting the last training result as the training characteristic of the prediction model.
Wherein, step S4 specifically includes:
s41, standardizing the emotion analysis characteristics and the technical index characteristics, and eliminating the dimensional relationship between stock transaction data and emotion data;
s42, establishing a prediction model for future stock price fluctuation based on the BP neural network;
s43, acquiring real-time stock market news information text and technical index data, extracting news content emotional characteristics and stock technical index characteristics, and performing standardization processing;
s44, inputting the normalized emotional characteristics and the normalized technical index characteristics into a prediction model, and outputting a prediction result;
s45, building a stock pool based on the prediction result;
s46, predicting the stocks in the stock pool again to obtain trading signals;
s47, constructing a trading instruction according to the position holding condition of the stock;
and S48, executing the transaction instruction.
When the historical news public opinion information and the market historical technical index data are obtained in the step S11, python is used for capturing information of news websites and forums with high attention, such as a 'New wave finance news network' and an 'east wealth network', and the like; news or subject postings which cannot express the emotion of the investor are removed by writing cleaning rules, wherein the cleaning rules comprise four related types of pictures without characters, links without characters, careless expressed random symbols and real disk combinations automatically generated by a system.
Wherein, the technical index data in the step S3 includes stock index, closing price, opening price, fluctuation range, volume of trade, index smooth iso-mean line MACD, random index KDJ, relative strength index RSI and change rate index ROC of large disk; the technical indexes such as MACD, KDJ, RSI, ROC and the like are selected as the characteristic values, because the trend of rising or falling presented by MACD is delayed compared with the release time of real-time quotation, and the lagging is a big disadvantage, so investors usually match with RSI to master the time of entering and exiting. KDJ can not only reflect the level of overbilling and overbilling in the market, but also send out buying and selling signals through cross breakthrough, meanwhile, KDJ has extremely high accuracy in the prediction of the large disk and the popular large disk stock, and the combined use of the three can well control the trend. In the weak market, ROC has a high value of actual combat.
As shown in FIG. 4, for the input technical index data, the invention adopts the LSTM neural network to capture the time series characteristics of the stock technical index. X for technical index dataiRepresenting input, input into the LSTM training model and trained by control of the "gate", i.e., D ═ xi,x2,...,x10)。yiThe sample at time i is indicated. Wherein the output h of the previous moment9Is x10And (4) inputting the time. The LSTM neural network has memory cells that can hold some meaningful information for long periods of time and control the characteristics of the information through a "gate" (output gate, control gate, forgetting gate) structure. In the process of correcting the weight, some errors can be directly forgotten through the control of the gate structure of the LSTM, and better performance is realized when the timing problem is processed.
Has the advantages that:
1. the invention combines news information and technical indexes, and utilizes a neural network model to predict the stock price fluctuation; the marginal benefit of the traditional method for predicting stock price trend, namely technical analysis and basic analysis, is gradually reduced, news public sentiment is used as effective investment information, the emotional characteristics can be extracted from the effective investment information by using a text analysis method to serve as incremental information, and the incremental information is combined with the technical analysis, so that the accuracy of stock price direction prediction can be improved to a great extent, and the investor is guided positively.
2. The invention selects the recurrent neural network method to carry out sentiment analysis on the collected news content text data, because the recurrent neural network is suitable for large data volume, has the characteristics of memorability and parameter sharing, and is a recurrent neural network which takes sequence data as input and all recurrent units are connected in a chain way to form a closed loop when recursion is carried out in the evolution direction of the sequence. Meanwhile, natural language data is typical sequence data, the cyclic neural network has inherent advantages in processing the sequence data, and the LSTM can extract more complex semantic feature information and has good expression on the emotion analysis task.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (6)
1. A stock price rise and fall prediction method based on news information is characterized by comprising the following steps:
s1, converting news content information into word vectors: performing word segmentation on the preprocessed text data by using a crust word segmentation tool and a self-built emotion dictionary; training Word vectors by using a Word2vec tool, and converting Word information after Word segmentation into Word vector representation through Word2 vec;
s2, extracting emotional features of news content: sequentially inputting the word vectors into the LSTM according to the sequence in the text, converting the variable-length input into the fixed-length output by the LSTM, and taking the output of the LSTM at the last moment as the emotional characteristic;
s3, stock technical index feature extraction: aiming at the input technical index data, an LSTM neural network is adopted to capture the time series characteristics of stock technical indexes, and vectors formed by the technical indexes are used as the input of the cyclic neural network at each moment;
s4, model prediction: and inputting the generated emotional characteristic value and the technical index data characteristics obtained by training into a BP neural network for prediction, finally outputting a prediction result, and generating a transaction instruction according to the prediction result.
2. The method for predicting stock price rise and fall based on news information as claimed in claim 1, wherein said step S1 specifically includes:
s11, acquiring historical news public opinion information and market historical technical index data;
s12, preprocessing the text data, and performing word segmentation on the preprocessed text data by using a crust word segmentation tool and combining with a self-built emotion dictionary to obtain word information;
s13, training Word information by using a Word2vec tool, and converting the Word information after Word segmentation into Word vector representation by using Word2 vec.
3. The method for predicting stock price rise and fall based on news information as claimed in claim 1, wherein said step S3 specifically includes:
s31, cleaning the technical indexes, including checking data consistency, processing invalid values and missing values, and preprocessing the cleaned data;
s32, capturing the time series characteristics of the preprocessed technical index data by using LSTM;
and S33, extracting the last training result as the training characteristic of the prediction model.
4. The method for predicting stock price rise and fall based on news information as claimed in claim 1, wherein said step S4 specifically includes:
s41, standardizing the emotion analysis characteristics and the technical index characteristics, and eliminating the dimensional relationship between stock transaction data and emotion data;
s42, establishing a prediction model for future stock price fluctuation based on the BP neural network;
s43, acquiring real-time stock market news information text and technical index data, extracting news content emotional characteristics and stock technical index characteristics, and performing standardization processing;
s44, inputting the normalized emotional characteristics and the normalized technical index characteristics into a prediction model, and outputting a prediction result;
s45, building a stock pool based on the prediction result;
s46, predicting the stocks in the stock pool again to obtain trading signals;
s47, constructing a trading instruction according to the position holding condition of the stock;
and S48, executing the transaction instruction.
5. The method for predicting stock price rise and fall based on news information as claimed in claim 1, wherein the step S11 is performed by capturing information of news websites and forums with high attention using python when obtaining historical news public opinion information and market historical technical index data; news or subject postings which cannot express the emotion of the investor are removed by writing cleaning rules, wherein the cleaning rules comprise four related types of pictures without characters, links without characters, careless random symbols and real disk combinations automatically generated by a system.
6. The method of claim 1, wherein the technical index data in step S3 includes stock index, closing price, opening price, rising and falling amplitude, volume of trades, index smooth iso-mean line MACD, random index KDJ, relative intensity index RSI, and change rate index ROC of large disk.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110143641.2A CN113435204A (en) | 2021-02-02 | 2021-02-02 | Stock price fluctuation prediction method based on news information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110143641.2A CN113435204A (en) | 2021-02-02 | 2021-02-02 | Stock price fluctuation prediction method based on news information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113435204A true CN113435204A (en) | 2021-09-24 |
Family
ID=77752729
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110143641.2A Pending CN113435204A (en) | 2021-02-02 | 2021-02-02 | Stock price fluctuation prediction method based on news information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113435204A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114048929A (en) * | 2022-01-12 | 2022-02-15 | 深圳希施玛数据科技有限公司 | Stock price data prediction method and device |
| CN114519613A (en) * | 2022-02-22 | 2022-05-20 | 平安科技（深圳）有限公司 | Price data processing method and device, electronic equipment and storage medium |
| WO2024021354A1 (en) * | 2022-07-28 | 2024-02-01 | 中国科学院深圳先进技术研究院 | Model training method, price prediction method, terminal device and storage medium |
-
2021
- 2021-02-02 CN CN202110143641.2A patent/CN113435204A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114048929A (en) * | 2022-01-12 | 2022-02-15 | 深圳希施玛数据科技有限公司 | Stock price data prediction method and device |
| CN114519613A (en) * | 2022-02-22 | 2022-05-20 | 平安科技（深圳）有限公司 | Price data processing method and device, electronic equipment and storage medium |
| CN114519613B (en) * | 2022-02-22 | 2023-07-25 | 平安科技（深圳）有限公司 | Price data processing method and device, electronic equipment and storage medium |
| WO2024021354A1 (en) * | 2022-07-28 | 2024-02-01 | 中国科学院深圳先进技术研究院 | Model training method, price prediction method, terminal device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Transformer-based attention network for stock movement prediction | |
| Zhang et al. | Deeplob: Deep convolutional neural networks for limit order books | |
| Serafini et al. | Sentiment-driven price prediction of the bitcoin based on statistical and deep learning approaches | |
| CN113435204A (en) | Stock price fluctuation prediction method based on news information | |
| Yujun et al. | A hybrid prediction method for stock price using LSTM and ensemble EMD | |
| Zou et al. | PreBit—A multimodal model with Twitter FinBERT embeddings for extreme price movement prediction of Bitcoin | |
| Wu et al. | Optimized deep learning framework for water distribution data-driven modeling | |
| Wang et al. | A prediction model of CNN-TLSTM for USD/CNY exchange rate prediction | |
| CN115423499A (en) | Model training method, price prediction method, terminal device, and storage medium | |
| Besarabov et al. | Predicting digital asset market based on blockchain activity data | |
| Negi et al. | Cryptocurrency Price Analysis using Deep Learning | |
| Yang | Research on financial credit evaluation and early warning system of internet of things driven by computer-aided technology | |
| CN117034019B (en) | Service processing method and device, electronic equipment and storage medium | |
| Cui et al. | A novel convolutional neural networks for stock trading based on DDQN algorithm | |
| Chantona et al. | News sentiment analysis in forex trading using r-cnn on deep recurrent q-network | |
| Sutkatti et al. | Stock market forecasting techniques: A survey | |
| Yangyudongnanxin | Financial credit risk control strategy based on weighted random forest algorithm | |
| Sakarwala et al. | Use advances in data science and computing power to invest in stock market | |
| Liu et al. | A Sentence‐Level Joint Relation Classification Model Based on Reinforcement Learning | |
| Girsang | Cryptocurrency Price Prediction Based Social Network Sentiment Analysis Using LSTM-GRU and FinBERT | |
| Ruhal et al. | A Comparative Study Of Statistical Methods And Machine Learning Approaches For Stock Price Prediction | |
| CN114328641A (en) | Prevention network gambling monitoring system based on multi-source information fusion | |
| Begicheva et al. | Bank transactions embeddings help to uncover current macroeconomics | |
| CN112348257A (en) | Election prediction method driven by multi-source data fusion and time sequence analysis | |
| Nakano et al. | Enhancing Sentiment Analysis based Investment by Large Language Models in Japanese Stock Market |
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: 20210924 |
|
| RJ01 | Rejection of invention patent application after publication |