Intelligent redisk method and system based on natural language comment technology

Abstract

The invention relates to an intelligent redisk method and system based on a natural language comment technology, and belongs to the field of financial science and technology services, and the method comprises the following steps: obtaining real-time investment data, dynamically visualizing an investment process, constructing a natural language generation system, redisk and comment. According to the invention, quantitative analysis is carried out on the delivery bill data of the user, scoring is carried out on the big stock judgment, individual stock selection, plate judgment and buying and selling point capabilities of the user, the comment description of a natural language is generated, and the investment process and result of the user are displayed in a dynamic visualization mode, so that the user experience is improved. The objective, comprehensive and intelligent evaluation of the stock investment management capability of the user is realized, and the investment level and efficiency of the user are improved.

Description

Technical Field

[0001] The present disclosure belongs to the field of financial technology services, and specifically relates to an intelligent review method and system based on natural language review technology. Background Art

[0002] Traditional financial investment review methods rely on investors to manually organize transaction records, analyze market conditions and personal decisions, which has problems such as high labor costs, low efficiency, and strong subjectivity. Although there are some auxiliary tools on the market, they usually lack the features of intelligence and natural language interaction.

[0003] Natural language processing technology has made significant progress in text analysis, semantic understanding and automatic generation, which can help computers understand and process human language. At the same time, natural language generation technology enables computers to generate natural and fluent text.

[0004] The financial investment knowledge base contains the laws of the stock market, investment strategies, financial analysis methods, etc. The construction of the financial investment knowledge base helps the system better understand and analyze investment behavior and improve the level of intelligence; dynamic visualization technology can present data in a dynamic and intuitive way to help users better understand data and trends. In the financial field, dynamic visualization can be used to display information such as transaction processes and stock trends, so that investors can understand their investment situation more intuitively; the delivery note is an important record of investors' stock transactions, which contains information on stocks bought and sold, transaction time, price and other data. Data processing and analysis of the delivery note can help investors understand their own trading behavior and investment returns.

[0005] By integrating the above technologies and methods, the present invention proposes an intelligent review method and system based on natural language processing technology. In view of the problems existing in the existing financial investment review methods, an intelligent review system is constructed using natural language processing technology and financial investment knowledge base, which realizes dynamic visualization and intelligent evaluation of investors' investment process. Summary of the invention

[0006] In view of the deficiencies in the prior art, the purpose of the present invention is to provide an intelligent review method and system based on natural language commentary technology, which solves the problem of investors' difficulty in reviewing in the prior art.

[0007] An intelligent review method based on natural language review technology includes the following steps:

[0008] Step 1, obtain real-time investment data;

[0009] Step 2: Dynamically visualize the investment process and determine the investor's holdings, individual stocks, returns, and market data based on the real-time investment data obtained;

[0010] Step 3: Build a natural language generation system. A natural language generation system is established with the delivery note as input and natural language comments as output. The delivery note is quantified and calculated to learn and identify typical operation behaviors, and generate corresponding descriptive language outputs.

[0011] Step 4: Review and comment. Review and comment on the investment process according to the time window. Match the dynamic visualization process with the time window of natural language commentary. During the review process, make real-time comments on typical operating behaviors.

[0012] A stock investment management capability evaluation system based on an intelligent system, comprising:

[0013] Input module, used to receive delivery notes and market data input by users;

[0014] The analysis module is used to conduct quantitative analysis on the delivery order, record the investment process of the entire delivery order, the income of each transaction, and compare the stock situation of the same industry in the same period according to the user's holdings;

[0015] The visualization module is used to dynamically visualize the delivery note through dynamic visualization technology, and at the same time, provide natural language comments on the typical operations of the user during the operation process;

[0016] The evaluation module is used to evaluate the user's stock investment management capabilities based on the user's delivery notes and comments.

[0017] The purpose of this disclosure can be achieved through the following technical solutions:

[0018] Beneficial effects of the present disclosure:

[0019] The present invention provides an intelligent review method based on natural language commentary technology, which can objectively, comprehensively and intelligently evaluate the user's stock investment management ability, including market judgment, individual stock selection, sector judgment and buying and selling point ability, as well as comprehensive ability;

[0020] By conducting quantitative analysis on the user's delivery note data, scoring the user's various capabilities and generating natural language review descriptions, and displaying the user's investment process and results in a dynamic and visual way, the user can clearly understand his or her investment level and efficiency, identify his or her strengths and weaknesses, and improve his or her investment capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 It is a flowchart of an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0024] Examples, such as Figure 1 As shown, an intelligent review method based on natural language review technology includes the following steps:

[0025] Step 1, get real-time investment data:

[0026] The investment process is input in the form of a delivery note, and the latest market data is obtained at the same time, the market data includes;

[0027] Stock historical data: This includes the stock's trading status over a period of time, such as opening price, closing price, highest price, lowest price, etc.

[0028] Real-time trading data: This refers to the latest trading status of stocks in the current market, including the latest transaction price, transaction volume, buy and sell order status, etc.

[0029] Basic stock information: This information describes the basic attributes of the stock itself, such as market value, price-earnings ratio, industry classification, concept classification, etc.

[0030] Financial statement data: This includes financial statement information of listed companies, such as balance sheet, income statement, cash flow statement, etc.

[0031] Step 2: Dynamic visualization of the investment process:

[0032] By inputting the delivery note and using visualization technology and K-line technology, the delivery note is processed into a visualized dynamic change process, and quantitative indicators such as positions, individual stocks, returns, and market data are calculated in real time.

[0033] Furthermore, the market data carries label data, which is used to indicate data of multiple dimensions. When comprehensive analysis of stocks is required, the market data of all stocks selected by the user and the band formula for analyzing the market data are first obtained from the market system.

[0034] In some embodiments, the output of the delivery note includes a description of the rise or fall or rate of return. The rise and fall of the held stocks are divided into severe losses, losses, small losses, small profits, profits, and large profits according to the threshold, and the income of the overall investment account is calculated at the same time. If the individual stocks held suffer severe losses and have a significant impact on the income of the overall account, the natural language description of the loss and degree of loss of the individual stocks will be retrieved, as well as the description of the impact of the individual stocks on the overall investment account. These description languages ​​are stored in a corpus of natural language descriptions and classified according to categories. Different categories have multiple language description templates, such as descriptions of the market, descriptions of losses, and descriptions of investment styles. When the calculation result reaches the threshold range of a certain category description, the description template in the category is called for description. The diagnosis module is called to give investment advice to the delivery note. The diagnosis module finds out the most typical behavior and reasons of the user according to the threshold range of the calculation result. For example, if a certain score of the rate of return, liquidity, risk control ability, etc. is relatively low, optimization suggestions are made for the indicator, and praise is made for the indicator with the relatively highest score, so that users can play to their strengths and avoid weaknesses and understand their own investment characteristics.

[0035] Furthermore, the specific steps for quantifying the benefits of the investment process are as follows:

[0036] Investment rate of return is one of the important indicators for evaluating investment performance in the investment field. This disclosure adopts a normalized calculation method that uses a unified rate of return during the investment period. The more commonly used method is the irregular cash flow internal rate of return. The calculation formula is:

[0037]

[0038] In the formula, Y is the rate of return, C0 is the market value at the beginning of the period, Ci is the funds flowing in or out at time ti, and Cn is the market value at the end of the period.

[0039] The maximum retracement rate of a stock refers to the maximum decline in the stock price within a period of time, reflecting the risk level of the stock. The technical formula for the maximum retracement rate of a stock is as follows:

[0040] R=Pmax / Pmin-Pmax

[0041] Among them, R represents the maximum retracement rate, Pmin represents the lowest point of the stock price, and Pmax represents the highest point of the stock price.

[0042] Step 3: Build a natural language generation system:

[0043] Establish a natural language generation system with delivery notes as input and natural language comments as output. Through quantitative calculation of delivery notes, learn and identify typical operation behaviors, and generate corresponding descriptive language output.

[0044] The user's delivery note needs to calculate some quantitative indicators of the user, including profitability, risk control ability, liquidity and performance attribution indicators, as well as the ability to choose industries, stocks, timing and positions, etc.

[0045] In the embodiment of the present disclosure, the establishment of a database for constructing a natural language generation system includes the following steps:

[0046] 3.1 Data preprocessing;

[0047] Obtain large-scale financial investment-related corpus, perform pre-processing such as data cleaning and deduplication, and store it in the database.

[0048] The data mainly includes stock market data and user transaction data. Market data mainly includes historical data of stocks, real-time transaction data and basic information of stocks. Historical data and transaction data of stocks mainly include high opening and low closing data. Basic information of stocks includes market value, price-earnings ratio, industry, concept, etc.

[0049] User delivery data mainly includes relevant transaction data such as buying, selling, price, quantity, transaction amount, etc. reflected in the delivery note.

[0050] In some embodiments, the data preprocessing uses a clustering algorithm to divide the data into several clusters, and then determines whether the data belongs to a certain cluster. If some data is sparsely distributed and does not belong to any cluster, it is considered an outlier. This method can effectively identify abnormal fluctuations, abnormal prices, abnormal trading volumes, etc. in stock data, and process them accordingly, such as deleting, replacing or retaining. The specific steps are as follows:

[0051] Choose a suitable clustering algorithm and select appropriate parameters according to the characteristics and goals of the data, such as the number of clusters, distance measurement, density threshold, etc.

[0052] Cluster the stock data to get several clusters and the label of the cluster to which each data point belongs. If a data point is not assigned to any cluster, it is marked as an outlier.

[0053] Process outliers and choose appropriate processing methods according to specific situations, such as deletion, replacement, or retention. If you delete outliers, remove them directly from the data set; if you replace outliers, replace them with other data, such as the mean, median, boundary value, etc.; if you retain outliers, do not do any processing, but pay attention to analyzing their possible causes and effects.

[0054] Analyze the processed data to evaluate the effects of clustering and outlier processing, such as using the silhouette coefficient, adjusted Rand index and other indicators to measure the quality of clustering, and using the mean absolute error, root mean square error and other indicators to measure the error of outlier processing.

[0055] 3.2 Build a knowledge base in the field of financial investment;

[0056] Construct financial investment-related data into knowledge and store it in the ontology knowledge base. At the same time, optimize the knowledge base structure to facilitate continuous updating and upgrading. Establish a natural language processing model.

[0057] The ontology knowledge base in the financial investment field is built by taking stocks as the center and using the relevant information of stocks as attributes to establish the ontology. Then, the ontologies are linked through the relationships between stocks, such as the same industry and the same concept, to form a knowledge base with stocks as the ontology unit. When new stocks or new concepts appear, they are learned and updated to the knowledge base. For example, after a new stock is issued, a new stock ontology unit is added to obtain and add the basic information, industry, and concept information of the new stock. The data in the database is mainly called with the stock code as the key information, and the relevant data of the ontology unit is called.

[0058] In some further embodiments, building a database of the natural language generation system may be performed before step 1, obtaining real-time investment data.

[0059] In the disclosed embodiments, typical operation behaviors include changes in stock positions, changes in individual stock returns, the rise or fall of the industry to which the held stocks belong or stocks in the same industry, etc. The magnitude of these changes is defined, and when the change reaches a defined threshold, the operation is described in natural language. The threshold can be a numerical value of the increase or decrease, or a score value.

[0060] Step 4: Review and comment:

[0061] The investment process is reviewed and commented on according to the time window, and the dynamic visualization process is matched with the time window of natural language commentary. During the review process, real-time commentary is performed on typical operating behaviors, and an intelligent review system based on natural language commentary technology is built.

[0062] It should be noted here that the intelligent replay system uses visualization technology to play the user's delivery order on the web page, and records the user's position and individual stock returns at a certain point in time according to the time window. If the current quantitative indicators, mainly the gains and losses, reach the threshold that requires commentary description, the natural language comment module will be triggered to describe the current status. Because it takes time for users to see and understand the language of the comments during the playback process, and not all operations require comments during the entire operation process, we set the threshold of the quantitative indicators, and only when typical investment behaviors and calculation results appear will appropriate natural language comments be provided. An intelligent replay system is built through visualization technology and natural language commentary technology.

[0063] In some implementations, generating a comprehensive operational evaluation of the output results includes the following steps:

[0064] Calculate the user's quantitative scores in four aspects: market judgment, individual stock selection, sector judgment, and buying and selling point ability, which are recorded as S1, S2, S3, and S4 respectively.

[0065] Then, the four scores are normalized so that they are all in the interval [0,1];

[0066] Furthermore, the present disclosure takes the maximum and minimum normalization as an example, and the normalized scores are recorded as S1', S2', S3', and S4', and the calculation formula is as follows:

[0067]

[0068] Among them, Smin and Smax represent the minimum and maximum values ​​of the four scores respectively.

[0069] The normalized scores are weighted averaged to obtain the comprehensive ability score, denoted as S, and the calculation formula is as follows:

[0070]

[0071] Among them, wi represents the weight of the i-th aspect. Different weights can be set according to different investment objectives and risk preferences.

[0072] Map the comprehensive ability score to the interval [0,100] to obtain the quantitative score of the comprehensive ability, denoted as S c , the calculation formula is as follows:

[0073] S c =S×100

[0074] This formula indicates that the quantitative score of comprehensive ability is proportional to the weighted average of the scores in the four aspects. The higher the score, the stronger the comprehensive ability, and the lower the score, the weaker the comprehensive ability.

[0075] In this embodiment, the scoring rules and formula for market judgment ability: Market judgment ability refers to the ability of investors to judge and predict the overall trend and changing trend of the stock market, reflecting the investor's sensitivity and grasp of the market. The scoring standard for market judgment ability is based on the difference between the user's investment return rate and the rise and fall of the market index, and then considers factors such as the user's investment risk and holding time, and comprehensively calculates a score from 0 to 100 points. The higher the score, the stronger the market judgment ability, and the lower the score, the weaker the market judgment ability. The scoring formula for market judgment ability is as follows:

[0076]

[0077] Among them, S 1 It represents the score of the ability to judge the market, Y represents the user's investment return rate, B represents the rise and fall of the market index, R represents the user's investment risk, Rmax represents the maximum risk during the user's investment period, T represents the user's holding time, and Tmax represents the maximum holding time during the user's investment period.

[0078] It should be noted that the basis of the scoring formula is the difference between the user's investment return rate and the rise and fall of the market index. The larger the difference, the stronger the user's ability to judge the market, and the smaller the difference, the weaker the user's ability to judge the market. The difference is multiplied by 10 to amplify the impact of the difference and make the score more sensitive.

[0079] Scoring rules for stock selection ability:

[0080] Stock selection ability refers to the ability of investors to select high-quality stocks in the stock market, reflecting the ability of investors to identify and analyze the value and potential of individual stocks. The scoring standard for stock selection ability is based on the difference between the user's investment return rate and the rise and fall of individual stocks, and then considers factors such as the user's investment risk and holding time, and comprehensively calculates a score from 0 to 100. The higher the score, the stronger the stock selection ability, and the lower the score, the weaker the stock selection ability. The scoring formula for stock selection ability is as follows:

[0081] S 2 =50+10(GB) / P-10(Rmax / R)-10(Tmax / T)

[0082] Among them, S 2 It represents the score of individual stock selection ability, G represents the rise or fall of individual stocks, B represents the rise or fall of the market, R represents the investment risk of the user, Rmax represents the maximum risk during the user's investment period, T represents the user's holding time, and Tmax represents the maximum holding time during the user's investment period.

[0083] Scoring rules and formula for sector judgment ability:

[0084] S 3 =50+10(FB) / P-10(Rmax / R)-10(Tmax / T)

[0085] Among them, S 3 It represents the score of sector judgment ability, F represents the rise and fall of the sector, B represents the rise and fall of the market, R represents the investment risk of the user, Rmax represents the maximum risk during the user's investment period, T represents the user's holding time, and Tmax represents the maximum holding time during the user's investment period.

[0086] Scoring rules and formulas for buy / sell point capabilities:

[0087] S 4 =50+10(YP) / P-10(Rmax / R)-10(Tmax / T)

[0088] Among them, S 4 It represents the score of buying and selling point ability, Y represents the user's investment return rate, P represents the rise and fall of individual stocks, R represents the user's investment risk, Rmax represents the maximum risk during the user's investment period, T represents the user's holding time, and Tmax represents the maximum holding time during the user's investment period.

[0089] Buy-sell point ability refers to the ability of investors to determine the timing of buying and selling individual stocks in the stock market, reflecting the investor's ability to grasp and control the fluctuations and trends of individual stocks. The scoring criteria for buy-sell point ability are based on the difference between the user's investment return rate and the maximum increase and minimum decrease of individual stocks, and then consider factors such as the user's investment risk and holding time, and comprehensively calculate a score from 0 to 100.

[0090] The technical solution of the present invention is an intelligent review method based on natural language commentary technology, which aims to help investors improve the efficiency and returns of stock investment while reducing investment risks and costs. The technical solution of the present invention includes the following steps:

[0091] Furthermore, the steps of review and comment are as follows:

[0092] The first step is to input the investor's delivery note into the intelligent system, which will conduct quantitative analysis on the delivery note, record the entire investment process of the delivery note, the profit of each transaction, and compare the stock situation of the same industry in the same period based on the user's holdings.

[0093] The second step is to use dynamic visualization technology to dynamically visualize the delivery note and provide natural language comments on the typical operations of the user during the operation.

[0094] The third step is to evaluate the user's stock investment management ability based on the user's delivery note and comments. The evaluation criteria are as follows:

[0095] Stock investment management capabilities are divided into four aspects: market judgment, stock selection, sector judgment, and buying and selling points. Each aspect of ability has a score ranging from 0 to 100 points. The higher the score, the stronger the ability.

[0096] The score of each aspect of capability is composed of multiple indicators, each indicator has a weight, and the sum of the weights is 1. The score of each indicator is calculated based on the data in the user's delivery note and comments, and the calculation method depends on different indicators.

[0097] In the fourth step, based on the user's score on stock investment management ability, the intelligent system provides suggestions for improvement in timing or stock selection. For example, since the stock held by AVIC Optronics has suffered heavy losses and the personal market value has continued to decline, it is recommended to adjust the investment strategy in a timely manner to improve stock selection ability.

[0098] Furthermore, a rise or fall of 5% and a score below 50 will trigger comments and alert investors.

[0099] For example, first, the user's delivery note is quantitatively analyzed, the entire delivery note investment process, the income of each transaction are recorded, and the stock situation of the same industry in the same period is compared according to the user's holdings. The delivery note is dynamically visualized through dynamic visualization technology, and at the same time, the typical operations of the user during the operation are commented on in natural language, such as the Ping An Bank holding rose sharply, the increase reached 7%, the New China Life Insurance holding fell more than other stocks in the insurance industry, the high-selling and low-buying operation of Oriental Fortune was successfully carried out, and the Shanghai Composite Index fell today. By continuously training and upgrading the intelligence of the system, investors can be given suggestions for improvement in timing or stock selection. For example, due to the serious loss of the stock held by AVIC Optronics and the continuous decline in personal market value, it is recommended to adjust the investment strategy in time and improve the ability to select stocks.

[0100] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0101] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. An intelligent review method based on natural language review technology, characterized in that: The following steps are involved: Step 1, obtain real-time investment data; Step 2: Dynamically visualize the investment process and determine the investor's holdings, individual stocks, returns, and market data based on the real-time investment data obtained; Step 3: Build a natural language generation system. A natural language generation system is established with the delivery note as input and natural language comments as output. The delivery note is quantified and calculated to learn and identify typical operation behaviors, and generate corresponding descriptive language outputs. Step 4: Review and comment. Review and comment on the investment process according to the time window. Match the dynamic visualization process with the time window of natural language commentary. During the review process, make real-time comments on typical operating behaviors.

2. According to claim 1, the intelligent review method based on natural language review technology is characterized in that: The comments include stock investment management capabilities, which are divided into four aspects: market judgment ability, individual stock selection ability, sector judgment ability and buying and selling point ability. Each aspect of ability has a score ranging from 0 to 100 points. The higher the score, the stronger the ability.

3. According to claim 1, the intelligent review method based on natural language review technology is characterized in that: The establishment of a database for building a natural language generation system includes the following steps: Obtain large-scale financial investment-related corpus, perform pre-processing such as data cleaning and deduplication, and store them in the database; The data mainly includes stock market data and user transaction data; Market data mainly includes historical data of stocks, real-time transaction data and basic information of stocks; Stock historical data and transaction data mainly include high opening and low closing data. Stock basic information includes market value, price-earnings ratio, industry, concept, and sector. User delivery data mainly includes the buy, sell, price, quantity and transaction amount reflected in the delivery note.

4. According to claim 3, the intelligent review method based on natural language review technology is characterized in that: The data preprocessing steps include; Select a clustering algorithm and choose appropriate parameters based on the characteristics and goals of the stock data; Cluster the stock data to obtain several clusters and the label of the cluster to which each data point belongs. If a data point is not assigned to any cluster, it is marked as an outlier. Process outliers and select appropriate processing methods according to specific circumstances, such as deletion, replacement, or retention; The processed data are analyzed to evaluate the effect of clustering and outlier processing.

5. According to claim 4, the intelligent review method based on natural language review technology is characterized in that: The method includes constructing a knowledge base in the field of financial investment, and the specific construction steps are as follows; Step 1, obtaining real-time investment data, which includes changes in stock positions, changes in individual stock returns, and the rise and fall of the industry to which the stock holdings belong or the same industry; Step 2: Based on the investment data, call the financial investment domain ontology knowledge base, wherein the financial investment domain ontology knowledge base is established by taking stocks as the center and using the relevant information of stocks as attributes, and then the ontologies are connected through the relationship between stocks to form a knowledge base with stocks as the ontology unit; Step 3, generating a natural language description based on the investment data and the financial investment domain ontology knowledge base, wherein the natural language description defines the magnitude of the change in the investment data, and when the change reaches a defined threshold, a natural language description of the operation is performed.

6. According to claim 1, the intelligent review method based on natural language review technology is characterized in that: The specific steps of step 4 are as follows: Obtain the user's delivery order data, including the stock code, buying and selling price, quantity, time, and transaction amount of the transaction; Generate a dynamic and visualized investment process based on the delivery note data, showing the user's position and individual stock returns; Based on the preset thresholds of quantitative indicators, determine whether typical operation behaviors occur during the investment process; Based on typical operating behaviors, natural language commentary descriptions are generated, corresponding to the dynamic and visualized investment process, and played to users in real time.

7. According to claim 1, the intelligent review method based on natural language review technology is characterized in that: The steps to generate a comprehensive operational evaluation include; Step 1: Calculate the user's quantitative scores in four aspects: market judgment, stock selection, sector judgment, and buying and selling point ability, which are recorded as S1, S2, S3, and S4 respectively; Step 2: normalize the four scores so that they are all in the interval [0,1]. The normalized scores are recorded as S1', S2', S3', and S4'; Step 3: Take a weighted average of the normalized scores to obtain a comprehensive ability score, denoted as S; Step 4: Map the comprehensive ability score to the interval [0,100] to obtain the quantitative score of the comprehensive ability, denoted as S c .

8. According to claim 5, the intelligent review method based on natural language review technology is characterized in that: The threshold is set at a rise or fall of more than 5% or a score below 50.

9. An intelligent review system based on natural language review technology, characterized in that: include: Input module, used to receive delivery notes and market data input by users; The analysis module is used to conduct quantitative analysis on the delivery order, record the investment process of the entire delivery order, the income of each transaction, and compare the stock situation of the same industry in the same period according to the user's holdings; The visualization module is used to dynamically visualize the delivery note through dynamic visualization technology, and at the same time, provide natural language comments on the typical operations of the user during the operation process; The evaluation module is used to evaluate the user's stock investment management capabilities based on the user's delivery notes and comments.

Images