CN111326254A - Methods, devices, equipment and media for indicator intervention - Google Patents

Abstract

The invention provides a method, a device, equipment and a medium for index intervention. Obtaining an arithmetic mechanism for providing an output result for an input sample, wherein the input sample comprises a plurality of indicators; determining a mapping relation between a controllable index and a response index in a plurality of indexes, wherein the controllable index is an index set of which the value can be adjusted, and the response index is an index set of which the value is difficult to adjust but can change according to the change of the controllable index; based on the mapping relation, the controllable indexes of the input samples are adjusted, so that the output result obtained by the operation of the operation mechanism aiming at the adjusted input samples meets the preset requirement; and providing the adjustment result of the controllable index for index intervention. Therefore, index intervention is carried out according to the adjustment result of the controllable index, expected operation output can be facilitated, and the method is easy to implement.

Description

Methods, devices, equipment and media for indicator intervention

technical field

The present invention generally relates to the technical field of data science, and more particularly, to a method for performing indicator intervention, an apparatus for performing indicator intervention, a computing device and a non-transitory machine-readable storage medium.

Background technique

Providing output results for input samples by constructing computing mechanisms (eg, predictive models (eg, machine learning models), rule sets of one or more rules, function operations, etc.) A technique commonly used in other fields. For example, it is possible to build a forecasting model to predict the output value or loss of machines to guide industrial production; it is also possible to build a profit forecasting model to forecast the user's profit to provide a basis for users to make decisions; it is also possible to build a forecast A model of the probability of disease, a model of predicting weight, etc., manage the physical condition of the user.

After the operation mechanism is established, for a given sample, the corresponding output result can be obtained. However, for the input object, the possible results are often obtained only by means of the computing mechanism, but they do not know how to make targeted adjustments to obtain the desired results.

SUMMARY OF THE INVENTION

Exemplary embodiments of the present invention are to provide a method and apparatus for index intervention to solve the above-mentioned technical problems.

According to a first aspect of the present invention, a method for index intervention is proposed, including: acquiring an operation mechanism for providing an output result for an input sample, wherein the input sample includes a plurality of indicators; determining the The mapping relationship between the controllable index and the response index among the multiple indexes, wherein the controllable index is the set of indexes whose values can be adjusted, and the response index is the value of which is difficult to adjust but can be adjusted according to the The set of indicators that changes with the change of the controllable indicators; based on the mapping relationship, the controllable indicators of the input samples are adjusted, so that the output results obtained by the operation mechanism operating on the adjusted input samples meet the predetermined requirements ; and provide adjustment results of controllable indicators for indicator intervention.

Optionally, the operation mechanism includes at least one of the following items: a machine learning model, a rule set, and a function operation.

Optionally, the operation mechanism is a prediction model that provides prediction results for the prediction problem, wherein the input sample is a prediction sample of the prediction model, and its features include multiple indicators related to the prediction problem.

Optionally, the step of determining the mapping relationship between the controllable index and the response index among the multiple indexes includes: constructing a value prediction based on the controllable index according to the multiple indexes included in the characteristics of the samples in the sample set. The model of the value of the response indicator to determine the mapping relationship between the controllable indicator and the response indicator; or according to the multiple indicators included in the characteristics of the samples in the sample set, construct a static indicator-based value and controllable indicator. A model for predicting the value of a response indicator by taking a value to determine the mapping relationship between a controllable indicator and a response indicator, wherein a static indicator is a set of indicators whose values cannot be changed.

Optionally, the step of determining the mapping relationship between the controllable index and the response index among the multiple indexes includes: determining the mapping relationship between the controllable index and the response index among the multiple indexes according to prior knowledge.

Optionally, the method further includes: grouping the controllable indicators, so as to divide the indicators with a linkage relationship among the controllable indicators into the same group, and the step of adjusting the controllable indicators among the characteristics of the predicted samples includes: Within the value range of the controllable indicators, the controllable indicators are adjusted on a group-by-group basis based on the linkage relationship between the indicators in the same group.

Optionally, the linkage relationship includes a correlation relationship and/or a causal relationship, and the step of grouping the controllable indicators includes: determining the indicators with a correlation relationship among the controllable indicators by means of calculation, so as to classify the correlation relationship among the controllable indicators. and/or according to external input, determine the indicators with causal relationship among the controllable indicators, so as to classify the indicators with causal relationship among the controllable indicators into the same group.

Optionally, the step of determining the indicators with correlation among the controllable indicators by means of calculation includes: calculating the correlation between indicators in the controllable indicators by means of independence verification.

Optionally, the step of adjusting the controllable indicators in the characteristics of the predicted samples includes: using the prediction results obtained by the prediction model for the adjusted predicted samples to meet the predetermined requirements and to minimize the adjustment cost of the controllable indicators as: The target is to adjust the controllable indicators among the characteristics of the predicted sample.

According to a second aspect of the present invention, there is also provided a method for index intervention, comprising: obtaining a disease probability prediction model; using the disease probability prediction model to predict the user's disease probability according to the user's characteristics, wherein, The characteristics of the user include multiple indicators related to the predicted probability of disease; in the case that the predicted probability of the user's disease is higher than a predetermined threshold, the mapping relationship between the static indicators and the controllable indicators to the response indicators is based on the multiple indicators , adjust the controllable indicators in the user's characteristics, so that the disease probability predicted by the disease probability prediction model for the adjusted user's characteristics is lower than the predetermined threshold, wherein the static index is its value that cannot be changed The controllable index is the set of indicators whose value can be adjusted, the response index is the set of indicators whose value itself is difficult to adjust but can be changed according to the change of the controllable index; and the adjustment result of the controllable index is provided to facilitate The user performs indicator intervention.

Optionally, the static indicators include at least one of the following: age, gender, height, and occupation; the controllable indicators include at least one of the following: weight, one or more lifestyle-related characteristics, blood pressure, and blood sugar; the response indicators include the following At least one item: uric acid, blood lipids.

Optionally, the method further includes: determining the mapping relationship between the static index and the controllable index among the multiple indexes to the response index.

Optionally, the step of determining the mapping relationship between the static index and the controllable index among the multiple indexes to the response index includes: according to the multiple indexes included in the characteristics of the samples in the sample set, constructing a value and The value of the controllable index predicts the model of the value of the response index to determine the mapping relationship; or according to prior knowledge, the mapping relationship between the static index and the controllable index among the multiple indexes to the response index is determined.

Optionally, the method further includes: grouping the controllable indicators, so as to divide the indicators with a linkage relationship among the controllable indicators into the same group, and the step of adjusting the controllable indicators among the characteristics of the predicted samples includes: Within the value range of the controllable indicators, the controllable indicators are adjusted on a group-by-group basis based on the linkage relationship between the indicators in the same group.

Optionally, the linkage relationship includes a correlation relationship and/or a causal relationship, and the step of grouping the controllable indicators includes: determining the indicators with a correlation relationship among the controllable indicators by means of calculation, so as to classify the correlation relationship among the controllable indicators. and/or according to external input, determine the indicators with causal relationship among the controllable indicators, so as to classify the indicators with causal relationship among the controllable indicators into the same group.

Optionally, the step of determining the indicators with correlation among the controllable indicators by means of calculation includes: calculating the correlation between indicators in the controllable indicators by means of independence verification.

Optionally, the step of adjusting the controllable index among the characteristics of the predicted sample includes: a disease probability obtained by predicting the characteristics of the adjusted user with a disease probability prediction model is lower than a predetermined threshold and minimizes the risk of disease. The adjustment cost of the control index is the target, and the controllable index among the characteristics of the prediction sample is adjusted.

According to a third aspect of the present invention, there is also provided an apparatus for performing index intervention, comprising: an acquisition unit for acquiring an operation mechanism for providing an output result for an input sample, wherein the input sample includes multiple a determination unit, configured to determine a mapping relationship between a controllable indicator and a response indicator among the multiple indicators, wherein the controllable indicator is a set of indicators whose values can be adjusted, and the response indicator is a set of indicators whose values can be adjusted. The index is an index set whose value itself is difficult to adjust but can be changed according to the change of the controllable index; the adjustment unit is used to adjust the controllable index of the input sample based on the mapping relationship, so that the calculation The mechanism operates on the adjusted input samples and obtains an output result that meets a predetermined requirement; and provides a unit for providing the adjustment result of the controllable index for index intervention.

Optionally, the operation mechanism includes at least one of the following items: a machine learning model, a rule set, and a function operation.

Optionally, the operation mechanism is a prediction model that provides prediction results for the prediction problem, wherein the input sample is a prediction sample of the prediction model, and its features include multiple indicators related to the prediction problem.

Optionally, the determining unit constructs a model for predicting the value of the response indicator based on the value of the controllable indicator according to multiple indicators included in the characteristics of the samples in the sample set, so as to determine the mapping between the controllable indicator and the response indicator. relationship, or the determining unit constructs a model for predicting the value of the response indicator based on the value of the static indicator and the value of the controllable indicator according to multiple indicators included in the characteristics of the samples in the sample set, so as to determine the value of the controllable indicator and the response indicator. The mapping relationship between indicators, where a static indicator is a set of indicators whose values cannot be changed.

Optionally, the determining unit determines the mapping relationship between the controllable index and the response index among the multiple indexes according to the prior knowledge.

Optionally, the device further includes: a grouping unit for grouping the controllable indicators, so as to divide the indicators with linkage relationship among the controllable indicators into the same group, and the adjustment unit is based on the value range of the controllable indicators. The linkage relationship between the indicators in the same group, and the controllable indicators are adjusted on a group basis.

Optionally, the linkage relationship includes a correlation relationship and/or a causal relationship, and the grouping unit determines the indicators with a correlation relationship among the controllable indicators by calculation, so as to divide the indicators with a correlation relationship among the controllable indicators into the same group, and /Or the grouping unit determines the indicators with causal relationship among the controllable indicators according to the external input, so as to divide the indicators with causal relationship among the controllable indicators into the same group.

Optionally, the grouping unit calculates the correlation between the indicators in the controllable indicators by means of independence check.

Optionally, the adjustment unit adjusts the controllable index among the characteristics of the predicted sample with the goal that the prediction result obtained by the prediction model for the adjusted prediction sample satisfies the predetermined requirement and minimizes the adjustment cost of the controllable index. .

According to a fourth aspect of the present invention, there is also provided an apparatus for performing index intervention, comprising: an acquisition unit for acquiring a disease probability prediction model; a prediction unit for using the disease probability prediction model according to the user's The feature predicts the probability of disease of the user, wherein the feature of the user includes a plurality of indicators related to the predicted probability of disease; the adjustment unit is used for predicting that the probability of disease of the user is higher than a predetermined threshold, based on multiple indicators The mapping relationship between the static indicators and the controllable indicators in the indicators to the response indicators, adjust the controllable indicators in the user's characteristics, so that the disease probability prediction model predicts the patient's disease according to the adjusted user's characteristics. The probability of disease is lower than a predetermined threshold, among which, the static index is the set of indicators whose value cannot be changed, the controllable index is the set of indicators whose value can be adjusted, and the response index is the value of which is difficult to adjust but can be adjusted according to the controllable index. A set of indicators that change with the change of the indicator; and a providing unit, which is used to provide the adjustment result of the controllable indicator so that the user can perform indicator intervention.

Optionally, the static indicators include at least one of the following: age, gender, height, and occupation; the controllable indicators include at least one of the following: weight, one or more lifestyle-related characteristics, blood pressure, and blood sugar; the response indicators include the following At least one item: uric acid, blood lipids.

Optionally, the apparatus further includes: a determining unit, configured to determine the mapping relationship between the static index and the controllable index among the multiple indexes to the response index.

Optionally, the determining unit constructs a model for predicting the value of the response indicator based on the value of the static indicator and the value of the controllable indicator according to multiple indicators included in the characteristics of the samples in the sample set, so as to determine the mapping relationship; or The determining unit determines the mapping relationship according to the prior knowledge.

Optionally, the device further includes: a grouping unit for grouping the controllable indicators, so as to divide the indicators with linkage relationship among the controllable indicators into the same group, and the adjustment unit is based on the value range of the controllable indicators. The linkage relationship between the indicators in the same group, and the controllable indicators are adjusted on a group basis.

Optionally, the linkage relationship includes a correlation relationship and/or a causal relationship, and the grouping unit determines the indicators with a correlation relationship among the controllable indicators by calculation, so as to divide the indicators with a correlation relationship among the controllable indicators into the same group, and /Or the grouping unit determines the indicators with causal relationship among the controllable indicators according to the external input, so as to divide the indicators with causal relationship among the controllable indicators into the same group.

Optionally, the grouping unit calculates the correlation between the indicators in the controllable indicators by means of independence check.

Optionally, the adjustment unit takes the disease probability predicted by the disease probability prediction model according to the characteristics of the adjusted user as the goal, which is lower than the predetermined threshold and reduces the adjustment cost of the controllable index as much as possible, among the characteristics of the predicted sample. controllable indicators.

According to a fifth aspect of the present invention, there is also provided a computing device, comprising: a processor; and a memory on which executable codes are stored, and when the executable codes are executed by the processor, the processor is caused to execute the method according to the present invention The method of the first aspect or the second aspect.

According to a sixth aspect of the present invention, there is also provided a non-transitory machine-readable storage medium on which executable codes are stored, and when the executable codes are executed by a processor of an electronic device, the processor is made to execute the A method as described in the first or second aspect of the invention.

According to the method and device for index intervention according to the exemplary embodiments of the present invention, a reasonable index adjustment scheme can be provided. Index intervention according to the adjustment result of the controllable index can promote the desired operation output. Compared with the index control scheme that directly uses the optimization method without distinguishing the characteristics of the predicted samples to obtain the desired operation output, the adjustment result of the controllable index provided by the present invention is more reasonable. The adjustment results of the indicators can be achieved and easy to achieve when the indicators are intervened.

Additional aspects and/or advantages of the present general inventive concept will be set forth in part in the ensuing description, and in part will be apparent from the description, or may be learned by practice of the present general inventive concept.

Description of drawings

The above and other objects and features of the exemplary embodiments of the present invention will become more apparent from the following description in conjunction with the accompanying drawings that exemplarily illustrate the embodiments, in which:

FIG. 1 shows a flowchart of a method for index intervention according to an exemplary embodiment of the present invention;

Fig. 2 shows the flow chart of the method in which the present invention is applied to intervene on the physical index of the user to reduce the risk of the user's disease;

FIG. 3 shows a block diagram of an apparatus for performing index intervention according to an exemplary embodiment of the present invention;

FIG. 4 shows a block diagram of an apparatus for performing index intervention according to another exemplary embodiment of the present invention;

FIG. 5 shows a schematic structural diagram of a computing device that can be used to implement the above method according to an exemplary embodiment of the present invention.

Detailed ways

Reference will now be made in detail to the embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like parts throughout. The embodiments will be described below in order to explain the present invention by referring to the figures.

For a given sample, it is input to a specific operation mechanism to obtain an output result, which is a typical process of using operation to obtain the expected result. Under certain constraints, it is an optimization process to find a suitable input set to optimize the objective function. How to give appropriate suggestions for the indicators contained in a given sample, so that the changes of the indicators will lead to the expected operation output, but it cannot be achieved by simply combining the optimization method and the operation mechanism. The main problem is that the index control scheme obtained by using the optimization method for the operation mechanism may not be realized or easy to realize.

This is because the input sample may include multiple indicators related to the operation problem, and the values of these multiple indicators may not necessarily be adjusted. , there may also be certain constraints, for example, there are some indicators that are difficult to be directly adjusted but will change with the changes of other indicators. Therefore, in the prior art, the indexes of the samples are not differentiated, and the optimization method is directly used to obtain the index control scheme that can promote the desired output.

Taking disease risk prediction as an example, it is assumed that the input samples of the disease risk prediction model include age, gender, height, weight, one or more lifestyle-related characteristics, blood pressure, blood sugar, uric acid, blood lipids and other indicators. The characteristics related to the lifestyle may include, but are not limited to, indicators that can be adjusted artificially, such as work and rest time, diet, and exercise. If the disease risk prediction model predicts and predicts that the user has a higher risk of disease according to the user's existing indicators, it is necessary to provide the user with an appropriate indicator adjustment suggestion, so that the user will have a lower probability of disease after adjusting the indicator according to the indicator adjustment suggestion. If the above indicators are not distinguished, the optimization method is directly used to calculate the indicator adjustment plan with the lowest or lower probability of disease. Obviously, adjustment for age, gender, and height is impossible, and adjustment for indicators such as uric acid and blood lipids is also difficult for users (because users generally do not know how to adjust uric acid and blood lipids). Therefore, the index adjustment scheme calculated directly by the optimization method is not necessarily reasonable.

In view of this, the present invention proposes a scheme for index intervention. FIG. 1 shows a flowchart of a method for index intervention according to an exemplary embodiment of the present invention. The method shown in FIG. 1 can be implemented by a computer program, or can also be implemented by a special device for performing index intervention.

In step S110, an operation mechanism for providing an output result for an input sample is obtained, wherein the input sample may include a plurality of indicators. The operation mechanism can be any operation process that can obtain corresponding output results through specific operations for input samples. As an example, the operation mechanism may include at least one of the following: a machine learning model, a set of rules (the set of rules may include at least one rule), a function operation. Various combinations can be made between the above items, such as the combination of machine learning models and expert rules. Furthermore, the metrics may indicate any attributes or characteristics used to describe the input sample itself or its external association.

Specifically, the computing mechanism may be a prediction model that provides a prediction result for a prediction problem, wherein the input sample is a prediction sample of the prediction model, and its features include multiple indicators related to the prediction problem.

Here, the prediction model may be a machine learning model trained based on a machine learning algorithm, a statistical model constructed based on a statistical method, or a model constructed based on other methods, which is not limited in the present invention. In this step, a pre-trained prediction model may be received from the outside, or a prediction model may be constructed according to the prediction problem. The construction process of the prediction model will not be repeated here.

The prediction model mentioned in the present invention may be a prediction model under various application scenarios. For example, it can be a model for predicting disease probability, a model for predicting body weight, a model for predicting machine wear and tear, a model for predicting profit, and the like. By inputting the features of the predicted samples into the prediction model, the prediction results provided by the prediction model for the prediction problem can be obtained.

The features of the prediction samples of the prediction model include multiple indicators related to the prediction problem. For the multiple indicators included in the characteristics of the predicted samples, these indicators can be some attribute information, and their values have their own characteristics based on factors such as business meaning and/or common sense (for example, whether it can be changed, the value range of the change, the cost of change, etc.). As an example, according to externally input prior knowledge (business meaning, common sense, etc.), multiple indicators included in the characteristics of the predicted samples can be divided into static indicators (if any), controllable indicators and/or response indicators. Among them, the static index is the index set whose value cannot be changed, the controllable index is the index set whose value can be adjusted, and the response index is the index set whose value is difficult to adjust but can be changed according to the change of the controllable index. .

Taking the prediction model being a disease risk prediction model as an example, the characteristics of the prediction sample may include multiple indicators related to disease risk prediction, such as but not limited to the user's age, gender, height, weight, one or more of the user's life style. Related characteristics (such as work and rest time, diet, exercise, etc.), blood pressure, blood sugar, uric acid, blood lipids and other indicators. According to prior knowledge (relevant medical knowledge and common sense), indicators such as age, gender, and height that cannot be changed artificially are static indicators. Values such as weight, lifestyle-related characteristics, blood pressure, and blood sugar can be The adjusted indicators can be divided into controllable indicators. The values such as uric acid and blood lipid are difficult for users to adjust, but they can change with the changes of controllable indicators such as weight, lifestyle-related characteristics, blood pressure, and blood sugar. , so uric acid and blood lipids can be divided into response indicators.

It should be noted that when multiple indicators related to the prediction problem included in the characteristics of the prediction samples are divided into static indicators, controllable indicators or response indicators based on prior knowledge (business meaning, common sense, etc.), specific indicators can also be combined. Scenario to determine the category to which the indicator belongs. For example, for the indicator "weight", its value can be adjusted, so it can be divided into controllable indicators, but "weight" can also be regarded as a change in indicators such as "diet" and "exercise". Therefore, “weight” can also be divided into response indicators, and whether it is divided into controllable indicators or response indicators can be determined according to the actual situation.

In step S120, the mapping relationship between the controllable index and the response index among the multiple indexes is determined.

在不考虑起始响应指标的影响的情况下，变化后的响应指标

f＝E(x)为可控指标和静态指标到响应指标的映射关系。When determining the mapping relationship between controllable indicators and response indicators, static indicators may not be considered, and the mapping relationship between controllable indicators and response indicators may be determined, or static indicators may be considered to determine the mapping of static indicators and controllable indicators to response indicators. relation. For example, the set of indicators included in the characteristics of the predicted sample is X, the static indicator is represented by x _s , the controllable indicator is represented by x _c , and the response indicator is represented by x _d , X=(x _s ,x _c ,x _d ), then A mapping of x _c to x _d can be constructed, and a mapping of {x _c ,x _s } to x _d can also be constructed. Taking the construction of the mapping from {x _c , x _s } to x _d as an example, assuming that Δx _c is the change value of the controllable index that satisfies the constraints, considering the influence of the initial response index, the changed response index

Without considering the impact of the initial response index, the changed response index

f=E(x) is the mapping relationship between controllable indexes and static indexes to response indexes.

When constructing the mapping from x _c to x _d , a static mapping of x _c +Δx _c →x _d +Δx _d can be constructed, and a dynamic mapping of Δx _c →Δx _d can also be constructed. Correspondingly, when constructing {x _c ,x When mapping _s } to x _d , a static mapping of {x _c +Δx _c ,x _s }→x _d +Δx _d can be constructed, or a dynamic mapping of {Δx _c , x _s }→Δx _d can be constructed. It should be noted that under static mapping or dynamic mapping, the influence of the initial response index may or may not be considered according to the actual situation.

The construction method of the mapping may include, but is not limited to, a machine learning model, a statistical model, and artificially formulated rules, and the specific construction method may be set according to actual needs. For example, in the case where the mapping relationship between the controllable index and the response index can be determined by prior knowledge, a rule that can characterize the mapping relationship between the controllable index and the response index can be formulated artificially.

As an example, the mapping relationship between the controllable index and the response index can be determined in any one of the following two ways: way 1, according to the multiple indexes included in the characteristics of the samples in the sample set, construct a A model for predicting the value of the response index based on the value of the control index to determine the mapping relationship between the controllable index and the response index. The value of the controllable index and the value of the controllable index are used to predict the value of the response index, so as to determine the mapping relationship between the controllable index and the response index. Among them, the constructed model can be a statistical model or a machine learning model, and the sample set can be the training sample set used when training the prediction model, or the sample set obtained based on other methods; Starting from the prediction problem targeted by the model, the mapping relationship between controllable indicators and response indicators among multiple indicators can be determined according to prior knowledge. way to determine the mapping relationship between controllable indicators and response indicators.

It should be noted that the mapping relationship between the controllable index and the response index mentioned in the present invention may include various relationships such as one-to-one, many-to-one, one-to-many, and many-to-many. When constructing a model to determine the mapping relationship between controllable indicators and response indicators, traditional statistical models are generally one-to-one or many-to-one, but the mapping relationship is relatively free in newer machine learning models, such as deep neural networks. In the network, the number of inputs and outputs are arbitrarily controllable, and the mapping relationship can be automatically adjusted by the model. The specific form of the mapping relationship is not limited in the present invention.

In step S130, based on the mapping relationship, the controllable index of the input sample is adjusted, so that the output result obtained by the operation mechanism operating on the adjusted input sample satisfies the predetermined requirement.

As an example, the indicators included in the input sample (for example, the prediction sample) can be divided into static indicators (if any), controllable indicators, and response indicators. A static indicator is a set of indicators whose values cannot be changed, and a response indicator is difficult A collection of metrics that are adjusted but can vary based on changes in controllable metrics. After the value of the controllable indicator changes, the value of the response indicator can be determined based on the mapping relationship. Therefore, when it is necessary to give appropriate suggestions for the indicators contained in the input samples, so that the changes in the indicators will lead to the expected operation output, only the controllable indicators can be adjusted, so that the adjusted input samples (by the new controllable output) can be adjusted. The control index, the new response index and the original static index (if any) are formed) and the output result obtained after the input operation mechanism can meet the predetermined requirements.

It should be noted that there may be some linked indicators among the controllable indicators. For example, "weight", "blood lipids" and "blood sugar" all have linkages. When "weight" increases, "blood lipids" and "blood sugar" generally It should also be increased. When increasing the weight, if blood lipids and blood sugar remain unchanged, or blood lipids and blood sugar decrease, it is obviously contrary to the real situation. therefore. In order to make the adjustment of the controllable indicators more accurate, the controllable indicators can also be grouped, so that the indicators with a linkage relationship among the controllable indicators are divided into the same group. When adjusting the controllable indicators, the controllable indicators can be Within the value range of , the controllable indicators are adjusted on a group-by-group basis based on the linkage relationship between the indicators in the same group. Adjusting the controllable indicators on a group basis means that, for the indicators in the same group, when any one of the indicators is adjusted, other indicators in the group should be adjusted at the same time. Among them, when adjusting the value of one of the indicators, the adjustment direction of the values of other indicators in the group can be determined according to the actual linkage relationship. For example, the linkage relationship between indicator A and indicator B in the same group is the same When both rise and fall, when the value of index A is increased, the value of index B should also be increased.

The linkage relationship mentioned in the present invention mainly includes a correlation relationship and/or a causal relationship. The indicators with a causal relationship among the controllable indicators may be divided into the same group, and/or the indicators with a correlation among the controllable indicators may also be divided into the same group.

The indicators with correlation among the controllable indicators can be determined by calculation, so as to divide the indicators with correlation among the controllable indicators into the same group. As an example, the correlation can be understood as the values of the two indicators have a fixed change relationship, such as a positive correlation change relationship (that is, the value rises and falls simultaneously), or a negative correlation change relationship (that is, the value increases once a drop). For example, the correlation between the indicators in the controllable indicators can be calculated by means of independence verification. The calculation method of the independence check may include, but is not limited to, the Pearson correlation coefficient, the Spearman correlation coefficient, the Kendall harmony coefficient, the t test, the chi-square test, and the like. Among them, when calculating the correlation between the indicators, the calculation can be performed based on the value of the indicator, or it can be calculated according to the change value of the indicator, that is, the correlation between the indicators in the controllable indicators can be calculated in two ways: static and dynamic. , so as to obtain the static and dynamic correlation between the indicators. The specific calculation process will not be repeated here.

The causal relationship mainly means that logically, the value change of one indicator will lead to the value change of another (or several other) indicators. The determination of causal relationship is closer to logical reasoning, and its changes are generally more complicated and difficult to determine by calculation. Therefore, the indicators with causal relationship among controllable indicators can be determined according to external input, so as to divide the indicators with causal relationship among controllable indicators. to the same group. For example, the indicators with causal relationship among the controllable indicators can be determined by artificial induction according to prior knowledge.

As an example, the indicators with strong correlation (for example, the calculated correlation degree is greater than a predetermined threshold) among the controllable indicators can be divided into the same group by means of calculation (such as independence check). Since there is no particularly good calculation method for the causal relationship, Therefore, the indicators with causal relationship among the control indicators can be identified by means of external input (such as artificial induction), and the indicators with causal relationship can be divided into the same group to make up for the deficiency of the calculation method.

Taking the computing mechanism as the prediction model as an example, when solving the adjustment results of the controllable indicators, the prediction results obtained by the prediction model for the adjusted prediction samples can meet the predetermined requirements and minimize the adjustment cost of the controllable indicators as the goal. , to adjust the controllable indicators among the characteristics of the predicted sample. The adjustment cost of the controllable index can be predetermined, for example, the adjustment cost of each index in the controllable index can be represented by a specific expression.

For example, after dividing the multiple indicators included in the characteristics of the predicted sample into static indicators, controllable indicators and response indicators, and obtaining the mapping relationship between the static indicators and the controllable indicators to the response indicators, based on the prediction The problem provides a prediction model for predicting the results. On the premise of satisfying the constraints related to the controllable indicators, the problem of solving the adjustment results of the controllable indicators can be transformed into the following form:

Among them, ΔX is the feature change of the predicted sample, that is, the change value of multiple indicators included in the feature, and C is the set of all possible values for which the feature change ΔX satisfies the constraint conditions, including the controllable indicators and the response indicator changes. Value range, and grouping constraints of controllable indicators determined based on linkage relationship.

Taking into account the cost (ΔX) of the change of controllable indicators, the above formula can finally be transformed into:

Among them, w ₁ and w ₂ are constants, which are used to weigh the final control effect and control cost, and the values of w ₁ and w ₂ can be set according to specific scenarios. The min in the above formula is only to indicate the optimization direction, not necessarily the minimum. For example, when w ₁ is positive, it means that the optimization direction is the smallest, and when w ₁ is negative, it means that the optimization direction is large. The above equation can be solved in a variety of computational ways. Among them, the specific solution method can be determined according to the final solution difficulty, such as linear programming, gradient descent, simplex method, if the function is more complex, you can use heuristic algorithm (such as genetic algorithm, simulated annealing, ant colony algorithm, etc.) , the solution process of the formula will not be repeated here.

Here, it should be noted that the above expression for solving the problem is only an example. For example, in the case of considering the influence of the initial response index, the above-mentioned changed response index needs to be replaced by

In addition, according to different requirements that need to be met, the above formula can also have certain deformations. For example, the above formula can not only be used as a solution formula for characterizing optimization to the limit value, but also can be deformed to transform the above formula The formula is converted into a solution formula used to characterize optimization to a specific value, such as a solution formula that can be converted into a minimum value of the difference between the true value and the specific value. The specific deformation form of the above formula will not be repeated.

In step S140, the adjustment result of the controllable index is provided for index intervention.

In this way, only the adjustment result of the controllable index can be provided, and the relevant parties can intervene in the index according to the adjustment result of the controllable index, so as to promote the desired operation output. Compared with not distinguishing the indicators of the input samples and directly using the optimization method to obtain the indicator control scheme that can promote the desired output, the adjustment results based on the controllable indicators provided by the present invention are more reasonable. It can be achieved and easy to achieve when the adjustment results of the index intervention are carried out.

For example, for a model that predicts the probability of disease, if the model predicts that the probability of the user's disease is high according to the user's existing indicators, the present invention can be used to give a suggestion for adjusting the controllable indicators in the user's indicators. Adjusting the controllable indicators according to the recommendations given can reduce the probability of disease. For another example, for a model for predicting machine wear and tear, the present invention can be used to give suggestions for adjusting the controllable indexes in the existing indexes, and relevant personnel can adjust the controllable indexes according to the given suggestions, so as to reduce machine loss and prolong the machine life. life.

FIG. 2 shows a flow chart of a method for intervening on a user's physical index to reduce the user's risk of disease according to an exemplary embodiment of the present invention. The method shown in FIG. 2 can be implemented by a computer program, or can also be implemented by a special device for performing index intervention.

Referring to Fig. 2, in step S210, a prediction model is obtained. The prediction model obtained here refers to the disease probability prediction model. The disease probability prediction model is used to predict the user's disease probability according to the user's characteristics. Optionally, the disease probability prediction model may be trained by a machine learning algorithm. . Among them, the pre-trained disease probability prediction model may be obtained from the outside, and the disease probability prediction model may also be trained online. The training process of the disease probability prediction model will not be repeated here.

In step S220, the disease probability of the user is predicted according to the characteristics of the user by using the disease probability prediction model.

The characteristics of the user include multiple indicators related to predicting the probability of disease. These multiple indicators mainly refer to the physical indicators related to the user, such as but not limited to age, gender, height, weight, one or more lifestyle indicators. Related characteristics (such as work and rest time, diet, exercise), blood pressure, blood sugar, uric acid, blood lipids and other indicators.

In step S230, it is determined whether the predicted disease probability of the user is higher than a preset threshold.

When the predicted probability of the user's disease is lower than the preset threshold, it can be considered that the current index of the user is normal, and no intervention is required. When the predicted probability of the user's disease is higher than the preset threshold, it can be considered that the user's current index is abnormal, and an appropriate index adjustment plan needs to be provided for the user, so that the user can adjust his index according to the given adjustment plan, and can Reduce the chance of getting sick.

As shown in FIG. 2 , when the predicted probability of the user's disease is higher than the preset threshold, steps S240 and S250 may be performed to provide the user with an adjustment result.

In step S240, based on the mapping relationship between the static index among the multiple indexes and the controllable index to the response index, adjust the controllable index among the characteristics of the user, so that the prediction model is adjusted according to the characteristics of the user after adjustment. The predicted disease probability is lower than a predetermined threshold.

For the concepts and division methods of static indicators, controllable indicators, and response indicators, please refer to the relevant description above, and will not be repeated here. As an example, the static indicators may include at least one of the following: age, gender, height, and occupation; the controllable indicators may include at least one of the following: weight, one or more lifestyle-related characteristics (such as daily life, diet, exercise) , blood pressure, blood sugar; response indicators may include at least one of the following: uric acid, blood lipids.

The mapping relationship between static indicators and controllable indicators to response indicators can be received from the outside, and the mapping relationship between static indicators and controllable indicators among multiple indicators to response indicators can also be determined.

在不考虑起始响应指标的影响的情况下，变化后的响应指标

f＝E(x)为可控指标和静态指标到响应指标的映射关系。其中，在构造{x_c,x_s}到x_d的映射时，可以构造{x_c+Δx_c,x_s}→x_d+Δx_d的静态映射，也可以构造{Δx_c，x_s}→Δx_d的动态映射。应注意的是，在静态映射或者动态映射下，都可根据实际情况，考虑或不考虑起始响应指标的影响。Taking the set of indicators included in the characteristics of the predicted sample as X, the static indicators as x _s , the controllable indicators as x _c , and the response indicators as x _d , X=(x _s , x _c , x _d ) as an example, A mapping of {x _c ,x _s } to x _d can be constructed. For example, assuming that Δx _c is the change value of the controllable index that satisfies the constraints, considering the influence of the initial response index, the changed response index

Without considering the impact of the initial response index, the changed response index

f=E(x) is the mapping relationship between controllable indexes and static indexes to response indexes. Among them, when constructing the mapping from {x _c , x _s } to x _d , a static mapping of {x _c +Δx _c ,x _s }→x _d +Δx _d can be constructed, or {Δx _c , x _s } → Dynamic mapping of Δx _d . It should be noted that under static mapping or dynamic mapping, the influence of the initial response index may or may not be considered according to the actual situation.

The construction method of the mapping may include, but is not limited to, a machine learning model, a statistical model, and artificially formulated rules, and the specific construction method may be set according to actual needs. For example, in the case where the mapping relationship between the controllable index and the response index can be determined by prior knowledge, a rule that can characterize the mapping relationship between the controllable index and the response index can be formulated artificially.

As an example, the mapping relationship between the controllable index and the response index can be determined in any one of the following two ways: way 1, according to the multiple indexes included in the characteristics of the samples in the sample set, construct a static The value of the indicator and the value of the controllable indicator are used to predict the value of the response indicator, so as to determine the mapping relationship between the controllable indicator and the response indicator. Among them, the constructed model can be a statistical model or a machine learning model, and the sample set can be the training sample set used in training the prediction model, or the sample set obtained based on other methods; According to the experimental knowledge, the mapping relationship between static indicators and controllable indicators to response indicators among multiple indicators is determined. For example, the mapping relationship between static indicators and controllable indicators to response indicators can be determined by constructing statistical models or formulating rules according to existing prior knowledge (such as relevant knowledge and common sense in the medical field).

In order to make the adjustment of the controllable indicators more in line with the real situation, the controllable indicators can also be grouped, so that the indicators with a linkage relationship among the controllable indicators are divided into the same group. Within the value range of the controllable indicators, the controllable indicators are adjusted on a group-by-group basis based on the linkage relationship between the indicators in the same group. For an implementation manner of dividing the indicators with a linkage relationship among the controllable indicators into the same group, reference may be made to the relevant description above, and details are not repeated here.

As an example, when finding the adjustment result of the controllable index, the disease probability predicted by the disease probability prediction model for the adjusted user sample can be lower than the predetermined threshold and the adjustment cost of the controllable index can be minimized as the goal. , to adjust the controllable indicators among the characteristics of the predicted sample. The adjustment cost of the controllable index may be predetermined, for example, the adjustment cost of each index in the controllable index may be represented based on a determined expression.

In step S250, the adjustment result of the controllable index is provided so that the user can perform index intervention.

When performing index intervention, users can only intervene on controllable indicators such as body weight, one or more lifestyle-related characteristics, blood pressure, blood sugar and other controllable indicators according to the adjustment results of the provided controllable indicators, so as to reduce the risk of disease. It does not need to consider indicators that are difficult for users to adjust, such as uric acid and blood lipids.

The method for performing index intervention of the present invention can also be implemented as a device for performing index intervention. FIG. 3 shows a block diagram of an apparatus for performing index intervention according to an exemplary embodiment of the present invention. The functional units of the apparatus for performing index intervention may be implemented by hardware, software, or a combination of hardware and software that implements the principles of the present invention. Those skilled in the art can understand that the functional units described in FIG. 3 can be combined or divided into sub-units, so as to realize the principle of the above invention. Accordingly, the description herein may support any possible combination, or division, or further definition of the functional units described herein.

The following briefly describes the functional units that the apparatus for performing index intervention can have and the operations that each functional unit can perform. For the details involved, reference may be made to the above related descriptions, which will not be repeated here.

Referring to FIG. 3 , the apparatus 300 for performing index intervention includes an acquiring unit 310 , a determining unit 320 , an adjusting unit 330 and a providing unit 340 .

The obtaining unit 310 is configured to obtain an operation mechanism for providing an output result for an input sample, wherein the input sample includes a plurality of indicators. The acquiring unit 310 may receive the operation mechanism from the outside, or may establish the operation mechanism by itself. Regarding the operation mechanism (eg, prediction model), indicators, etc., reference may be made to the description of step S110 above, and details are not repeated here.

The determining unit 320 is configured to determine the mapping relationship between the controllable index and the response index among the multiple indexes. Among them, the controllable index is the index set whose value can be adjusted, and the response index is the index set whose value itself is difficult to adjust but can be changed according to the change of the controllable index.

For the determination of the mapping relationship, reference may be made to the description of step S120 above, and details are not repeated here. As an example of the present invention, the determining unit 320 may construct a model for predicting the value of the response indicator based on the value of the controllable indicator according to multiple indicators included in the characteristics of the samples in the sample set, so as to determine the controllable indicator and the response indicator. The mapping relationship between the indicators, or the determining unit 320 can also construct a model for predicting the value of the response indicator based on the value of the static indicator and the value of the controllable indicator according to multiple indicators included in the characteristics of the samples in the sample set , to determine the mapping relationship between the controllable index and the response index, wherein the static index is a set of indexes whose value cannot be changed. In addition, the determining unit 320 may also determine the mapping relationship between the controllable index and the response index according to prior knowledge.

The adjustment unit 330 is configured to adjust the controllable index of the input sample based on the mapping relationship, so that the output result obtained by the operation mechanism operating on the adjusted input sample satisfies the predetermined requirement.

For the process of adjusting the controllable index, reference may be made to the description of step S130 above, which will not be repeated here. As an example of the present invention, the adjustment unit 330 may target the prediction results obtained by the prediction model for the adjusted prediction samples to meet the predetermined requirements and reduce the adjustment cost of the controllable indicators as much as possible. Adjustable indicators.

Optionally, the apparatus 300 for performing index intervention may further include a grouping unit (not shown in the figure), and the grouping unit is used for grouping the controllable indexes, so as to divide the indexes with a linkage relationship among the controllable indexes into the same group. group, the adjustment unit 330 may adjust the controllable index in the unit of group based on the linkage relationship between the indexes in the same group within the value range of the controllable index. For the implementation process of grouping the controllable indicators by the grouping unit, reference may be made to the relevant description above, and details are not repeated here.

The providing unit 340 is configured to provide the adjustment result of the controllable index for index intervention.

It should be understood that the specific implementation manner of the apparatus 300 for performing index intervention according to an exemplary embodiment of the present invention may be implemented with reference to the relevant specific implementation manner described in conjunction with FIG. 1 , which will not be repeated here.

FIG. 4 shows a block diagram of an apparatus for performing index intervention according to another exemplary embodiment of the present invention. The functional units of the apparatus for performing index intervention may be implemented by hardware, software, or a combination of hardware and software that implements the principles of the present invention. Those skilled in the art can understand that the functional units described in FIG. 4 can be combined or divided into sub-units, so as to realize the principle of the above invention. Accordingly, the description herein may support any possible combination, or division, or further definition of the functional units described herein.

The following briefly describes the functional units that the apparatus for performing index intervention can have and the operations that each functional unit can perform. For the details involved, reference may be made to the above related descriptions, which will not be repeated here.

Referring to FIG. 4 , the apparatus 400 for performing index intervention includes an acquiring unit 410 , a predicting unit 420 , an adjusting unit 430 and a providing unit 440 .

The obtaining unit 410 is configured to obtain a disease probability prediction model. For details, reference may be made to the description of step S210 above, which will not be repeated here.

The predicting unit 420 is configured to use the disease probability prediction model to predict the user's disease probability according to the user's characteristics. The characteristics of the user include multiple indicators related to predicting the probability of disease, such as but not limited to age, gender, height, weight, lifestyle, blood pressure, blood sugar, uric acid, blood lipids and other indicators.

The adjustment unit 430 is configured to adjust the controllable among the characteristics of the user based on the static index among the multiple indexes and the mapping relationship between the controllable index and the response index when the predicted probability of the user's disease is higher than the predetermined threshold. The indicator is adjusted so that the disease probability obtained by the disease probability prediction model based on the adjusted user characteristics is lower than the predetermined threshold.

For the concepts and division methods of static indicators, controllable indicators, and response indicators, please refer to the relevant description above, and will not be repeated here. As an example, the static index may include at least one of the following: age, gender, height, and occupation; the controllable index may include at least one of the following: weight, lifestyle, blood pressure, and blood sugar; the response index may include at least one of the following: uric acid, blood lipids.

For the process of adjusting the controllable index, reference may be made to the description of step S240 above, which will not be repeated here. As an example of the present invention, the adjustment unit 430 may use the disease probability predicted by the disease probability prediction model according to the characteristics of the adjusted user to be lower than the predetermined threshold and the adjustment cost of minimizing the controllable index as The target is to adjust the controllable indicators among the characteristics of the predicted sample.

The providing unit 440 is configured to provide the adjustment result of the controllable index so that the user can perform index intervention.

Optionally, the apparatus 400 for performing index intervention may further include a determining unit (not shown in the figure). The determining unit is used to determine the mapping relationship between the static index and the controllable index among the multiple indexes to the response index. For the construction method of the mapping, reference may be made to the above related description, which will not be repeated here.

It should be understood that the specific implementation manner of the apparatus 400 for performing index intervention according to an exemplary embodiment of the present invention may be implemented with reference to the relevant specific implementation manner described in conjunction with FIG. 2 , which will not be repeated here.

FIG. 5 shows a schematic structural diagram of a computing device that can be used to implement the above method according to an exemplary embodiment of the present invention.

Referring to FIG. 5 , computing device 500 includes memory 510 and processor 520 .

The processor 520 may be a multi-core processor, or may include multiple processors. In some embodiments, processor 520 may comprise a general-purpose main processor and one or more special-purpose co-processors, such as a graphics processing unit (GPU), a digital signal processor (DSP), and the like. In some embodiments, the processor 520 may be implemented using customized circuits, such as Application Specific Integrated Circuits (ASICs) or Field Programmable Gate Arrays (FPGAs).

Memory 510 may include various types of storage units, such as system memory, read only memory (ROM), and persistent storage. The ROM may store static data or instructions required by the processor 520 or other modules of the computer. Persistent storage devices may be readable and writable storage devices. Permanent storage may be a non-volatile storage device that does not lose stored instructions and data even if the computer is powered off. In some embodiments, persistent storage devices employ mass storage devices (eg, magnetic or optical disks, flash memory) as persistent storage devices. In other embodiments, persistent storage may be a removable storage device (eg, a floppy disk, an optical drive). System memory can be a readable and writable storage device or a volatile readable and writable storage device, such as dynamic random access memory. System memory can store some or all of the instructions and data that the processor needs at runtime. In addition, memory 510 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read only memory), and magnetic and/or optical disks may also be employed. In some embodiments, memory 510 may include a removable storage device that is readable and/or writable, such as a compact disc (CD), a read-only digital versatile disc (eg, DVD-ROM, dual-layer DVD-ROM), Read-only Blu-ray Discs, Ultra-Density Discs, Flash Cards (eg SD Cards, Min SD Cards, Micro-SD Cards, etc.), Magnetic Floppy Disks, etc. Computer readable storage media do not contain carrier waves and transient electronic signals transmitted over wireless or wire.

Executable code is stored on the memory 510, and when the executable code is executed by the processor 520, the processor 520 can be made to execute the above-mentioned method for performing index intervention.

The method, apparatus and computing device for performing index intervention according to the present invention have been described in detail above with reference to the accompanying drawings.

Furthermore, the method according to the invention can also be implemented as a computer program or computer program product comprising computer program code instructions for carrying out the above-mentioned steps defined in the above-mentioned method of the invention.

Alternatively, the present invention can also be implemented as a non-transitory machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) having executable codes (or computer programs, or computer instruction codes stored thereon) ), when the executable code (or computer program, or computer instruction code) is executed by the processor of the electronic device (or computing device, server, etc.), the processor is caused to perform the various steps of the above-mentioned method according to the present invention .

Those skilled in the art will also appreciate that the various exemplary logical blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein may be implemented as electronic hardware, computer software, or combinations of both.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems and methods in accordance with various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more functions for implementing the specified logical function(s) executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented in dedicated hardware-based systems that perform the specified functions or operations , or can be implemented in a combination of dedicated hardware and computer instructions.

Various embodiments of the present invention have been described above, and the foregoing descriptions are exemplary, not exhaustive, and not limiting of the disclosed embodiments. Numerous modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the various embodiments, the practical application or improvement over the technology in the marketplace, or to enable others of ordinary skill in the art to understand the various embodiments disclosed herein.

Claims (10)

1. A method for performing index intervention, comprising:

obtaining an arithmetic mechanism for providing an output result for an input sample, wherein the input sample comprises a plurality of metrics;

determining a mapping relation between a controllable index and a response index in the plurality of indexes, wherein the controllable index is an index set of which the value can be adjusted, and the response index is an index set of which the value is difficult to adjust but can change according to the change of the controllable index;

based on the mapping relation, the controllable indexes of the input samples are adjusted, so that the output result obtained by the operation of the operation mechanism aiming at the adjusted input samples meets the preset requirement; and

and providing the adjustment result of the controllable index for index intervention.

2. The method of claim 1, wherein the operational mechanism comprises at least one of: machine learning model, rule set, function operation.

3. The method of claim 1, wherein the computing mechanism is a predictive model that provides a prediction result for a predictive problem, and wherein the input samples are predictive samples of the predictive model, the characteristics of which include a plurality of metrics related to the predictive problem.

4. The method of claim 3, wherein determining a mapping between a controllable metric and a responsive metric among the plurality of metrics comprises:

according to a plurality of indexes included in the characteristics of the samples in the sample set, constructing a model for predicting the value of the response index based on the value of the controllable index so as to determine the mapping relation between the controllable index and the response index; or

And according to a plurality of indexes included in the characteristics of the samples in the sample set, constructing a model for predicting the value of the response index based on the value of a static index and the value of the controllable index so as to determine the mapping relation between the controllable index and the response index, wherein the static index is an index set of which the value is unchangeable.

5. The method of claim 3, wherein determining a mapping between a controllable metric and a responsive metric among the plurality of metrics comprises:

and determining a mapping relation between a controllable index and a response index in the plurality of indexes according to the prior knowledge.

6. A method for performing index intervention, comprising:

acquiring a disease probability prediction model;

predicting the prevalence probability of a user according to the characteristics of the user by using the prevalence probability prediction model, wherein the characteristics of the user comprise a plurality of indexes related to the predicted prevalence probability;

under the condition that the prevalence probability of the user is predicted to be higher than a preset threshold, adjusting a controllable index in the characteristics of the user based on a mapping relation between a static index and a controllable index to a response index in the multiple indexes, so that the prevalence probability predicted by the prevalence probability prediction model for the adjusted characteristics of the user is lower than the preset threshold, wherein the static index is an index set of which the value is unchangeable, the controllable index is an index set of which the value is adjustable, and the response index is an index set of which the value is difficult to adjust and can change according to the change of the controllable index; and

and providing an adjustment result of the controllable index so as to facilitate index intervention of a user.

7. An apparatus for performing index intervention, comprising:

an obtaining unit configured to obtain an operation mechanism for providing an output result for an input sample, wherein the input sample includes a plurality of indexes;

the device comprises a determining unit, a calculating unit and a processing unit, wherein the determining unit is used for determining a mapping relation between a controllable index and a response index in the plurality of indexes, the controllable index is an index set of which the value can be adjusted, and the response index is an index set of which the value is difficult to adjust but can change according to the change of the controllable index;

the adjusting unit is used for adjusting the controllable indexes of the input samples based on the mapping relation so that the output result obtained by the operation of the operation mechanism aiming at the adjusted input samples meets the preset requirement; and

and the providing unit is used for providing the adjustment result of the controllable index for index intervention.

8. An apparatus for performing index intervention, comprising:

an acquisition unit for acquiring a disease probability prediction model;

a prediction unit for predicting the prevalence probability of a user according to a feature of the user using the prevalence probability prediction model, wherein the feature of the user includes a plurality of indexes related to the predicted prevalence probability;

an adjusting unit, configured to, when it is predicted that the prevalence probability of the user is higher than a predetermined threshold, adjust a controllable index among the features of the user based on a mapping relationship between a static index and a controllable index to a response index among the multiple indexes, so that the prevalence probability predicted by the prevalence probability prediction model for the adjusted features of the user is lower than the predetermined threshold, where the static index is an index set whose value is unchangeable, the controllable index is an index set whose value is adjustable, and the response index is an index set whose value is difficult to adjust but is changeable according to a change of the controllable index; and

and the providing unit is used for providing the adjustment result of the controllable index so that the user can perform index intervention.

9. A computing device, comprising:

a processor; and

a memory having executable code stored thereon, which when executed by the processor, causes the processor to perform the method of any of claims 1 to 5.

10. A non-transitory machine-readable storage medium having stored thereon executable code, which when executed by a processor of an electronic device, causes the processor to perform the method of any of claims 1-5.

Images