Cloud computing-based investor abnormal behavior detection method and system

Abstract

The invention discloses an investor abnormal behavior detection method and system based on cloud computing, and relates to the technical field of behavior detection, and the main scheme is that the method comprises the steps: collecting investment account data, investment frequency data, stock investment data and investment scale data of an investor in a current detection period; performing multi-round calculation on the collected data to obtain an investor comprehensive behavior robustness score, a comprehensive investment health score, an investment variety risk score and an investment scale comprehensive score, further calculating a comprehensive investment behavior abnormality score, presetting an investor abnormal behavior threshold value, comparing the comprehensive investment behavior abnormality score with the preset investor abnormal behavior threshold value, and determining the investor abnormal behavior according to the preset investor abnormal behavior threshold value. And judging whether the investor behavior is abnormal and generating a report.

Description

Technical Field

[0001] The present invention relates to the field of behavior detection technology, and in particular to a method and system for detecting abnormal behavior of investors based on cloud computing. Background Art

[0002] As the financial market continues to develop and grow, the number of investors continues to rise and trading activities become increasingly frequent and complex. In such an environment, accurately identifying abnormal investor behavior is crucial to maintaining the stability and fairness of the financial market. The rise of cloud computing technology provides an opportunity to solve these problems. It can integrate large-scale distributed computing resources to achieve efficient collection, storage and rapid processing of investor transaction data, market data and rich financial information data.

[0003] Traditional investor behavior detection methods rely on a single data resource and lack analytical depth, resulting in poor detection accuracy and insufficient ability to identify complex and changeable abnormal behavior patterns. Manual analysis accounts for a large proportion and is highly subjective, making it prone to omissions and misjudgments. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the deficiencies in the prior art, the present invention provides a method and system for detecting abnormal behavior of investors based on cloud computing. By collecting the investment account data, investment frequency data, stock investment data and investment scale data of the investor in the current detection period, multiple rounds of calculations are performed on the collected data to obtain the investor's comprehensive behavior robustness score, comprehensive investment health score, investment product risk score and investment scale comprehensive score, and further calculate the comprehensive investment behavior abnormality score, preset the investor abnormal behavior threshold, compare the comprehensive investment behavior abnormality score with it, judge whether the investor behavior is abnormal and generate a report, which solves the problems of relying on a single data resource, insufficient analysis depth, resulting in poor detection accuracy, insufficient ability to recognize complex and changeable abnormal behavior patterns, a large proportion of manual analysis, strong subjectivity, and easy omissions and misjudgments.

[0005] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for detecting abnormal behavior of investors based on cloud computing, comprising the following steps: Step 1: Collect the investor's investment account data, investment frequency data, stock investment data and investment scale data in the current detection period; Step 2: Calculate the investment reliability index RPE and the account funds stability index REW based on the investment frequency data and the investment account data; calculate the transaction-fund health index QWC based on the investment reliability index RPE and the account funds stability index REW; calculate the investor's comprehensive behavior robustness score IBR based on the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC; Calculate the investment rhythm stability index CVB and market adaptability index VHG based on the investment frequency data, and further calculate the comprehensive investment health score MAV; Calculate the investment product risk score RCS based on stock investment data; Calculate the investment fund liquidity ratio EMC and the holding market value volatility index EGD based on the investment scale data and further calculate the investment scale comprehensive score MEB; The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB; Step 3: Preset an investor abnormal behavior threshold GFN, compare the comprehensive investment behavior abnormality score CDQ with the investor abnormal behavior threshold GFN, determine whether the investor behavior is abnormal based on the comparison result, and generate a corresponding detection report.

[0006] In the preferred embodiment of the above-mentioned method for detecting abnormal behavior of investors based on cloud computing: the method for calculating the investment reliability index RPE and the account funds stability index REW is: Investment frequency data includes the number of investment transactions QAW i ; Investment account data includes actual account balance QAZ and cash flow VDW; According to the number of investment transactions QAW i , the actual account balance QAZ and the cash flow VDW are used to calculate the investment reliability index RPE, and the formula is as follows:

[0007] Among them, QAW i is the transaction number of the i-th stock investment transaction, i is the serial number corresponding to the number of different stock investment transactions, and its value is [1, n], and n is the number of stock investment transactions, and its value is a positive integer; The investment frequency data also include stock trading price EDA; Investment account data also includes the initial account fund balance JGP; The account fund stability index REW is calculated based on the stock transaction price EDA, the actual account balance QAZ and the initial account fund balance JGP. The formula is as follows: .

[0008] In the preferred embodiment of the above-mentioned method for detecting abnormal behavior of investors based on cloud computing: the method for calculating the transaction-fund health index QWC is: The trading-fund health index QWC is calculated based on the investment reliability index RPE and the account fund stability index REW, and the formula is as follows: .

[0009] In the preferred embodiment of the above-mentioned method for detecting abnormal behavior of investors based on cloud computing: the method for calculating the comprehensive behavior robustness score IBR of investors is: The investor's comprehensive behavioral robustness score IBR is calculated based on the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC. The formula is as follows: .

[0010] In the preferred embodiment of the above-mentioned method for detecting abnormal investor behavior based on cloud computing, the method for calculating the comprehensive investment health score MAV is: The investment frequency data also include the standard deviation of transaction time intervals JUT and the average transaction time interval AKH; The investment rhythm stability index CVB is calculated based on the standard deviation of transaction time intervals JUT and the average transaction interval time AKH. The formula is as follows:

[0011] Investment frequency data also includes stock trading frequency GMC j and market conditions duration FNW j ; According to the stock trading frequency GMC j and market conditions duration FNW j The market adaptability index VHG is calculated based on the following formula:

[0012] Among them, GMC j is the stock trading frequency under the j-th market situation, FNW j is the duration of the jth market situation, j is the serial number corresponding to different market situations, and its value is [1, h]; h is the number of market situation types, and its value is a positive integer; The comprehensive investment health score MAV is calculated based on the investment rhythm stability index CVB and the market adaptability index VHG. The formula is as follows: .

[0013] In the preferred embodiment of the above-mentioned method for detecting abnormal behavior of investors based on cloud computing, the method for calculating the risk score RCS of investment products is: Stock investment data including expected return BSI k and risk value LBF k ; Based on the expected rate of return BSI k and risk value LBF k The formula for calculating the investment product risk score RCS is as follows:

[0014] Among them, BSI k is the expected return of the stock corresponding to the k-th stock code, LBF k is the risk value of the stock corresponding to the kth stock code, k is the serial number corresponding to the stock code, and its value is [1, q], q is the total number of stock codes, and its value is a positive integer.

[0015] In the preferred embodiment of the above-mentioned method for detecting abnormal behavior of investors based on cloud computing: the method for calculating the comprehensive score MEB of investment scale is: The investment scale data include the initial capital inflow amount KGF, the initial capital outflow amount NEJ and the total investment amount PAH; The investment funds turnover ratio EMC is calculated based on the initial capital inflow amount KGF, the initial capital outflow amount NEJ and the total investment amount PAH, and the formula is as follows:

[0016] Among them, ω 1 is the weight coefficient of the initial capital inflow amount KGF, ranging from 0.1 to 0.4; ω 2 is the weight coefficient of the initial capital outflow amount NEJ, ranging from 0.6 to 0.9; and ω 1 +ω 2 =1; The investment scale data also includes the initial holdings value GDP and the ending holdings market value SAM; The holding value volatility index EGD is calculated based on the initial holding value GDP and the ending holding value SAM. The formula is as follows:

[0017] Among them, α 1 is the weight coefficient of the initial position value GDP, ranging from 0.2 to 0.4; α 2 The weight coefficient of the end-of-period holding market value SAM is 0.6~0.8; and α 1 +α2 =1; The comprehensive investment scale score MEB is calculated based on the investment fund liquidity ratio EMC and the holding market value volatility index EGD. The formula is as follows:

[0018] In the preferred embodiment of the above-mentioned method for detecting abnormal investor behavior based on cloud computing, the method for calculating the comprehensive investment behavior abnormality score CDQ is: The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB. The formula is as follows: .

[0019] In the preferred embodiment of the above-mentioned method for detecting abnormal investor behavior based on cloud computing: the method for determining whether the investor behavior is abnormal is: The comprehensive investment behavior abnormality score CDQ is compared with the investor abnormal behavior threshold GFN, and the investor behavior is judged to be abnormal based on the comparison result. The criteria are as follows: .

[0020] The present invention also discloses an investor abnormal behavior detection system based on cloud computing, comprising: The data collection module is used to collect the investment account data, investment frequency data, stock investment data and investment scale data of the investor in the current detection period; A scoring calculation module is used to calculate the investment reliability index RPE and the account funds stability index REW according to the investment frequency data and the investment account data; calculate the transaction-fund health index QWC according to the investment reliability index RPE and the account funds stability index REW; calculate the investor comprehensive behavior robustness score IBR according to the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC; Calculate the investment rhythm stability index CVB and market adaptability index VHG based on the investment frequency data, and further calculate the comprehensive investment health score MAV; Calculate the investment product risk score RCS based on stock investment data; Calculate the investment fund liquidity ratio EMC and the holding market value volatility index EGD based on the investment scale data and further calculate the investment scale comprehensive score MEB; The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB; The judgment module is used to preset an investor abnormal behavior threshold GFN, compare the comprehensive investment behavior abnormality score CDQ with the investor abnormal behavior threshold GFN, judge whether the investor behavior is abnormal based on the comparison result, and generate a corresponding detection report.

[0021] (III) Beneficial effects The present invention provides a method and system for detecting abnormal behavior of investors based on cloud computing, which has the following beneficial effects: (1) By collecting the investment account data, investment frequency data, stock investment data and investment scale data of the investors in the current detection period, comprehensive data collection is achieved, providing basic information for subsequent analysis, ensuring that the investment situation of investors can be understood from multiple dimensions and avoiding inaccurate analysis due to missing data.

[0022] (2) By calculating multi-dimensional indicators based on the collected data, including various indexes such as reliability, stability, health, risk, capital flow and volatility, and further calculating multiple comprehensive scores, it is possible to quantify investors' investment behavior from multiple angles, which can more accurately reflect the performance of investors in the investment process and help discover potential investment problems and risks.

[0023] (3) By presetting a threshold, the comprehensive investment behavior anomaly score is compared with the threshold to determine whether the investor behavior is abnormal and generate a report. This can detect abnormal situations in a timely manner and provide risk warnings for investors. It also provides an effective tool for financial institutions to supervise investor behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The figure is a schematic diagram of the working steps of the method for detecting abnormal behavior of investors based on cloud computing of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] See also Figure 1 The present invention provides a method for detecting abnormal behavior of investors based on cloud computing, comprising the following steps: Step 1: Collect the investor's investment account data, investment frequency data, stock investment data and investment scale data for the current detection period.

[0027] When using, combine the content of step 1: By collecting the investment account data, investment frequency data, stock investment data and investment scale data of the investors' current testing period, we have achieved comprehensive data collection, provided basic information for subsequent analysis, ensured that we can understand the investors' investment situation from multiple dimensions, and avoided inaccurate analysis due to missing data.

[0028] Step 2: Calculate the investment reliability index RPE and the account funds stability index REW based on the investment frequency data and investment account data; calculate the transaction-fund health index QWC based on the investment reliability index RPE and the account funds stability index REW; calculate the investor's comprehensive behavioral robustness score IBR based on the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC.

[0029] The investment rhythm stability index CVB and the market adaptability index VHG are calculated based on the investment frequency data, and the comprehensive investment health score MAV is further calculated.

[0030] Calculate the investment product risk score RCS based on stock investment data.

[0031] Based on the investment scale data, the investment cash flow ratio EMC and the holding market value volatility index EGD are calculated, and the comprehensive investment scale score MEB is further calculated.

[0032] The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB.

[0033] Step 201: The method for calculating the investment reliability index RPE and the account funds stability index REW is: Investment frequency data includes the number of investment transactions QAW i .

[0034] It should be noted that the number of investment transactions is QAW i Represents the transaction number of the i-th stock investment transaction, obtained from the investor's transaction records.

[0035] Investment account data includes actual account balance QAZ and cash flow VDW.

[0036] It should be noted that the actual account balance QAZ represents the investor's current actual account balance, which is obtained from the investor's securities account information and reflects the actual amount of funds in the investor's current account; the cash flow VDW represents the inflow and outflow of investors' funds. In stock trading, the cash flow is reflected in the number of shares and is obtained through the transaction records of financial institutions.

[0037] According to the number of investment transactions QAWi , the actual account balance QAZ and the cash flow VDW are used to calculate the investment reliability index RPE, and the formula is as follows:

[0038] Among them, QAW i is the transaction number of the i-th stock investment transaction, i is the serial number corresponding to different stock investment transaction numbers, and its value is [1, n], and n is the number of stock investment transactions, and its value is a positive integer.

[0039] It should be noted that in this formula: the numerator represents the sum of the number of all stock investment transactions, divided by the product of the actual account balance QAZ and the cash flow VDW and multiplied by 100 to obtain the investment reliability index RPE, and multiplied by 100 is a proportional coefficient. The ratio calculated previously is multiplied by 100 to enlarge the result to a numerical range that is easier to observe and compare. When the total number of transactions of investors is large and the denominator is relatively small, it means that investors have made more transactions on the basis of relatively less funds, which may imply that the investment behavior is more aggressive and the investment reliability may be lower; when investors have more capital reserves and larger capital flows, and the numerator is relatively small, it means that investors have made relatively fewer transactions with stronger financial strength, which may mean that the investment behavior is more conservative and the investment reliability is relatively high.

[0040] Investment frequency data also include stock trading price EDA.

[0041] It should be noted that the stock transaction price EDA represents the transaction price in the stock transaction, which is obtained from the stock transaction record.

[0042] Investment account data also includes the initial account fund balance JGP.

[0043] It should be noted that the initial account balance JGP represents the balance of funds in the investor's account at the time of initial investment, which is obtained from the investor's own investment records.

[0044] The account fund stability index REW is calculated based on the stock transaction price EDA, the actual account balance QAZ and the initial account fund balance JGP. The formula is as follows: .

[0045] It should be noted that in this formula: the actual account balance QAZ divided by the initial account balance JGP represents the proportional relationship between the actual account balance and the initial account balance, reflecting the relative changes in account funds. When the two are close, it means that the account funds change less and the funds stability is higher; when the stock trading price is higher, the denominator is larger and the entire fraction value is smaller, which means that a higher stock trading price may have a negative impact on the stability of account funds.

[0046] Step 202: The method for calculating the transaction-fund health index QWC is: The trading-fund health index QWC is calculated based on the investment reliability index RPE and the account fund stability index REW, and the formula is as follows: .

[0047] It should be noted that in this formula: the trading-fund health index QWC is obtained by multiplying the investment reliability index RPE and the account funds stability index REW and taking the square root. When the product of the two is large, the trading-fund health index will be higher, indicating that the investor is in a relatively healthy state in terms of trading and fund management; when the product of the two is small, the trading-fund health index will be lower, reflecting that the investor may face risks in trading and fund management.

[0048] Step 203: The method for calculating the investor comprehensive behavior robustness score IBR is: The investor's comprehensive behavioral robustness score IBR is calculated based on the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC. The formula is as follows: .

[0049] It should be noted that this formula takes into account three factors: investor reliability, account fund stability and transaction-fund health. The investor's comprehensive behavior robustness score IBR is obtained by multiplying the three indexes and taking the cube root. This avoids the problem of the IBR value deviating excessively from the actual situation due to one of the values ​​being too large. It can more reasonably reflect the robustness of the investor's comprehensive behavior. Generally speaking, the larger the IBR value, the more stable the investor's comprehensive behavior.

[0050] Step 204: The method for calculating the comprehensive investment health score MAV is: The investment frequency data also include the standard deviation of transaction time intervals JUT and the average transaction interval time AKH.

[0051] It should be noted that the standard deviation of transaction time interval JUT represents the time interval between two adjacent transactions, which is obtained by recording the time of each transaction and calculating the standard deviation of the time interval between adjacent transactions; the average transaction interval AKH represents the average time interval between each transaction of investors within a period of time, which is obtained by obtaining the transaction time from the transaction records and calculating the average value of all adjacent transaction time intervals.

[0052] The investment rhythm stability index CVB is calculated based on the standard deviation of transaction time intervals JUT and the average transaction interval time AKH. The formula is as follows:

[0053] It should be noted that this formula takes into account the standard deviation of the transaction time interval and the average transaction time interval. The result is converted into a percentage through the proportional relationship between the numerator and the denominator and multiplied by 100%, which makes it easier to intuitively compare and understand the stability of the investment rhythm. In general, the larger the CVB value, the more stable the investment rhythm.

[0054] Investment frequency data also includes stock trading frequency GMC j and market conditions duration FNW j .

[0055] It should be noted that the stock trading frequency GMC j It represents the stock trading frequency under the jth market situation, including bull market, bear market, shock market and other market situations. The trading frequency is obtained by counting the number of transactions under different market situations and then dividing it by the duration of the situation. The market situation duration FNW j Represents the duration of the j-th market situation, provided by a professional financial data provider.

[0056] According to the stock trading frequency GMC j and market conditions duration FNW j The market adaptability index VHG is calculated based on the following formula:

[0057] Among them, GMC j is the stock trading frequency under the j-th market situation, FNW j is the duration of the jth market situation, j is the serial number corresponding to different market situations, and its value is [1, h]; h is the number of market situation types, and its value is a positive integer.

[0058] It should be noted that in this formula: the numerator represents the sum of the products of trading frequency and market duration under different market conditions. By multiplying the trading frequency and the duration of the market, the trading activity of investors under different market conditions and the duration of the market are comprehensively considered; the denominator represents the sum of the duration of different market conditions. The calculation result is converted into a percentage by dividing the numerator and the denominator and multiplying it by 100%, which is convenient for intuitive comparison and understanding of the market adaptability index.

[0059] The comprehensive investment health score MAV is calculated based on the investment rhythm stability index CVB and the market adaptability index VHG. The formula is as follows: .

[0060] It should be noted that in this formula: the comprehensive investment health score MAV is obtained by multiplying the investment rhythm stability index CVB and the market adaptability index VHG and taking the square root. When both CVB and VHG are high, the MAV value will also be high, which indicates that the investor not only has a stable trading rhythm during the investment process, but also can adapt well to the market conditions, and the overall investment health is good; when CVB or VHG is low, the MAV value will also be low, indicating that the investor has an unstable trading rhythm or is not well adapted to the market conditions during the investment process, and the investment health is poor.

[0061] Step 205: The method for calculating the investment product risk score RCS is: Stock investment data including expected return BSI k and risk value LBF k .

[0062] It should be noted that the expected rate of return BSI k The expected rate of return of the stock corresponding to the k-th stock code is obtained through the research report released by the Asset Management Association of China; the risk value LBF k It represents the risk level of the stock corresponding to the kth stock code, and measures the risk value of the stock corresponding to each stock code by collecting historical volatility data.

[0063] Based on the expected rate of return BSI k and risk value LBF k The formula for calculating the investment product risk score RCS is as follows:

[0064] Among them, BSI k is the expected return of the stock corresponding to the k-th stock code, LBF kis the risk value of the stock corresponding to the kth stock code, k is the serial number corresponding to the stock code, and its value is [1, q], q is the total number of stock codes, and its value is a positive integer.

[0065] It should be noted that in this formula: the numerator represents the sum of the products of the expected rate of return and the inverse of the risk value for stocks corresponding to different stock codes. The product of the expected rate of return and the inverse of the risk value comprehensively considers the relationship between the return and risk of the investment. The smaller the risk value, the greater the contribution to the numerator, which means that under the same expected rate of return, the smaller the risk, the more advantageous it is. The denominator represents the ratio of the expected rate of return of different stock codes to the number of stock codes, comprehensively considering the expected rate of return and the risk value of stocks corresponding to different stock codes. Through this proportional relationship between the numerator and the denominator, a score RCS that can reflect the risk level of the investment product is calculated. The larger the RCS value, the lower the overall risk of the investment product may be when considering the expected rate of return and the risk value; the smaller the RCS value, the higher the risk of the investment product may be.

[0066] Step 206: The method for calculating the comprehensive score of investment scale MEB is: The investment scale data include the initial capital inflow amount KGF, the initial capital outflow amount NEJ and the total investment amount PAH.

[0067] It should be noted that the initial capital inflow amount KGF represents the amount of capital inflow from the time when the investor makes an investment decision and the capital begins to enter the investment market to the time when the initial asset allocation is completed, which is obtained from the initial capital record of the investment account; the initial capital outflow amount NEJ represents the amount of capital outflow from the time when the investor makes an investment decision and the capital begins to enter the investment market to the time when the initial asset allocation is completed, which is obtained by checking the capital expenditure in the initial investment stage; the total investment amount PAH represents the total amount of capital invested by the investor in the investment project, which is obtained from the total capital record of the investment account.

[0068] The investment funds turnover ratio EMC is calculated based on the initial capital inflow amount KGF, the initial capital outflow amount NEJ and the total investment amount PAH, and the formula is as follows:

[0069] Among them, ω 1 is the weight coefficient of the initial capital inflow amount KGF, which is determined according to the impact of the initial capital inflow amount KGF on the investment capital turnover ratio EMC, and the value is 0.1~0.4; ω 2is the weight coefficient of the initial capital outflow amount NEJ, which is determined according to the impact of the initial capital outflow amount NEJ on the investment capital turnover ratio EMC, and takes a value of 0.6~0.9, and ω1+ω2=1; for example, if the stability and continuity of capital inflow are more critical to the success of the investment project, then ω1 can be appropriately increased; if the control of outflow funds, such as controlling costs and reducing unnecessary expenditures, is more meaningful to the healthy operation of the project, then ω2 should be increased. For example, in a 3-year long-term investment project, the initial capital inflow in the first year is of great significance to the start-up, and it can be given a weight of 0.6, and the weight of subsequent additional inflow funds is set to 0.4; and for outflow funds, cost outflow control is more important near the end of the project, then the weight of outflow funds in the last year is given to 0.7, and the weight of the early year is set to 0.3.

[0070] It should be noted that in this formula: the numerator part What is calculated is the net capital inflow under the influence of the weight coefficient, that is, the difference between the initial inflow and the initial outflow under the influence of the weight coefficient. The denominator PAH is the total investment amount. By dividing the net capital inflow by the total investment amount, we can get the investment capital turnover ratio EMC, which can be used to measure the flow efficiency of funds in investment projects.

[0071] Investment scale data also includes initial holding value GDP and ending holding market value SAM.

[0072] It should be noted that the initial holding value GDP indicates the amount of assets held by investors from the time when investors make investment decisions and funds begin to enter the investment market to the completion of the initial asset allocation, which is obtained from the initial holding record of the investment account; the ending holding value SAM indicates the market value of assets held by investors at the end of the period from the completion of the initial asset allocation, through the entire investment process, to the final decision to end the investment and complete the asset liquidation, which is obtained from the ending asset valuation record of the investment account.

[0073] The holding value volatility index EGD is calculated based on the initial holding value GDP and the ending holding value SAM. The formula is as follows:

[0074] Among them, α 1 is the weight coefficient of the initial holding value GDP, which is determined according to the impact of the initial holding value GDP on the holding market value volatility index EGD, and the value is 0.2~0.4; α 2 The weight coefficient of the end-of-period holdings market value SAM is determined according to the impact of the end-of-period holdings market value SAM on the holdings market value volatility index EGD, and the value is 0.6~0.8; and α 1 +α 2 =1.

[0075] It should be noted that in this formula: the initial holding value and the end-of-period holding market value are comprehensively considered, and their impact on the holding market value volatility index is adjusted through weight coefficients and respectively. When the numerator is larger, when the denominator is 2, the EGD value is larger, indicating that the volatility of the holding market value is larger.

[0076] The comprehensive investment scale score MEB is calculated based on the investment fund liquidity ratio EMC and the holding market value volatility index EGD. The formula is as follows:

[0077] It should be noted that in this formula: the investment capital flow ratio and the holding market value volatility index are comprehensively considered. When EMC and EGD are both large and their values ​​are close, the numerator is larger, the denominator is smaller, and the MEB value is larger, which indicates that when the performance in investment capital flow and holding market value volatility is better, the comprehensive score of investment scale is higher; when one of EMC and EGD is smaller, the denominator is larger and the MEB value is smaller, indicating that the overall performance of investment scale is poor.

[0078] Step 207: The method for calculating the comprehensive investment behavior abnormality score CDQ is: The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB. The formula is as follows: .

[0079] It should be noted that in this formula: the comprehensive investment behavior abnormality score CDQ is calculated by adding the four scores and multiplying them as the denominator, with the numerator being 1. When the four scores of IBR, MAV, RCS, and MEB are all high, the value of the denominator is large, resulting in a small value of CDQ. This means that when the comprehensive behavior of investors is stable, the comprehensive investment is healthy, the investment product risk is low, and the comprehensive performance of the investment scale is good, the comprehensive investment behavior abnormality score is low, that is, the investment behavior is relatively normal; when one or more of the four scores is low, the value of the denominator is small, and the value of CDQ is large, indicating that there may be abnormalities in the investment behavior.

[0080] When used, combine the contents of step 201 to step 207: By calculating multi-dimensional indicators based on the collected data, including various indexes such as reliability, stability, health, risk, capital flow and volatility, and further comprehensively calculating multiple comprehensive scores, it is possible to quantify investors' investment behavior from multiple angles, which can more accurately reflect investors' performance in the investment process and help discover potential investment problems and risks.

[0081] Step 3: Preset an investor abnormal behavior threshold GFN, compare the comprehensive investment behavior abnormality score CDQ with the investor abnormal behavior threshold GFN, determine whether the investor behavior is abnormal based on the comparison result, and generate a corresponding detection report.

[0082] Step 301: Preset an investor abnormal behavior threshold GFN by: The investor abnormal behavior threshold GFN collects a large number of abnormal investment data from investors in different periods, calculates the comprehensive investment behavior abnormality score CDQ for different periods, screens out the lower 50% comprehensive investment behavior abnormality scores and calculates their average value as the investor abnormal behavior threshold GFN.

[0083] Step 302: The method for determining whether the investor's behavior is abnormal is: The comprehensive investment behavior abnormality score CDQ is compared with the investor abnormal behavior threshold GFN, and the investor behavior is judged to be abnormal based on the comparison result. The criteria are as follows:

[0084] It should be noted that a corresponding detection report is generated based on the results, and the report content may include: the investor's investment account data, investment frequency data, stock investment data and investment scale data during the current detection period, comprehensive investment behavior abnormality score, behavior judgment results and risk warnings.

[0085] When used, combine the contents of step 301 to step 302: By presetting the threshold, the comprehensive investment behavior anomaly score is compared with the threshold to determine whether the investor behavior is abnormal and generate a report. This can detect abnormal situations in a timely manner, provide risk warnings for investors, and also provide an effective tool for financial institutions to supervise investor behavior.

[0086] On the other hand, the present invention also discloses an investor abnormal behavior detection system based on cloud computing, comprising: The data collection module is used to collect the investment account data, investment frequency data, stock investment data and investment scale data of the investor in the current detection period; A scoring calculation module is used to calculate the investment reliability index RPE and the account funds stability index REW according to the investment frequency data and the investment account data; calculate the transaction-fund health index QWC according to the investment reliability index RPE and the account funds stability index REW; calculate the investor comprehensive behavior robustness score IBR according to the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC; Calculate the investment rhythm stability index CVB and market adaptability index VHG based on the investment frequency data, and further calculate the comprehensive investment health score MAV; Calculate the investment product risk score RCS based on stock investment data; Calculate the investment fund liquidity ratio EMC and the holding market value volatility index EGD based on the investment scale data and further calculate the investment scale comprehensive score MEB; The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB; The judgment module is used to preset an investor abnormal behavior threshold GFN, compare the comprehensive investment behavior abnormality score CDQ with the investor abnormal behavior threshold GFN, judge whether the investor behavior is abnormal based on the comparison result, and generate a corresponding detection report.

[0087] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware or in combination with computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0088] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0089] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. The method for detecting abnormal investor behavior based on cloud computing is characterized by: The following steps are involved: Step 1: Collect the investor's investment account data, investment frequency data, stock investment data and investment scale data in the current detection period; Step 2: Calculate the investment reliability index RPE and the account funds stability index REW based on the investment frequency data and the investment account data; calculate the transaction-fund health index QWC based on the investment reliability index RPE and the account funds stability index REW; calculate the investor's comprehensive behavior robustness score IBR based on the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC; Calculate the investment rhythm stability index CVB and market adaptability index VHG based on the investment frequency data, and further calculate the comprehensive investment health score MAV; Calculate the investment product risk score RCS based on stock investment data; Calculate the investment fund liquidity ratio EMC and the holding market value volatility index EGD based on the investment scale data, and further calculate the investment scale comprehensive score MEB; The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB; Step 3: Preset an investor abnormal behavior threshold GFN, compare the comprehensive investment behavior abnormality score CDQ with the investor abnormal behavior threshold GFN, determine whether the investor behavior is abnormal based on the comparison result, and generate a corresponding detection report.

2. The method for detecting abnormal investor behavior based on cloud computing according to claim 1 is characterized in that: The method for calculating the investment reliability index RPE and the account funds stability index REW is: Investment frequency data includes the number of investment transactions QAW i ; Investment account data includes actual account balance QAZ and cash flow VDW; According to the number of investment transactions QAW i , the actual account balance QAZ and the cash flow VDW are used to calculate the investment reliability index RPE, and the formula is as follows: Among them, QAW i is the transaction number of the i-th stock investment transaction, i is the serial number corresponding to the number of different stock investment transactions, and its value is [1, n], and n is the number of stock investment transactions, and its value is a positive integer; The investment frequency data also include stock trading price EDA; Investment account data also includes the initial account fund balance JGP; The account fund stability index REW is calculated based on the stock transaction price EDA, the actual account balance QAZ and the initial account fund balance JGP. The formula is as follows: 。 3. The method for detecting abnormal investor behavior based on cloud computing according to claim 2 is characterized in that: The method for calculating the transaction-fund health index QWC is: The trading-fund health index QWC is calculated based on the investment reliability index RPE and the account fund stability index REW, and the formula is as follows: 。 4. The method for detecting abnormal investor behavior based on cloud computing according to claim 3 is characterized in that: The method for calculating the investor's comprehensive behavioral robustness score IBR is: The investor comprehensive behavior robustness score IBR is calculated based on the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC. The formula is as follows: 。 5. The method for detecting abnormal investor behavior based on cloud computing according to claim 4 is characterized in that: The method for calculating the comprehensive investment health score MAV is: The investment frequency data also include the standard deviation of transaction time intervals JUT and the average transaction time interval AKH; The investment rhythm stability index CVB is calculated based on the standard deviation of transaction time intervals JUT and the average transaction interval time AKH. The formula is as follows: Investment frequency data also includes stock trading frequency GMC j and market conditions duration FNW j ; According to the stock trading frequency GMC j and market conditions duration FNW j The market adaptability index VHG is calculated based on the following formula: Among them, GMC j is the stock trading frequency under the j-th market situation, FNW j is the duration of the jth market situation, j is the serial number corresponding to different market situations, and its value is [1, h]; h is the number of market situation types, and its value is a positive integer; The comprehensive investment health score MAV is calculated based on the investment rhythm stability index CVB and the market adaptability index VHG. The formula is as follows: 。 6. The method for detecting abnormal investor behavior based on cloud computing according to claim 5 is characterized in that: The method for calculating the investment product risk score RCS is: Stock investment data including expected return BSI k and risk value LBF k ; Based on the expected rate of return BSI k and risk value LBF k The formula for calculating the investment product risk score RCS is as follows: Among them, BSI k is the expected return of the stock corresponding to the k-th stock code, LBF k is the risk value of the stock corresponding to the kth stock code, k is the serial number corresponding to the stock code, and its value is [1, q], q is the total number of stock codes, and its value is a positive integer.

7. The method for detecting abnormal investor behavior based on cloud computing according to claim 6 is characterized in that: The method for calculating the comprehensive score of investment scale MEB is: The investment scale data include the initial capital inflow amount KGF, the initial capital outflow amount NEJ and the total investment amount PAH; The investment funds turnover ratio EMC is calculated based on the initial capital inflow amount KGF, the initial capital outflow amount NEJ and the total investment amount PAH, and the formula is as follows: Among them, ω1 is the weight coefficient of the initial capital inflow amount KGF, which is 0.1~0.4; ω2 is the weight coefficient of the initial capital outflow amount NEJ, which is 0.6~0.9; and ω1+ω2=1; The investment scale data also includes the initial holdings value GDP and the ending holdings market value SAM; The holding value volatility index EGD is calculated based on the initial holding value GDP and the ending holding value SAM. The formula is as follows: Among them, α1 is the weight coefficient of the initial holding value GDP, which is 0.2~0.4; α2 is the weight coefficient of the end-of-period holding market value SAM, which is 0.6~0.8; and α1+α2=1; The comprehensive investment scale score MEB is calculated based on the investment fund liquidity ratio EMC and the holding market value volatility index EGD. The formula is as follows: 。 8. The method for detecting abnormal investor behavior based on cloud computing according to claim 7 is characterized in that: The method for calculating the comprehensive investment behavior abnormality score CDQ is: The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB. The formula is as follows: 。 9. The method for detecting abnormal investor behavior based on cloud computing according to claim 8 is characterized in that: The method to determine whether investor behavior is abnormal is: The comprehensive investment behavior abnormality score CDQ is compared with the investor abnormal behavior threshold GFN, and the investor behavior is judged to be abnormal based on the comparison result. The criteria are as follows: 。 10. The investor abnormal behavior detection system based on cloud computing is characterized by: include: The data collection module is used to collect the investment account data, investment frequency data, stock investment data and investment scale data of the investor in the current detection period; A scoring calculation module is used to calculate the investment reliability index RPE and the account funds stability index REW according to the investment frequency data and the investment account data; calculate the transaction-fund health index QWC according to the investment reliability index RPE and the account funds stability index REW; calculate the investor comprehensive behavior robustness score IBR according to the investment reliability index RPE, the account funds stability index REW and the transaction-fund health index QWC; Calculate the investment rhythm stability index CVB and market adaptability index VHG based on the investment frequency data, and further calculate the comprehensive investment health score MAV; Calculate the investment product risk score RCS based on stock investment data; Calculate the investment fund liquidity ratio EMC and the holding market value volatility index EGD based on the investment scale data and further calculate the investment scale comprehensive score MEB; The comprehensive investment behavior abnormality score CDQ is calculated based on the investor's comprehensive behavior robustness score IBR, the comprehensive investment health score MAV, the investment product risk score RCS and the investment scale comprehensive score MEB; The judgment module is used to preset an investor abnormal behavior threshold GFN, compare the comprehensive investment behavior abnormality score CDQ with the investor abnormal behavior threshold GFN, judge whether the investor behavior is abnormal based on the comparison result, and generate a corresponding detection report.

Images