Financial transaction intelligent analysis system

Abstract

The invention discloses a financial transaction intelligent analysis system, and belongs to the technical field of financial transactions. The system comprises a data acquisition module, a data preprocessing and feature engineering module, an intelligent analysis and decision module, a user preference and strategy configuration module, a signal matching and reminding generation module and a notification pushing module. According to the invention, multi-dimensional data related to financial transactions are collected, after preprocessing and feature extraction, a transaction signal is generated by using an algorithm model in the intelligent analysis and decision module, and a precise transaction prompt is generated and pushed to a user in time in combination with personalized preferences and strategies set by the user. The method can help the user to overcome the dilemma of information asymmetry and irrational decision, improves the decision efficiency and potential return rate of financial transactions, and has the characteristics of intelligence, individuation, initiative and timeliness.

Description

Technical Field

[0001] The present invention relates to the field of financial transaction technology, and in particular to a financial transaction intelligent analysis system. Background Art

[0002] Financial markets are volatile and complex. Accurately determining buy and sell timing is crucial for investors to achieve profitability. Traditional financial trading analysis relies on the investor's experience, interpretation of technical indicators, and the ability to access and interpret market information. However, for ordinary investors, lacking the expertise and time to analyze massive amounts of market data and news often leads to missed opportunities or losses caused by emotional decisions.

[0003] Currently, there are several financial trading analysis software programs on the market that can provide market data, technical charts, and information, and some also offer early warning functions based on specific algorithms. However, these programs often suffer from the following shortcomings: information overload and difficulty in filtering, requiring users to make their own judgments from a large amount of information, which places heavy pressure on decision-making; lack of timeliness and proactiveness, with some alerts not being timely enough or requiring users to actively query for advice rather than the system actively pushing it; and limited comprehensive analytical capabilities, with most tools focusing on single technical indicator analysis or simple news aggregation, lacking comprehensive deep learning and intelligent judgment of multi-dimensional information (such as fundamentals, technical analysis, news, and market sentiment), hindering user use, quick decision-making, and investment profitability in the stock market. Summary of the Invention

[0004] The main purpose of this invention is to overcome the shortcomings of the existing technology and provide a financial transaction intelligent analysis system, which is a comprehensive market analysis and trading assistance tool designed for investors in stocks, futures, options and financial markets. The system can comprehensively analyze various market data, use artificial intelligence algorithms to identify potential buying and selling points, and actively and promptly send trading reminders to users based on user preferences to help users improve decision-making efficiency and return on investment.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] The financial transaction intelligent analysis system includes: a data collection module for collecting multi-dimensional data related to financial transactions;

[0007] A data preprocessing and feature engineering module is used to process the collected data and extract effective features, perform preprocessing operations such as cleaning, denoising, normalization, and missing value processing on the collected raw data, and extract or construct features from it that are valuable for determining the timing of financial transactions;

[0008] An intelligent analysis and decision-making module, configured to analyze and generate financial transaction signals based on the effective features through an algorithmic model;

[0009] User preference and strategy configuration module, where users set their investment preferences and trading strategy parameters;

[0010] A signal matching and reminder generation module, used to match the transaction signal with the parameters set by the user and generate transaction reminder information that meets the conditions;

[0011] The notification push module is used to push the transaction reminder information to the user.

[0012] As a preferred technical solution of the present invention, the data collected by the data collection module includes at least one or more combinations of market data, technical indicator data, fundamental data, news information and announcement data;

[0013] Data collection module: used to collect various types of data related to financial transactions from multiple data sources in real time or periodically. The data includes at least:

[0014] Market data: financial transaction prices (opening price, closing price, highest price, lowest price), trading volume, turnover rate, etc.

[0015] Technical indicator data: such as moving average (MA), relative strength index (RSI), moving average convergence divergence (MACD), Bollinger Bands (BOLL), etc.

[0016] Fundamental data: company financial reports, price-to-earnings ratio (PE), price-to-book ratio (PB), earnings per share (EPS), etc.

[0017] News and announcement data: various news and announcements related to listed companies, industry trends, and macroeconomic policies;

[0018] Market sentiment data: A market sentiment index obtained by analyzing text data such as social media and stock reviews using natural language processing technology;

[0019] User behavior data: user historical transaction records, watchlists, risk preference settings, etc.

[0020] As a preferred technical solution of the present invention, the algorithmic model used in the intelligent analysis and decision-making module includes a rule-based expert system, a machine learning model, a deep learning model, or a fusion model thereof. This module is the core and is used to analyze and make decisions based on preprocessed data and extracted features using one or more algorithmic models to identify potential buy and sell signals. The algorithmic model may include but is not limited to:

[0021] Rule-based expert system: trading rules set by combining classic technical analysis theory and expert experience;

[0022] Machine learning models: such as support vector machines (SVMs), decision trees, random forests, gradient boosting machines (GBDTs), and long short-term memory networks (LSTMs), train models using historical data to predict price trends or identify specific patterns; randomly select a sample or a mini-batch of samples in each iteration, calculate the gradient of the loss function, and update the model parameters

[0023] Parameter update formula:

[0024]

[0025] Among them, θ is the model parameter, η is the learning rate, L is the loss function, xi and yi are the features and labels of the i-th sample;

[0026] The momentum term is introduced to enable the model to converge faster and cross the local minimum by adding the direction of the previous gradient to the parameter update.

[0027] Parameter update formula:

[0028]

[0029] Where v is the momentum term and γ is the momentum coefficient (usually 0.9).

[0030] Deep learning models: Utilize neural networks for more complex pattern recognition and trend prediction;

[0031] Fusion model: Fusion of the output results of multiple models to improve the accuracy and robustness of judgment;

[0032] The output of this module is a recommendation signal to buy, sell, or hold a specific financial transaction, along with the corresponding confidence level or trigger conditions.

[0033] As a preferred technical solution of the present invention, the system also includes a user interaction and feedback module for receiving user feedback on transaction reminders and using the feedback information to optimize the algorithm model of the intelligent analysis and decision-making module; and for allowing users to set their investment preferences, such as risk tolerance, expected rate of return, list of financial transactions of interest, trading strategy (e.g., short-term, medium- to long-term), and reminder sensitivity.

[0034] The Signal Matching and Alert Generation Module matches the trading signals generated by the Intelligent Analysis and Decision-Making Module with the user settings in the User Preferences and Strategy Configuration Module. When the signals meet the user-defined conditions, a specific trading alert is generated. This alert may include the financial transaction code, name, recommended action (buy / sell), recommended price range, and a brief description of the triggering reason.

[0035] As a preferred technical solution of the present invention, the notification push module is used to push the generated reminder information to the user in a timely manner through one or more channels, which may include: in-application push (App Push), text message (SMS), email, voice reminder, etc.

[0036] As a preferred technical solution of the present invention, the user interaction and feedback module provides a user interface to display financial transaction information, analysis results, historical reminder records, etc., and allows users to provide feedback on reminders (such as "adopt" or "ignore"). The feedback information can be used to optimize the model in the intelligent analysis and decision-making module.

[0037] A financial transaction intelligent analysis system comprises the following steps:

[0038] S1. Data Collection: The system automatically obtains financial transaction-related market data, technical indicators, fundamental data, news information, market sentiment data, etc. from pre-set data sources;

[0039] S2. Data preprocessing and feature extraction: Clean and format the collected data, and extract or construct features for model analysis;

[0040] S3. User profile loading: Obtain the user's investment preferences, list of preferred financial transactions, and strategy parameters;

[0041] S4. Intelligent Analysis and Signal Generation: The pre-processed data and features are fed into the intelligent analysis and decision-making module, which runs a pre-set algorithm model to analyze financial transactions in the user's watchlist or other financial transactions that meet the screening criteria, generating preliminary buy and sell signals and confidence levels.

[0042] S5. Signal Filtering and Matching: Filter preliminary signals based on user-configured strategy parameters (such as risk level and alert sensitivity) to select trading signals that meet the user's personalized needs;

[0043] S6. Alert Generation and Push: For successfully matched trading signals, a reminder message containing specific action suggestions and reasons is generated and pushed to the user through the selected notification channel;

[0044] S7. User feedback collection and model iteration: Collect user feedback on reminders, and regularly use new data and feedback information to retrain and optimize the models in the intelligent analysis and decision-making module.

[0045] Beneficial effects:

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. Intelligent Decision Support: By integrating advanced AI algorithms and multi-dimensional data analysis, it can more intelligently identify potential trading opportunities, reducing the burden of in-house analysis on investors;

[0048] 2. Personalized service: Users can configure their investment strategies based on their own risk preferences, and the system provides more targeted reminders;

[0049] 3. Proactive and timely reminders: The system can proactively monitor market changes and send timely reminders to help users seize fleeting trading opportunities;

[0050] 4. Improve decision-making efficiency and quality: Structured reminder information and concise reasons help users quickly understand and make more rational investment decisions;

[0051] 5. Continuous learning and optimization: Through user feedback and learning from new data, the accuracy and adaptability of the system model can be continuously improved.

[0052] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 It is a structural block diagram of the financial transaction intelligent analysis system of the present invention.

[0055] Figure 2 It is a schematic diagram of the interface of the financial transaction intelligent analysis system of the present invention. DETAILED DESCRIPTION

[0056] In order to make the technical means, creative features, purpose and efficacy of the present invention easy to understand, the present invention is further described below in conjunction with specific examples, but the following examples are only preferred embodiments of the present invention, not all. Based on the examples in the embodiments, other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention. The experimental methods in the following examples, unless otherwise specified, are conventional methods, and the materials, reagents, etc. used in the following examples, unless otherwise specified, can be obtained from commercial channels.

[0057] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] Reference Figure 1-2 The financial transaction intelligent analysis system 100 in the embodiment of the present invention includes: a data acquisition module 10, a data preprocessing and feature engineering module 20, an intelligent analysis and decision-making module 30, a user preference and strategy configuration module 40, a signal matching and reminder generation module 50, a notification push module 60, and a user interaction and feedback module 70.

[0059] The data collection module 10 is responsible for acquiring data from APIs of major stock exchanges, financial data service providers (such as Wind, Bloomberg, and Eastmoney), news portals, and social media platforms. For example, it can obtain real-time minute-level K-line data for Shanghai, Shenzhen, and Hong Kong A-shares, daily financial statement summaries, and important news announcements.

[0060] The data preprocessing and feature engineering module 20 processes the collected data. For example, it calculates commonly used technical indicators such as the MACD golden cross and dead cross signals and the RSI overbought and oversold areas from K-line data; performs sentiment analysis on news text to quantify market sentiment; and calculates year-on-year and month-on-month financial data.

[0061] The intelligent analysis and decision-making module 30 is the core of the system. For example, a LSTM-based neural network model can be deployed, using historical K-line data, technical indicator sequences, and news sentiment sequences as input to predict the probability of stock prices rising or falling in the future. At the same time, a rule engine can be combined, such as "when the LSTM predicts a rising probability greater than 70%, the MACD forms a golden cross, and a favorable announcement is released recently, a buy signal is generated." This module can use different sub-models or parameters for different types of financial transactions (such as growth stocks, value stocks) or different market conditions (bull market, bear market, volatile market).

[0062] The user preference and strategy configuration module 40 allows user Zhang San to set his risk preference to "conservative", focus on the futures pool as "Shanghai Tin 24XX, Glass 25XX", select the "medium-term holding" strategy, and set the reminder threshold to "high confidence signal".

[0063] The signal matching and reminder generation module 50 receives the "buy signal, confidence level 85%" generated by the intelligent analysis and decision-making module 30 for "Shanghai Tin 24XX". Since 85% is higher than the "high confidence level" set by user Zhang San (e.g., defined as 80%), and "Shanghai Tin 24XX" is in his watch list, an reminder is generated: "[Buy Reminder] Shanghai Tin Solder 24XX (SN24XX) has a mid-term buy signal. The current price is XXXX yuan. It is recommended to pay attention. Trigger reason: technical indicators resonate, and capital inflow is obvious."

[0064] The notification push module 60 sends the above reminder information to the notification bar of Zhang San's mobile phone through the push service bound to the App on his mobile phone.

[0065] The user interaction and feedback module 70 displays this reminder within the app. After seeing it, if Zhang San actually takes action, he can click "Accepted" and the system will record this feedback. If a large number of users have a high adoption rate for a certain type of signal, the weight of this type of signal can be increased in subsequent model optimization.

[0066] Reference Figure 2 The process of the financial transaction intelligent analysis system in the embodiment of the present invention is as follows:

[0067] First, the system continuously acquires and processes the latest market data through the data acquisition module 10 and the data preprocessing and feature engineering module 20 (S1, S2). The user sets his or her own parameters in the user preference and strategy configuration module 40 through the user interaction and feedback module 70 (S3). The intelligent analysis and decision-making module 30 analyzes the latest data and stocks that the user may be interested in (or stocks screened by strategy after a full market scan) to generate trading signals (S4). The signal matching and reminder generation module 50 compares these signals with the user's configuration (S5). If the match is successful, a reminder message is generated and sent to the user through the notification push module 60 (S6). After the user receives the reminder, his or her actions or feedback can be collected for future model iterations (S7).

[0068] For example, when the market is generally declining, but the intelligent analysis and decision-making module 30 detects strong buying at a key support level for a user-focused stock X and significant favorable policies for its related industry, the model comprehensively determines this as a potential counter-trend buy opportunity. At this point, if the user's risk appetite allows and the signal strength reaches the set threshold, the system will generate and push a buy reminder.

[0069] Those skilled in the art will understand that the above embodiments are merely illustrative of the present invention and are not intended to limit the present invention. The scope of protection of the present invention is defined by the claims. Various modifications and variations may be made to the above embodiments without departing from the spirit and scope of the present invention. For example, the intelligent analysis module may adopt a more advanced AI model, the data source may be further expanded, and the reminder methods may be more diverse.

Claims

1. Financial transaction intelligent analysis system, characterized by: include: Data collection module, used to collect multi-dimensional data related to financial transactions; Data preprocessing and feature engineering module, used to process the collected data and extract effective features; An intelligent analysis and decision-making module, configured to analyze and generate financial transaction signals based on the effective features through an algorithmic model; User preference and strategy configuration module, where users set their investment preferences and trading strategy parameters; A signal matching and reminder generation module, used to match the transaction signal with the parameters set by the user and generate transaction reminder information that meets the conditions; The notification push module is used to push the transaction reminder information to the user.

2. The system according to claim 1, wherein: The data collected by the data collection module includes at least one or more combinations of market data, technical indicator data, fundamental data, news information and announcement data.

3. The system according to claim 2, characterized in that The data collected by the data collection module also includes market sentiment data.

4. The system according to claim 1, wherein: The algorithm model adopted by the intelligent analysis and decision-making module includes a rule-based expert system, a machine learning model, a deep learning model or a fusion model thereof.

5. The system according to claim 4, characterized in that The machine learning model is selected from one or more of a support vector machine, a decision tree, a random forest, a gradient boosting machine, or a long short-term memory network.

6. The system according to claim 1, wherein: The system also includes a user interaction and feedback module for receiving user feedback on transaction reminders and using the feedback information to optimize the algorithm model of the intelligent analysis and decision-making module.

7. The system according to claim 1, wherein: The notification push module pushes transaction reminder information via at least one of in-application push, SMS or email.

8. The system according to claim 1, wherein: The following steps are involved: (a) Collect multi-dimensional data related to financial transactions, and obtain real-time market data such as the opening price, closing price, highest price, lowest price, and trading volume of financial transactions; obtain news headlines, content, release time, and other information related to the financial transaction market through crawler technology or cooperation with news providers; (b) preprocessing the collected data and extracting effective features, cleaning and standardizing the market data, and calculating commonly used technical indicators, such as the moving average (MA) and relative strength index (RSI); performing word segmentation, stop word removal, and part-of-speech tagging on the news text to extract key information, and performing sentiment analysis on the news text using a sentiment analysis model (such as BERT, LLaMA, etc.) to extract sentiment polarity features; (c) loading the user’s set investment preferences and trading strategy parameters; (d) analyzing and generating financial transaction signals based on the effective features and algorithmic models; (e) matching the trading signals with the parameters set by the user and screening out trading signals that meet the conditions; (f) generating transaction reminder information based on the qualified transaction signals and pushing the reminder information to the user.

9. The system according to claim 8, characterized in that The algorithm model used in step (d) includes a rule-based expert system, a machine learning model, a deep learning model, or a fusion model thereof.

10. The system according to claim 8, wherein: Also includes the steps: (g) Collect user feedback on trading alerts and use this feedback and new data to iteratively optimize the algorithm model, displaying real-time financial trading market data in the form of charts, including price trend charts, technical indicator charts, etc. When the model generates a buy or sell signal, notify the user via one or more methods such as SMS, email, and APP push notifications.

Images