CN117726444A - Investment decision processing method and device, storage medium and electronic equipment - Google Patents
Investment decision processing method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN117726444A CN117726444A CN202311737559.8A CN202311737559A CN117726444A CN 117726444 A CN117726444 A CN 117726444A CN 202311737559 A CN202311737559 A CN 202311737559A CN 117726444 A CN117726444 A CN 117726444A
- Authority
- CN
- China
- Prior art keywords
- investment
- accounts
- decision
- demand data
- investment decision
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 49
- 238000003066 decision tree Methods 0.000 claims abstract description 44
- 238000007477 logistic regression Methods 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 15
- 230000006870 function Effects 0.000 claims description 18
- 238000001514 detection method Methods 0.000 claims description 15
- 230000002159 abnormal effect Effects 0.000 claims description 12
- 230000035772 mutation Effects 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 8
- 238000005516 engineering process Methods 0.000 abstract description 11
- 230000008569 process Effects 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 7
- 238000010276 construction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
Abstract
The invention discloses an investment decision processing method, an investment decision processing device, a storage medium and electronic equipment. Relates to the field of financial science and technology, and the method comprises the following steps: acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively; based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively; based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain investment probabilities corresponding to the N accounts respectively; and obtaining second investment decision results corresponding to the N accounts based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively. The invention solves the technical problem of low accuracy of decision prediction results in the investment decision method in the related technology.
Description
Technical Field
The invention relates to the field of financial science and technology, in particular to an investment decision processing method, an investment decision processing device, a storage medium and electronic equipment.
Background
The quantized investment is an investment method for asset management by means of a quantized financial analysis method, and not only can the effect of investment be analyzed and verified using history data, but also can be selected at the execution stage of investment. The series of processes are automatically executed by a computer. In view of the problem that whether the product is recommended to the user or not is one of the most basic problems in quantitative investment, namely, the problem of recommendation and non-recommendation, in the related art, whether the user performs prediction of investment is mainly performed in batches through a single quantitative investment model, so as to determine whether to recommend the corresponding investment product to the client or not. However, such methods have incomplete consideration, resulting in lower accuracy of investment decision prediction for the user.
In view of the above problems, no effective solution has been proposed at present.
Disclosure of Invention
The embodiment of the invention provides an investment decision processing method, an investment decision processing device, a storage medium and electronic equipment, which at least solve the technical problem of low decision prediction result accuracy in the investment decision processing method in the related technology.
According to an aspect of an embodiment of the present invention, there is provided an investment decision processing method including: acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively, and M, N is an integer greater than or equal to 2; based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively; based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain investment probabilities corresponding to the N accounts respectively; and obtaining second investment decision results corresponding to the N accounts respectively based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not.
According to another aspect of the embodiment of the present invention, there is also provided an investment decision processing apparatus including: the first acquisition module is used for acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively, and M, N is an integer greater than or equal to 2; the first prediction module is used for obtaining first investment decision results corresponding to the N accounts respectively by adopting a decision tree model based on the investment demand data corresponding to the N accounts respectively; the second prediction module is used for obtaining investment probabilities corresponding to the N accounts respectively by adopting a logistic regression model based on the investment demand data corresponding to the N accounts respectively; the decision module is used for obtaining second investment decision results corresponding to the N accounts respectively based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not.
According to another aspect of embodiments of the present invention, there is also provided a non-volatile storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform any one of the investment decision processing methods.
According to another aspect of the embodiments of the present invention, there is also provided an electronic device including one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement any one of the investment decision processing methods.
In the embodiment of the invention, the investment demand data respectively corresponding to N accounts is obtained, wherein the investment demand data comprises index values respectively corresponding to M investment indexes, and M, N is an integer greater than or equal to 2; based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively; based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain investment probabilities corresponding to the N accounts respectively; based on the first investment decision results respectively corresponding to the N accounts and the investment probabilities respectively corresponding to the N accounts, a second investment decision result respectively corresponding to the N accounts is obtained, wherein the second investment decision result is used for indicating whether the corresponding accounts are invested or not, the purposes of respectively predicting the first investment decision results and the investment probabilities of the plurality of accounts in batches based on the decision tree model and the logistic regression model and accurately determining the investment probabilities of the users in batches based on the first investment decision results and the investment probabilities are achieved, and therefore the technical effects of improving the batch investment decision prediction efficiency and the prediction result accuracy are achieved, and the technical problem that the decision prediction result accuracy of the investment decision method in the related technology is low is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
FIG. 1 is a flow chart of an investment decision processing method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an alternative decision tree according to an embodiment of the invention;
fig. 3 is a schematic diagram of an investment decision processing apparatus in accordance with an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The quantized investment is an investment method for asset management by means of a quantized financial analysis method, and not only can the effect of investment be analyzed and verified using history data, but also can be selected at the execution stage of investment. The series of processes are automatically executed by a computer. In view of the fact that quantitative investments are based on the background knowledge of mathematics, statistics, information technology to manage portfolios, they naturally play an important role in intelligent consultation. After quantitative investors gather and analyze a large amount of data, advanced mathematical models are adopted to replace artificial subjective judgment by means of powerful information processing capacity of a computer system, and investment opportunities are captured in the market by means of computer programs and put into practice. After the quantitative investment model is introduced, the negative influence of investors on investment due to emotion fluctuation can be well reduced. For example, in the case of extremely enthusiastic or pessimistic markets, the quantified results of coldness can well avoid investors from making irrational investment decisions for personal emotions and the like, so as to ensure that the benefit maximization is achieved under the premise of controlling risk. In one sentence, the quantitative investment is to program the investment strategy to better utilize massive information. The quantized investment may assist investors in optimizing the investment. The most familiar method in machine learning is also the most common method if the theory related to the quantized investment needs to be combed and the classification technology is said to be the most familiar method in machine learning. In view of the problem that whether the product is recommended to the user is one of the most basic problems in quantitative investment, namely, the problem of recommendation and non-recommendation, prediction for whether to perform investment is mainly performed through a single quantitative investment model in the related art, so as to determine whether to recommend the corresponding investment product to the client. However, such methods have incomplete consideration, resulting in lower accuracy of investment decision prediction for the user.
In view of the foregoing, embodiments of the present invention provide a method embodiment for investment decision processing, it should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system, such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
Fig. 1 is a flow chart of an investment decision processing method according to an embodiment of the present invention, as shown in fig. 1, the method comprising the steps of:
step S102, acquiring investment demand data corresponding to the N accounts respectively, wherein the investment demand data comprises index values corresponding to the M investment indexes respectively, and M, N is an integer greater than or equal to 2.
Optionally, the investment index data includes index values of the corresponding account under M investment indexes, where the M investment indexes may include, but are not limited to: age, occupation, credit rating, risk preference, investment deadline, investment preference, revenue index, and the like. And carrying out batch prediction of investment requirements based on the index data.
Step S104, based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively.
Optionally, in the decision tree model, one investment index of M investment indexes is taken as a root node, the attribute corresponding to M-1 indexes except one investment index is taken as an intermediate node, the first investment result is taken as a leaf node, fig. 2 is a schematic diagram of an alternative decision tree model according to an embodiment of the invention, as shown in fig. 2, in the decision tree model, the age is taken as the root node, the attribute corresponding to other nodes is taken as the leaf node, the corresponding attribute judgment condition is taken as a path between nodes, for example, a student is an attribute corresponding to investment index occupation, the credit level can also be taken as an attribute alone, and the judgment condition is less than or equal to 30, 31-40 and >40 is the attribute judgment condition of investment index age; general, good attribute determination conditions that can be used as investment index credit levels, etc.; the corresponding leaf node is whether to purchase the corresponding investment product, i.e. whether to invest. The decision tree model provides a clear decision path, can learn the basis of each decision node, so that the decision process is better understood, and can be used for identifying the most important features to help understand the key factors influencing investment decisions.
In an alternative embodiment, before the decision tree model is adopted to obtain the first investment decision results corresponding to the N accounts respectively based on the investment demand data corresponding to the N accounts respectively, the method further includes: acquiring investment demand data and actual investment decision results corresponding to L accounts respectively, wherein L is an integer greater than N; based on the investment demand data and the actual investment decision result corresponding to the L accounts respectively, one investment index of the M investment indexes is taken as a root node, the attributes corresponding to the M-1 indexes except the one investment index are taken as intermediate nodes, and the first investment result is taken as a leaf node, so that a decision tree model is constructed.
Optionally, the construction process of the decision tree model, that is, the training process of the decision tree model, is implemented in the model training process, specifically, by layer-by-layer splitting of the nodes. In order to solve the problem of batch decision making, the node is split layer by layer based on the investment indexes such as age, occupation, credit level, risk preference, investment period, investment preference, income index and the like corresponding to the L accounts respectively and the actual investment decision result, and the trained decision tree model based on the node is capable of learning the relation between each investment index and the first investment decision result, so that the accurate prediction of the first investment decision result is realized.
In an alternative embodiment, based on the investment requirement data corresponding to the L accounts respectively, taking one of the M investment indexes as a root node, taking the properties corresponding to the M-1 indexes except the one investment index respectively as intermediate nodes, taking the first investment result as a leaf node, and constructing a decision tree model, wherein the decision tree model comprises the following steps: the node splitting criteria are determined as: node splitting is carried out based on the foundation coefficients corresponding to the M investment indexes respectively; based on the investment demand data and the actual investment decision result which are respectively corresponding to the L accounts and the node splitting standard, taking one of the M investment indexes as a root node, taking the attribute respectively corresponding to the M-1 indexes as an intermediate node, and taking the first investment result as a leaf node to carry out node splitting to construct a decision tree model.
Alternatively, the key to decision tree model construction is the choice of node splitting criteria. The coefficient of the base corresponding to each investment index can be used as the standard of node splitting. In decision trees, the coefficient of base (giniimatrix) is an indicator used to measure node purity (i matrix). The role of the coefficient of the kunit is to help the decision tree algorithm determine the best splitting point, i.e. which feature and feature value to choose for splitting of the node. Specifically, the kunit evaluates the node's non-purity by measuring the distribution of the various categories in a node, with smaller values indicating higher purity of the node. In the construction process of the decision tree, the algorithm tries different features and feature values as splitting points, and then calculates the coefficient of the kennel of the sub-node after splitting, so that the splitting point with the lowest degree of non-purity can be selected as the optimal splitting strategy.
Alternatively, for each investment index, there will typically be multiple categories (e.g., K, one attribute for each category), and the probability of the kth category is p k The expression of the coefficient of the foundation corresponding to the investment index is:
where Gini (p) represents a coefficient of keni, k=1, 2, …, and K represents any one of the categories corresponding to the investment index.
In the class two classification problem, i.e. if the investment index corresponds to two classes, the calculation is simpler, if the probability of belonging to the output of the first sample class is p, the expression of the coefficient of the foundation is: gini (p) =2p (1-p).
Intuitively, the Gini coefficient Gini (p) reflects the purity of sample data, which is made up of all index values corresponding to the same investment index. If the sample data in the branch belong to the same class, the coefficient of the radix is 0, and the purity is highest, the branch is a leaf node, and calculation is not needed; if the attribute values of all the characteristics of the samples in the branches are multiple, the algorithm adopts a mode of 'minority compliance majority', and marks the category as one of the most samples in the current branch; if none of the above is met, the above operation should be repeated for each set of sample data, and branches continue to be split down in a recursive manner until the sample data for each branch has the same class. And so on until finally all nodes are classified.
And S106, based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain the investment probability corresponding to the N accounts respectively.
Optionally, in the logistic regression model, a plurality of investment indexes are taken as independent variables, corresponding investment probabilities are taken as dependent variables, the logistic regression model learns the relationships between the plurality of investment indexes and the investment probabilities, based on the logistic regression model, the relationships between the plurality of investment indexes and the investment probabilities can be modeled as an S-shaped curve, and under the condition that the investment demand data corresponding to the N accounts are known respectively, namely, under the condition that the index values of the plurality of investment indexes of the N accounts are known respectively, the investment probabilities corresponding to the N accounts can be obtained predictably based on the logistic regression model.
In an alternative embodiment, before the investment probability corresponding to each of the N accounts is obtained by adopting the logistic regression model based on the investment demand data corresponding to each of the N accounts, the method further includes: acquiring investment demand data and actual investment probability corresponding to P accounts respectively, wherein P is an integer greater than N; and carrying out logistic regression training based on the investment demand data and the actual investment probability respectively corresponding to the P accounts to obtain a logistic regression model.
Optionally, in order for the logistic regression model (i.e., the logic regression model) to learn the relationships between the multiple investment indexes and the investment probabilities, investment demand data and actual investment probabilities corresponding to the P accounts respectively are obtained in advance to perform logistic regression training, so as to obtain the logistic regression model, and the specific implementation process is as follows: the loss function is determined to be a logistic function (i.e., a sigmoid function) formulated as follows:
where z represents investment probability and phi (z) represents a loss function value.
The initial model is determined as follows:
z＝w T x＝w 0 x 0 +w 1 x 1 +…+w m x m
wherein w is T A weight matrix consisting of weight values corresponding to a plurality of investment indexes respectively, w i (i=1, 2, …, m) represents an index weight corresponding to any one of a plurality of investment indices, x i (i=1, 2, …, m) represents an index value corresponding to any one of the investment indices.
Based on the loss function, investment demand data and actual investment probability corresponding to the P accounts respectively, performing logistic regression training on the initial model to obtain a logistic regression model.
It should be noted that the logic regression does not rigidly set the classification result to 0 or 1, but gives a score in the interval 0-1, and that the closer the score is to 1 means the greater the probability of investment by the user.
Step S108, based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, obtaining second investment decision results corresponding to the N accounts respectively, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not.
Optionally, the first investment decision result is an investment decision result of a corresponding account predicted by a decision tree model and is used for indicating whether to perform investment, however, in a practical situation, a problem that a model is over-fitted may exist in the prediction of a single model for performing investment decision, so that the accuracy of the prediction result of the investment decision is lower, based on the prediction result, the determination of the second investment result can be comprehensively performed by combining the first investment decision result predicted by the decision tree model and the investment probability predicted by the logistic regression model, and the accuracy of the prediction of the investment decision result is improved.
Optionally, for obtaining the second investment result, whether the first investment result is consistent with the investment trend corresponding to the investment probability or not can be judged, and if so, the first investment result is used as a final second investment result; if the investment decision results are inconsistent, adjustment of the investment decision results can be performed, for example, the first investment result indicates investment, the corresponding investment probability is only 0.2, and if the investment probability is smaller, the investment tendencies of the two are inconsistent; correspondingly, if the first investment result indicates that the investment is performed, the corresponding investment probability is 0.9, and the investment probability is larger, the investment trends of the first investment result and the second investment result are consistent. Specifically, under the condition that the first investment result and the investment tendency corresponding to the investment probability are inconsistent, the first investment decision result can be corrected based on the corresponding investment probability, and the corrected first investment result is used as the second investment result, so that the accuracy of investment decision result prediction is improved.
In an alternative embodiment, based on the first investment decision results respectively corresponding to the N accounts and the investment probabilities respectively corresponding to the N accounts, obtaining the second investment decision results respectively corresponding to the N accounts includes: based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, a second investment decision result corresponding to any one of the N accounts is obtained in the following manner: detecting whether the corresponding investment probability is in a preset first probability interval or not under the condition that the first investment decision result of any account is investment; or detecting whether the corresponding investment probability is in a preset second probability interval under the condition that the first investment decision result of any account is that investment is not performed, wherein the upper limit value of the preset first probability interval is larger than the upper limit value of the preset second probability interval, and the lower limit value of the preset first probability interval is larger than the lower limit value of the preset second probability interval; and under the condition that the first investment decision result of any one account is investment, and the corresponding investment probability is in a preset first probability interval, determining the second investment decision result of any one account as follows: investment is carried out; or when the first investment decision result of any one account is investment, and the corresponding investment probability is not in the preset first probability interval, determining the second investment decision result of any one account as follows: investment is not performed; or when the first investment decision result of any one account is that no investment is performed, and the corresponding investment probability is in a preset second probability interval, determining the second investment decision result of any one account as follows: investment is not performed; or under the condition that the first investment decision result of any one account is that no investment is performed and the corresponding investment probability is not in a preset second probability interval, determining the second investment decision result of any one account as follows: and (5) investment is carried out.
In the above manner, according to the first investment decision result and the probability interval to which the corresponding investment probability belongs, whether the investment trends predicted based on the decision tree model and the logistic regression model are consistent or not may be determined, for example, if the first investment decision result indicates that the investment is performed, the corresponding investment probability is within a preset first probability interval (such as an interval greater than 0.5); or the first investment decision result indicates that the investment is not performed, and if the corresponding investment probability is in a preset second probability interval (such as an interval smaller than 0.5), the investment trends are determined to be consistent, and the first investment decision result is taken as a second investment decision result. If the first investment decision result indicates that investment is performed, the corresponding investment probability is not in a preset first probability interval; or the first investment decision result indicates that the investment is not carried out, the corresponding investment probability is not in a preset second probability interval, the inconsistent investment trend is determined at the moment, the first investment prediction result is required to be corrected, and the corrected first investment prediction result is used as a second investment prediction result. By the method, accuracy of investment decision result prediction can be improved.
In an alternative embodiment, after obtaining the second investment decision results corresponding to the N accounts based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, the method further includes: performing anomaly detection on the second investment decision results corresponding to the N accounts respectively to obtain an anomaly investment decision result; and correcting the abnormal investment decision results in the second investment decision results corresponding to the N accounts respectively to obtain third investment decision results corresponding to the N accounts respectively.
Optionally, in the process of carrying out investment decision prediction based on the decision tree model and the logistic regression model, there may be a situation that index data varies, so that a certain error exists in the second investment decision result obtained by the prediction, based on the error, abnormality detection and repair are carried out on the obtained second investment decision result, and the corrected third investment result is used as a final investment decision result, thereby improving the accuracy of investment decision prediction. And the method can be used for carrying out anomaly detection on the second investment decision results corresponding to the N accounts respectively based on a differential evolution algorithm to obtain an anomaly detection result.
In an alternative embodiment, performing anomaly detection on the second investment decision results corresponding to the N accounts respectively to obtain an abnormal investment decision result, including: determining an initialization population and a fitness function based on the investment demand data respectively corresponding to the N accounts and the second investment decision results respectively corresponding to the N accounts; performing mutation, crossover and selection operations on individuals included in the initialized population in sequence to generate new individuals; based on the fitness function, determining fitness values respectively corresponding to individuals included in the initialized population and fitness values corresponding to new individuals; updating the individuals included in the initialized population based on the fitness values respectively corresponding to the individuals included in the initialized population and the fitness values corresponding to the new individuals to obtain a new population; repeatedly executing the operation until reaching a preset termination condition; determining the prediction investment decision results corresponding to the N accounts respectively based on the individuals included in the new population output when the preset termination condition is reached; and determining the predicted investment decision result inconsistent with the corresponding predicted investment decision result in the second investment decision results corresponding to the N accounts as an abnormal investment decision result.
Alternatively, in the anomaly detection of the second investment decision result based on the differential evolution algorithm, first, it is necessary to initialize the population. Each individual in the population represents one possible solution. The initial setting of the population should be covered as much as possibleThe space is searched so that the algorithm can find a globally optimal solution. The fitness function is used to evaluate the quality of each individual in the population. In variation monitoring, the fitness function may be the score or accuracy of the anomaly detection model. By calculating the fitness function, the fitness of each individual in the population for the problem can be evaluated. For each individual, a new individual is generated by performing a mutation operation on other individuals in the population. Mutation operations are typically performed by linear combination of three individuals in a population, and the resulting new individual is referred to as a "variant". Through mutation operation, new genetic information can be introduced, so that the diversity of the population is increased. For example, in the case of performing a mutation operation based on a differential evolution algorithm, three individuals X need to be randomly selected from the population in the g-th iteration p1(g) ，X p2(g) ，X p3(g) And p1+.p2+.p3+.p1, wherein p1, p2, p3 represent the index or number of three randomly selected individuals. The resulting variation vector (i.e., variant) is X i(g) ＝X p1(g) +F×(X p2(g) -X p3(g) ) Wherein X is p2(g) -X p3(g) Can be understood as a differential vector, i.e. individual X p2(g) And X p3(g) Differences in corresponding positional elements in the g generation; f is a scaling factor, and the size of the differential step length of the population individuals is determined. The smaller F can influence the difference among population individuals, so that an algorithm result falls into local optimum, the larger F can enhance the global searching capability of the algorithm, is favorable for searching an optimum solution, and can influence the convergence speed of the algorithm. For F is generally in the range of [0,2 ]]And is typically 0.5.
Alternatively, for this scaling factor F, an adaptive adjustment may also be made, i.e. based on two individual adaptive changes that generate the differential vector. For example, three individuals randomly selected in the mutation operator are ranked from good to bad to obtain X b 、X m 、X w Corresponding fitness f b 、f m 、f w . The mutation operator is changed into V i ＝X b +F i +(X m -X w ) Wherein V is i Representing mutation operator, i representing mutation times, and F value according to the generated differential directionTwo individual adaptive changes in amount:wherein F is l =0.1 is the first weight coefficient; f (F) u =0.9 is the second weight coefficient.
Optionally, the variant is further cross-manipulated with the original individual to produce a progeny individual. Through crossover operations, new individuals may be generated that may contain genetic information from multiple parents. The selection operation is selected according to the fitness of the individual. By comparing fitness of the offspring individuals and the original individuals, the individuals with higher fitness are selected as the next generation population. In each iteration, new individuals are generated by mutation, crossover and selection operations, and the worst individuals in the population are then replaced with new individuals. This process continues until a stop condition is met (e.g., a preset number of iterations is reached or a satisfactory solution is found). After each iteration, the anomaly detection model is updated with the new population. And then, using a new model to monitor variation of the second investment decision results corresponding to the N accounts respectively. If the monitoring result of a certain second investment decision result is inconsistent with the model predicted result, the second investment decision result is considered abnormal. By the method, the abnormal investment result can be accurately identified in the second investment decision results corresponding to the N accounts respectively.
As an optional embodiment, after obtaining the third investment decision results corresponding to the N accounts respectively, determining whether to recommend the target investment product to the corresponding account based on the corresponding third investment decision results, if the third investment result indicates that the investment is performed, the target account is more likely to purchase the target investment product, and recommending the target investment product to the target account at this time, wherein the target account is any one account of the N accounts; if the third investment result indicates that no investment is being made, it indicates that the target account is less likely to purchase the target investment product, at which point the target investment product is not recommended to the target account. By the method, the effect of improving the recommendation accuracy of the investment products can be achieved, and the user experience is further improved.
Through the steps S102 to S108, the purposes of predicting the first investment decision result and the investment probability of a plurality of accounts in batches based on the decision tree model and the logistic regression model respectively and accurately determining the investment probability of the users in batches based on the first investment decision result and the investment probability can be achieved, so that the technical effects of improving the prediction efficiency and the prediction result accuracy of the batch investment decision are achieved, and the technical problem of low decision prediction result accuracy of the investment decision method in the related technology is solved.
Based on the above embodiments and optional embodiments, the present invention proposes an implementation of an optional investment decision processing method, the method comprising:
step S1, acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively, and M, N is an integer greater than or equal to 2;
step S2, based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively;
step S3, based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain investment probabilities corresponding to the N accounts respectively;
step S4, based on the first investment decision results respectively corresponding to the N accounts and the investment probabilities respectively corresponding to the N accounts, obtaining second investment decision results respectively corresponding to the N accounts, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not;
s5, carrying out anomaly detection on second investment decision results corresponding to the N accounts respectively based on a differential evolution algorithm to obtain an anomaly investment decision result;
and S6, correcting the abnormal investment decision results in the second investment decision results corresponding to the N accounts respectively to obtain third investment decision results corresponding to the N accounts respectively.
Through the steps S1 to S6, the purposes of predicting the first investment decision result and the investment probability of a plurality of accounts in batches based on the decision tree model and the logistic regression model respectively and accurately determining the investment probability of the users in batches based on the first investment decision result and the investment probability can be achieved, so that the technical effects of improving the prediction efficiency and the prediction result accuracy of the batch investment decision are achieved, and the technical problem of low decision prediction result accuracy of the investment decision method in the related technology is solved.
In this embodiment, an investment decision processing apparatus is further provided, and the apparatus is used to implement the foregoing embodiments and preferred embodiments, and will not be described in detail. As used below, the terms "module," "apparatus" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
According to an embodiment of the present invention, there is further provided an apparatus for implementing the above investment decision processing method, and fig. 3 is a schematic structural diagram of an investment decision processing apparatus according to an embodiment of the present invention, as shown in fig. 3, where the above investment decision processing apparatus includes: a first acquisition module 300, a first prediction module 302, a second prediction module 304, a decision module 306, wherein:
The first obtaining module 300 is configured to obtain investment demand data corresponding to N accounts, where the investment demand data includes index values corresponding to M investment indexes, and M, N is an integer greater than or equal to 2;
the first prediction module 302 is connected to the first obtaining module 300, and is configured to obtain first investment decision results corresponding to the N accounts respectively based on the investment demand data corresponding to the N accounts respectively by adopting a decision tree model;
the second prediction module 304 is connected to the first prediction module 302, and is configured to obtain investment probabilities corresponding to the N accounts respectively based on the investment demand data corresponding to the N accounts respectively by adopting a logistic regression model;
the decision module 306 is connected to the second prediction module 304, and is configured to obtain a second investment decision result corresponding to the N accounts based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, where the second investment decision result is used to indicate whether the corresponding accounts perform investment.
In the embodiment of the present invention, the first obtaining module 300 is configured to obtain investment demand data corresponding to N accounts, where the investment demand data includes index values corresponding to M investment indexes, where M, N is an integer greater than or equal to 2; the first prediction module 302 is connected to the first obtaining module 300, and is configured to obtain first investment decision results corresponding to the N accounts respectively based on the investment demand data corresponding to the N accounts respectively by adopting a decision tree model; the second prediction module 304 is connected to the first prediction module 302, and is configured to obtain investment probabilities corresponding to the N accounts respectively based on the investment demand data corresponding to the N accounts respectively by adopting a logistic regression model; the decision module 306 is connected to the second prediction module 304, and is configured to obtain a second investment decision result corresponding to the N accounts based on the first investment decision result corresponding to the N accounts and the investment probability corresponding to the N accounts, where the second investment decision result is used to indicate whether the corresponding accounts are invested, so as to predict the first investment decision result and the investment probability of the multiple accounts in batches based on the decision tree model and the logistic regression model, and accurately determine the investment probability of the user in batches based on the first investment decision result and the investment probability, thereby achieving the technical effect of improving the prediction efficiency and the prediction result accuracy of the batch investment decision, and further solving the technical problem of low accuracy of the decision prediction result in the investment decision method in the related technology.
In an alternative embodiment, the decision module includes: the first detection sub-module is used for detecting whether the corresponding investment probability is in a preset first probability interval or not under the condition that the first investment decision result of any account is investment; or detecting whether the corresponding investment probability is in a preset second probability interval under the condition that the first investment decision result of any account is that investment is not performed, wherein the upper limit value of the preset first probability interval is larger than the upper limit value of the preset second probability interval, and the lower limit value of the preset first probability interval is larger than the lower limit value of the preset second probability interval; the first determining submodule is used for determining the second investment decision result of any one account as follows when the first investment decision result of any one account is investment and the corresponding investment probability is in a preset first probability interval: investment is carried out; the second determining submodule is used for determining that the second investment decision result of any one account is when the first investment decision result of any one account is investment, and the corresponding investment probability is not in a preset first probability interval, and the second investment decision result of any one account is: the first investment decision result of any account is that the investment is not performed, and the second investment decision result of any account is determined under the condition that the corresponding investment probability is in a preset second probability interval: investment is not performed; a fourth determining submodule, configured to determine, when the first investment decision result of any one account is that no investment is performed and the corresponding investment probability is not within a preset second probability interval, that the second investment decision result of any one account is that: and (5) investment is carried out.
In an alternative embodiment, the apparatus further comprises: the second detection sub-module is used for carrying out abnormal detection on second investment decision results corresponding to the N accounts respectively to obtain abnormal investment decision results; the first correction sub-module is used for correcting the abnormal investment decision results in the second investment decision results corresponding to the N accounts respectively to obtain third investment decision results corresponding to the N accounts respectively.
In an alternative embodiment, the second detection sub-module includes: the population updating sub-module is used for determining an initialized population and a fitness function based on the investment demand data corresponding to the N accounts respectively and the second investment decision results corresponding to the N accounts respectively; performing mutation, crossover and selection operations on individuals included in the initialized population in sequence to generate new individuals; based on the fitness function, determining fitness values respectively corresponding to individuals included in the initialized population and fitness values corresponding to new individuals; updating the individuals included in the initialized population based on the fitness values respectively corresponding to the individuals included in the initialized population and the fitness values corresponding to the new individuals to obtain a new population; repeatedly executing the operation until reaching a preset termination condition; a fifth determining submodule, configured to determine predicted investment decision results corresponding to the N accounts respectively based on the individuals included in the new population output when the preset termination condition is reached; and the sixth determining submodule is used for determining a predicted investment decision result which is inconsistent with the corresponding predicted investment decision result in the second investment decision results corresponding to the N accounts as an abnormal investment decision result.
In an alternative embodiment, the apparatus further comprises: the first acquisition submodule is used for acquiring investment demand data and actual investment decision results corresponding to L accounts respectively, wherein L is an integer greater than N; the first construction submodule is used for constructing a decision tree model based on the investment demand data and the actual investment decision result which are respectively corresponding to the L accounts, taking one investment index of the M investment indexes as a root node, taking the attributes which are respectively corresponding to the M-1 indexes except the one investment index as intermediate nodes, and taking the first investment result as a leaf node.
In an alternative embodiment, the first building sub-module includes: a seventh determining submodule, configured to determine a node splitting criterion as: node splitting is carried out based on the foundation coefficients corresponding to the M investment indexes respectively; the second construction submodule is used for carrying out node splitting by taking one investment index of M investment indexes as a root node and the attribute corresponding to M-1 indexes as an intermediate node based on the investment demand data and the actual investment decision result which are respectively corresponding to the L accounts and the node splitting standard, and the first investment result is taken as a leaf node to carry out node splitting to construct a decision tree model.
In an alternative embodiment, the apparatus further comprises: the second acquisition sub-module is used for acquiring the investment demand data and the actual investment probability corresponding to the P accounts respectively, wherein P is an integer greater than N; and the first training sub-module is used for carrying out logistic regression training based on the investment demand data and the actual investment probability corresponding to the P accounts respectively to obtain a logistic regression model.
It should be noted that each of the above modules may be implemented by software or hardware, for example, in the latter case, it may be implemented by: the above modules may be located in the same processor; alternatively, the various modules described above may be located in different processors in any combination.
It should be noted that, the first obtaining module 300, the first predicting module 302, the second predicting module 304, and the decision module 306 correspond to steps S102 to S108 in the embodiment, and the modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the disclosure of the foregoing embodiments. It should be noted that the above modules may be run in a computer terminal as part of the apparatus.
It should be noted that, the optional or preferred implementation manner of this embodiment may be referred to the related description in the embodiment, and will not be repeated herein.
The investment decision processing apparatus may further include a processor and a memory, wherein the first obtaining module 300, the first predicting module 302, the second predicting module 304, the deciding module 306, etc. are stored in the memory as program modules, and the processor executes the program modules stored in the memory to implement corresponding functions.
The processor comprises a kernel, the kernel accesses the memory to call the corresponding program module, and the kernel can be provided with one or more than one. The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.
According to an embodiment of the present application, there is also provided an embodiment of a nonvolatile storage medium. Optionally, in this embodiment, the nonvolatile storage medium includes a stored program, where the device in which the nonvolatile storage medium is located is controlled to execute any one of the investment decision processing methods when the program runs.
Alternatively, in this embodiment, the above-mentioned nonvolatile storage medium may be located in any one of the computer terminals in the computer terminal group in the computer network or in any one of the mobile terminals in the mobile terminal group, and the above-mentioned nonvolatile storage medium includes a stored program.
Optionally, the program controls the device in which the nonvolatile storage medium is located to perform the following functions when running: acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively, and M, N is an integer greater than or equal to 2; based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively; based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain investment probabilities corresponding to the N accounts respectively; and obtaining second investment decision results corresponding to the N accounts respectively based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not.
According to an embodiment of the present application, there is also provided an embodiment of a processor. Optionally, in this embodiment, the processor is configured to run a program, where any one of the investment decision processing methods is executed when the program runs.
According to an embodiment of the present application, there is also provided an embodiment of a computer program product adapted to perform a program initialized with the steps of any one of the investment decision processing methods described above when executed on a data processing device.
Optionally, the computer program product mentioned above, when executed on a data processing device, is adapted to perform a program initialized with the method steps of: acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively, and M, N is an integer greater than or equal to 2; based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively; based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain investment probabilities corresponding to the N accounts respectively; and obtaining second investment decision results corresponding to the N accounts respectively based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not.
The embodiment of the invention provides an electronic device, which comprises a processor, a memory and a program stored on the memory and capable of running on the processor, wherein the following steps are realized when the processor executes the program: acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively, and M, N is an integer greater than or equal to 2; based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively; based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain investment probabilities corresponding to the N accounts respectively; and obtaining second investment decision results corresponding to the N accounts respectively based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not.
The above-described order of embodiments of the invention is merely for illustration and does not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the modules may be a logic function division, and there may be another division manner when actually implemented, for example, a plurality of modules or components may be combined or may be integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with respect to each other may be through some interface, module or indirect coupling or communication connection of modules, electrical or otherwise.
The modules described above as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, i.e., may be located in one place, or may be distributed over a plurality of modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist alone physically, or two or more modules may be integrated into one module. The integrated modules may be implemented in hardware or in software functional modules.
The integrated modules described above, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable non-volatile storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a non-volatile storage medium, including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of the embodiments of the present invention. And the aforementioned nonvolatile storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (10)
1. A method of investment decision processing comprising:
acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively, and M, N is an integer greater than or equal to 2;
based on the investment demand data corresponding to the N accounts respectively, a decision tree model is adopted to obtain first investment decision results corresponding to the N accounts respectively;
based on the investment demand data corresponding to the N accounts respectively, adopting a logistic regression model to obtain investment probabilities corresponding to the N accounts respectively;
and obtaining second investment decision results corresponding to the N accounts respectively based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not.
2. The method of claim 1, wherein the obtaining the second investment decision results for the N accounts based on the first investment decision results for the N accounts, respectively, and the investment probabilities for the N accounts, respectively, comprises:
based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, a second investment decision result corresponding to any one of the N accounts is obtained in the following manner:
detecting whether the corresponding investment probability is in a preset first probability interval or not under the condition that the first investment decision result of any account is investment; or detecting whether the corresponding investment probability is in a preset second probability interval under the condition that the first investment decision result of any account is that investment is not performed, wherein the upper limit value of the preset first probability interval is larger than the upper limit value of the preset second probability interval, and the lower limit value of the preset first probability interval is larger than the lower limit value of the preset second probability interval;
and determining that the second investment decision result of any one account is the following under the condition that the first investment decision result of any one account is investment and the corresponding investment probability is in the preset first probability interval: investment is carried out; or alternatively
And determining that the second investment decision result of any one account is the following under the condition that the first investment decision result of any one account is investment and the corresponding investment probability is not within the preset first probability interval: investment is not performed; or alternatively
And under the condition that the first investment decision result of any one account is that no investment is performed and the corresponding investment probability is in the preset second probability interval, determining the second investment decision result of any one account as follows: investment is not performed; or alternatively
And under the condition that the first investment decision result of any one account is not invested and the corresponding investment probability is not within the preset second probability interval, determining the second investment decision result of any one account as follows: and (5) investment is carried out.
3. The method of claim 1, wherein after the obtaining the second investment decision results respectively corresponding to the N accounts based on the first investment decision results respectively corresponding to the N accounts and the investment probabilities respectively corresponding to the N accounts, the method further comprises:
performing anomaly detection on the second investment decision results corresponding to the N accounts respectively to obtain an anomaly investment decision result;
And correcting the abnormal investment decision results in the second investment decision results corresponding to the N accounts respectively to obtain third investment decision results corresponding to the N accounts respectively.
4. The method of claim 3, wherein the performing anomaly detection on the second investment decision results corresponding to the N accounts respectively to obtain an anomaly investment decision result comprises:
determining an initialization population and a fitness function based on the investment demand data respectively corresponding to the N accounts and the second investment decision results respectively corresponding to the N accounts; performing mutation, crossover and selection operations on individuals included in the initialized population in sequence to generate new individuals; determining fitness values corresponding to the individuals included in the initialized population respectively and the fitness values corresponding to the new individuals based on the fitness function; updating the individuals included in the initial population based on the fitness values respectively corresponding to the individuals included in the initial population and the fitness values corresponding to the new individuals to obtain a new population; repeatedly executing the operation until reaching a preset termination condition;
Determining the prediction investment decision results corresponding to the N accounts respectively based on the individuals included in the new population output when the preset termination condition is reached;
and determining the predicted investment decision result inconsistent with the corresponding predicted investment decision result in the second investment decision results respectively corresponding to the N accounts as the abnormal investment decision result.
5. The method according to any one of claims 1 to 4, wherein before the obtaining, based on the investment demand data respectively corresponding to the N accounts, a decision tree model, a first investment decision result respectively corresponding to the N accounts, the method further comprises:
acquiring investment demand data and actual investment decision results corresponding to L accounts respectively, wherein L is an integer greater than N;
and based on the investment demand data and the actual investment decision result respectively corresponding to the L accounts, taking one of the M investment indexes as a root node, taking the attributes respectively corresponding to the M-1 indexes except the one investment index as intermediate nodes, and taking the first investment result as a leaf node to construct the decision tree model.
6. The method of claim 5, wherein the constructing the decision tree model based on the investment demand data corresponding to the L accounts respectively, with one of the M investment indices as a root node, the M-1 indices other than the one investment index respectively corresponding to the attributes as intermediate nodes, and the first investment result as a leaf node, comprises:
The node splitting criteria are determined as: node splitting is carried out based on the base coefficient corresponding to each of the M investment indexes;
based on the investment demand data and the actual investment decision result which are respectively corresponding to the L accounts and the node splitting standard, taking one of the M investment indexes as a root node, taking the attribute respectively corresponding to the M-1 indexes as an intermediate node, and taking the first investment result as a leaf node to carry out node splitting, so as to construct the decision tree model.
7. The method according to any one of claims 1 to 4, wherein before the obtaining, based on the investment demand data corresponding to the N accounts respectively, an investment probability corresponding to the N accounts respectively using a logistic regression model, the method further comprises:
acquiring investment demand data and actual investment probability corresponding to P accounts respectively, wherein P is an integer greater than N;
and carrying out logistic regression training based on the investment demand data and the actual investment probability respectively corresponding to the P accounts to obtain the logistic regression model.
8. An investment decision processing apparatus, comprising:
the first acquisition module is used for acquiring investment demand data corresponding to N accounts respectively, wherein the investment demand data comprises index values corresponding to M investment indexes respectively, and M, N is an integer greater than or equal to 2;
The first prediction module is used for obtaining first investment decision results corresponding to the N accounts respectively by adopting a decision tree model based on the investment demand data corresponding to the N accounts respectively;
the second prediction module is used for obtaining investment probabilities corresponding to the N accounts respectively by adopting a logistic regression model based on the investment demand data corresponding to the N accounts respectively;
the decision module is used for obtaining second investment decision results corresponding to the N accounts respectively based on the first investment decision results corresponding to the N accounts respectively and the investment probabilities corresponding to the N accounts respectively, wherein the second investment decision results are used for indicating whether the corresponding accounts invest or not.
9. A non-volatile storage medium, characterized in that it stores a plurality of instructions adapted to be loaded by a processor and to perform the investment decision processing method of any one of claims 1 to 7.
10. An electronic device comprising one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the investment decision processing method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311737559.8A CN117726444A (en) | 2023-12-17 | 2023-12-17 | Investment decision processing method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311737559.8A CN117726444A (en) | 2023-12-17 | 2023-12-17 | Investment decision processing method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117726444A true CN117726444A (en) | 2024-03-19 |
Family
ID=90206554
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311737559.8A Pending CN117726444A (en) | 2023-12-17 | 2023-12-17 | Investment decision processing method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117726444A (en) |
-
2023
- 2023-12-17 CN CN202311737559.8A patent/CN117726444A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829057B (en) | Knowledge graph entity semantic space embedding method based on graph second-order similarity | |
| CN112734128B (en) | 7-day power load peak prediction method based on optimized RBF | |
| CN109858714B (en) | Tobacco shred quality inspection index prediction method, device and system based on improved neural network | |
| CN105786711A (en) | Data analysis method and device | |
| CN112529683A (en) | Method and system for evaluating credit risk of customer based on CS-PNN | |
| CN111797320A (en) | Data processing method, device, equipment and storage medium | |
| CN115394358A (en) | Single cell sequencing gene expression data interpolation method and system based on deep learning | |
| CN115409292A (en) | Short-term load prediction method for power system and related device | |
| CN113449919A (en) | Power consumption prediction method and system based on feature and trend perception | |
| CN117726444A (en) | Investment decision processing method and device, storage medium and electronic equipment | |
| CN115713389A (en) | Financial product recommendation method and device | |
| EP3929928A1 (en) | Associating pedigree scores and similarity scores for plant feature prediction | |
| CN113129108B (en) | Product recommendation method and device based on Double DQN algorithm | |
| CN114820074A (en) | Target user group prediction model construction method based on machine learning | |
| CN113537693A (en) | Personnel risk level obtaining method, terminal and storage device | |
| CN117539520B (en) | Firmware self-adaptive upgrading method, system and equipment | |
| CN117422314B (en) | Enterprise data evaluation method and equipment based on big data analysis | |
| CN117974209A (en) | XGBoost model-based user behavior prediction method | |
| CN113837266B (en) | Software defect prediction method based on feature extraction and Stacking ensemble learning | |
| CN116721695B (en) | Identification method, device, equipment and medium of candidate gene for regulating bacterial shape | |
| CN112163714B (en) | XGboost-based campus client group load release algorithm | |
| CN112988564B (en) | SRGM decision model considering cost-reliability and construction method thereof | |
| CN117114735A (en) | Marketing data statistics method and device | |
| CN117635182A (en) | Target recommendation method and device | |
| CN115049495A (en) | Fluctuation rate prediction method based on feature selection and learning model fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |