CN119149037B - An intelligent decision-making system combining algorithms and business rules - Google Patents

Abstract

The present invention relates to an intelligent decision-making system combining algorithms and business rules, which is applied to the technical field of online car-hailing platforms, including: managing diversified data assets and algorithm models through a unified platform, thereby avoiding the problem of difficulty in tracing historical legacy problems caused by personnel mobility due to the lack of a unified tool in traditional methods; splitting policy tasks into multiple functional parts, each functional part is responsible for editing corresponding content, and the visual arrangement of intelligent decision-making rules and complex business decision-making processes is realized through the cooperation of various functional components, so that non-technical personnel can also participate in the rule formulation and optimization process, greatly improving development efficiency and reducing dependence on program developers; generating test forms through automatic mapping of rule data, and instantly verifying the effects of business rules, which can quickly respond to changes in market demand and improve overall efficiency.

Description

Intelligent decision system combining algorithm and business rule

Technical Field

The invention relates to the technical field of network taxi-taking platforms, in particular to an intelligent decision system combining an algorithm and business rules.

Background

With the rapid development of the internet and mobile communication technology, the internet about car industry is rapidly increased, the internet about car platform provides convenient and efficient travel service by integrating passenger demands and driver resources, however, with the expansion of service scale and the diversification of service scenes, the service logic of the internet about car industry becomes more and more complex, and in order to improve the operation efficiency and user experience, each large platform is introduced into an algorithm model in a dispute, and intelligent decision is realized through data driving.

In the prior art, the data definition and the business rule input by the algorithm model under different business scenes are different, so that each application scene needs to be subjected to customized packaging, the customized development is time-consuming and labor-consuming, the conventional business rule management is biased to programming due to lack of a unified management tool, the rapid-change market demand is difficult to adapt, in addition, the history carry-over problem is difficult to trace back due to frequent personnel flow, and the difficulty of system maintenance and optimization is further increased.

Currently, service rule landing and optimization seriously depend on program development, each personalized requirement needs to undergo a series of processes from requirement analysis, carding and review to arrangement, development, testing, online and the like, the traditional software development mode leads to long service production period, the service rule is changeable and flexible in combination, and more repetitive work exists in the program landing, so that technical resources are seriously and ineffectively occupied, and the overall efficiency is reduced.

At present, most network taxi platforms adopt a customized development mode to meet the demands under different service scenes, while the method can provide a solution for specific problems, the method has the defects of long development period, high cost and difficulty in adapting to rapidly-changing market environments, and some platforms realize partial automation operation by writing scripts, however, the method still needs a great amount of manpower investment to carry out script maintenance and updating, and meanwhile, a unified management tool is also lacking, so that the complexity of the whole system is increased, and part of platforms try to modularize the system functions so as to reuse the modules under different scenes, but the method has limited effect in practical application due to the problems of non-uniform interface standards among the modules, inconsistent data formats and the like.

Disclosure of Invention

In view of the above, the present invention aims to provide an intelligent decision system for combining an algorithm and a business rule, so as to solve the problems of high dependency on program developers, difficult traceability of history carry-over problems, long landing period of the business rule, low resource reuse rate in the process of developing the business rule, and more repetitive work in the existing method.

According to a first aspect of an embodiment of the present invention, there is provided an intelligent decision system for combining an algorithm with business rules, the system comprising:

The feature management module is used for generating data features according to the appointed service field or generating model features according to labels corresponding to the model services built in the platform;

The interface management module is used for acquiring the characteristic variables of the characteristic management module and configuring variable values of the corresponding characteristic variables, and is also used for acquiring the output interface variables of other interfaces and configuring the variable values of the corresponding interface variables;

The method is also used for outputting characteristic variables containing variable values, interface variables and custom variables as interface characteristics;

The rule management module is used for acquiring the data characteristics, the model characteristics and the interface characteristics of the characteristic management module and the interface management module and classifying the data characteristics, the model characteristics and the interface characteristics into character string type characteristics, numerical value type characteristics and model type characteristics according to data types;

The method comprises the steps of carrying out a character string type feature according to a character string logic operation word, carrying out specific condition judgment on the character string type feature according to the character string logic operation word to generate a character string type feature condition, carrying out specific condition judgment on the numerical value type feature according to a numerical value logic operation word to generate a numerical value type feature condition, and carrying out condition judgment on the model type feature according to a model logic operation word to generate a model type feature condition;

the method is also used for combining the selected character string type characteristic conditions, numerical value type characteristic conditions or model type characteristic conditions, and setting different rule expressions into a list to obtain a rule list by setting output types and output values;

the flow management module is used for selecting one or more rule expressions from the rule list, executing the rule expressions according to a specific sequence in a connecting mode, and configuring an ending condition to obtain a strategy task;

and the batch task testing module is used for respectively testing the strategy tasks in the task list by uploading the data set in the appointed format, issuing the strategy tasks which are successfully executed, and deleting or reconfiguring the strategy tasks which are failed to be executed.

Preferably, the method comprises the steps of,

The attributes of the data features comprise a data feature ID, a name, a feature type, a service field of a large data warehouse associated with the data features, a name of an associated service field, a numerical value type of the associated service field and a data feature tag;

the attributes of the model features comprise model feature IDs, names, feature types, model services, mapping tags and model feature tags;

the data characteristic label or the model characteristic label is used for inquiring the data characteristic or the model characteristic;

The feature management module is also used for freely editing all attribute fields of the data features or model features which are not referenced by the rule expressions, freely editing the data feature IDs and names and data feature labels for the data features which are referenced by any rule expression, and freely editing the model feature IDs and names and model feature labels for the model features which are referenced by any rule expression;

The feature management module is also used for deleting the data features or model features which are not referenced by the rule expression on line;

the feature management module is also used for querying any data feature or model feature and acquiring attribute fields and all rule expressions being referenced.

Preferably, the method comprises the steps of,

The variables acquired by the interface management module comprise variable names, numerical value types, variable values, example values and variable descriptions;

The variable value type comprises a characteristic variable, an interface variable and a custom variable;

the example value is used for a default variable value when the interface management module performs interface debugging;

When the interface management module debugs the interface, returning debugging parameters, judging whether the necessary filling items of the variables are filled or not and whether the same interface exists or not and whether the request address is available or not according to the debugging parameters;

The interface management module is also used for editing and modifying interfaces which are not referenced by any rule expression;

The interface management module is further configured to delete interfaces that are not referenced by any of the rule expressions.

Preferably, the method comprises the steps of,

The rule expression comprises a rule name, a rule label and a rule description;

The rule management module further comprises a rule configuration plate which is used for performing configuration editing on the rule expression;

The rule configuration plate is used for generating character string type characteristic conditions by selecting character string type characteristics and judging specific conditions of the character string type characteristics according to the character string logic operation words, wherein the character string logic operation words comprise equal, unequal, containing, not containing, matching regular expressions, not matching regular expressions, beginning, ending, being empty and not being empty;

the rule configuration plate is used for selecting a numerical value type characteristic, and judging a specific condition of the numerical value type characteristic according to the numerical value logic operation words to generate a numerical value type characteristic condition, wherein the numerical value logic operation words comprise equal to, unequal to, greater than, less than, greater than or equal to, less than or equal to, null and not null;

the rule configuration plate is used for generating model type characteristic conditions by selecting model type characteristics and carrying out condition judgment on the model type characteristics according to the model logic operation words, wherein the model logic operation words comprise hit, miss, at least hit and maximum hit;

The rule configuration plate is also used for combining one or more character string type characteristic conditions, numerical value type characteristic conditions and model type characteristic conditions through connecting words, and setting an output type and an output value at the same time, wherein the connecting words comprise or;

The rule management module further comprises a rule verification plate, wherein the rule verification plate is used for verifying the generated rule expression;

The condition that the verification result of the rule expression fails comprises that the operation expression is wrong, any characteristic condition structure in the rule expression is missing, the rule expression is output in an unconfigured mode and the rule expression is output abnormally, and the rule expression which passes the verification result is stored and placed in a rule list.

Preferably, the method comprises the steps of,

The rule management module is also used for inquiring the target rule expression through any one or more of the name, the state, the rule label and the rule creation date of the rule expression;

the state of the rule expression comprises enabling and disabling, and the rule expression passing verification and being saved is defaulted to be a disabled state;

the rule management module is also used for editing rule expressions in a forbidden state, and the edited new rule expressions can be saved and added into the rule list after passing the verification of the rule verification plate;

the rule management module is also used for deleting the rule expressions which are not referenced by any policy tasks.

Preferably, the method comprises the steps of,

The policy tasks comprise basic information and interface information, wherein the basic information comprises a policy ID, a name, a policy tag, a policy description, a policy state, a policy creator and a policy creation time;

for the strategy task in the released state, the flow management module only allows to check the basic information and interface information of the strategy task and does not allow editing and modification;

The interface information comprises interface addresses of all rule expressions referenced by the policy task;

The flow management module is used for editing, modifying and deleting strategy tasks in an unpublished state.

Preferably, the method comprises the steps of,

The strategy tasks comprise a start node, a rule node and an end node;

the method comprises the steps of setting a start node as a flow start and a finish node as a flow finish, wherein the finish node comprises one or more rule nodes as flow intermediate steps, the start node is connected with one rule node, the rule nodes can be simultaneously connected with a plurality of rule nodes, and one rule node is connected with one or more finish nodes;

And when the output values are configured for both the satisfied and unsatisfied branches in the rule expression corresponding to any rule node, the rule node allows two connecting lines to be initiated and is connected with different post-arranged rule nodes or end nodes, otherwise, the rule node only supports the initiation of one connecting line.

Preferably, the method comprises the steps of,

The batch task testing module is used for testing policy tasks which are successfully configured and stored;

When the batch task testing module performs batch testing on the strategy task by acquiring the data set with the specified format, checking the data format in the uploaded data set, and if the data format contains all the data source fields required by the rule expression contained in the strategy task, performing testing if the data source fields pass the checking;

If the test result fails, the batch task test module displays which characteristic condition in which rule expression corresponds to the data execution failure;

The batch task testing module is also used for finishing the process of the strategy task which is being tested, deleting the stored result data from the execution queue and deleting the task record;

the batch task testing module is also used for deleting the strategy task waiting for executing the test from the waiting queue and deleting the task record;

The batch task test module is also used for deleting the stored result data and deleting the task record for the strategy task which fails to execute the test;

the batch task test module is also used for deleting the stored result data and deleting the task record for the task which is successfully tested.

The technical scheme provided by the embodiment of the invention can comprise the following beneficial effects:

the method manages diversified data assets and algorithm models through the unified platform, so that the problem that historical carry-over problems caused by personnel flow due to lack of unified tools are difficult to trace in the traditional method is avoided, a strategy task is split into a plurality of functional parts, each functional part is responsible for editing corresponding content, and intelligent decision rules and visual arrangement of complex business decision processes are realized through cooperation of functional components, so that non-technicians can participate in rule making and optimizing processes, development efficiency is greatly improved, dependence on program developers is reduced, test forms are generated through automatic mapping rule data, the business rule effect is verified immediately, market demand change can be responded rapidly, and overall efficiency is improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.

FIG. 1 is a system diagram illustrating an intelligent decision system combining algorithms and business rules according to an exemplary embodiment;

FIG. 2 is a schematic diagram of data features shown according to another exemplary embodiment;

FIG. 3 is a schematic diagram of model features shown according to another exemplary embodiment;

FIG. 4 is a schematic diagram of variables shown according to another exemplary embodiment;

FIG. 5 is a diagram illustrating interface debug return parameters according to another exemplary embodiment;

FIG. 6 is a schematic diagram of a rule configuration shown according to another exemplary embodiment;

FIG. 7 is a diagram illustrating string type feature conditions according to another exemplary embodiment;

FIG. 8 is a diagram illustrating numeric type feature conditions according to another exemplary embodiment;

FIG. 9 is a model-type feature condition diagram shown in accordance with another exemplary embodiment;

FIG. 10 is a rule verification schematic shown in accordance with another exemplary embodiment;

FIG. 11 is a schematic diagram of a strategic task shown in accordance with another exemplary embodiment;

In the drawing, a 1-feature management module, a 2-interface management module, a 3-rule management module, a 4-flow management module and a 5-batch task test module are shown.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the invention. Rather, they are merely examples of apparatus and methods consistent with aspects of the invention as detailed in the accompanying claims.

Example 1

FIG. 1 is a system diagram of an intelligent decision system for combining algorithms and business rules, as shown in FIG. 1, according to an exemplary embodiment, comprising:

the feature management module 1 is used for generating data features according to the appointed service field or generating model features according to labels corresponding to model services built in the platform;

The interface management module 2 is used for acquiring the characteristic variable of the characteristic management module 1 and configuring the variable value of the corresponding characteristic variable, and is also used for acquiring the output interface variable of other interfaces and configuring the variable value of the corresponding interface variable;

The method is also used for outputting characteristic variables containing variable values, interface variables and custom variables as interface characteristics;

The rule management module 3 is used for acquiring the data characteristics, the model characteristics and the interface characteristics of the characteristic management module 1 and the interface management module 2, and classifying the data characteristics, the model characteristics and the interface characteristics into character string type characteristics, numerical value type characteristics and model type characteristics according to data types;

The method comprises the steps of carrying out a character string type feature according to a character string logic operation word, carrying out specific condition judgment on the character string type feature according to the character string logic operation word to generate a character string type feature condition, carrying out specific condition judgment on the numerical value type feature according to a numerical value logic operation word to generate a numerical value type feature condition, and carrying out condition judgment on the model type feature according to a model logic operation word to generate a model type feature condition;

the method is also used for combining the selected character string type characteristic conditions, numerical value type characteristic conditions or model type characteristic conditions, and setting different rule expressions into a list to obtain a rule list by setting output types and output values;

The flow management module 4 is used for selecting one or more rule expressions from the rule list, executing the rule expressions according to a specific sequence in a connecting mode, and configuring an ending condition to obtain a strategy task;

The batch task testing module 5 is used for respectively testing the strategy tasks in the task list by uploading the data set with the specified format, issuing the strategy tasks which are successfully executed, and deleting or reconfiguring the strategy tasks which are failed to be executed;

It can be understood that, as shown in fig. 2, when selecting a new data feature in the feature management module 1, a dialog box is popped up, and a mapping field, a custom mapping name, a specified field type are required to be set, and feature tags are defined according to requirements;

The mapping field is a service field mapped by the current newly-built feature, and the option list is automatically associated with a large data warehouse service field list;

Mapping names, namely corresponding to selected service fields, customizing English names with readability, and limiting English or English and underline combination with the length of 64 characters;

the field type is the numerical value type of the selected service field and comprises the types of character strings, numerical values, boolean and the like;

Feature labels, namely, unnecessary items are arranged, and the feature labels are marked according to the needs and are used for classifying query features;

for a network booking platform, the data features may be driver number, dispatch distance, whether to talk, induce cancellation of orders, etc.

As shown in fig. 3, when selecting newly built model features in the feature management module 1, selecting model services built in a platform, selecting mapped model tags, defining feature tags according to requirements, wherein a tag list supporting automatic identification by each model service is fixed, a tag selection list of the mapped tags is automatically associated with a tag list supporting identification by the selected model service, and the model features can be actively asked good, smoking in a car, call recording and the like.

The feature management module 1 supports quick query of target features by feature names, feature types, feature labels, and feature creation dates, clicks a "query" button, queries target features according to set query conditions, clicks a "reset" button, and resets the query conditions to restore default values.

Inputting a feature name in an input box, inquiring the feature of a designated name, and supporting fuzzy search of the feature name by inputting an inquiry keyword;

supporting selecting the characteristics of the designated type of the query, and defaulting all types of characteristics of the query;

Querying by creation date, supporting querying features created within a specified date range by selecting the date range;

And clicking a drop-down button to select the feature tags from all feature tag lists, inquiring the features of the appointed tags, supporting the one-time selection of a plurality of feature tags, and supporting the fuzzy inquiry of the feature tags for selection through a search box.

Unreferenced data features, all fields support online editing;

Unreferenced model features, all fields support online editing;

the cited data features and model features only support editing feature names and feature labels, and other field information cannot be edited;

Deleting the referenced feature, wherein the referenced feature does not allow deletion, clicking a 'delete' button, popping up a delete prompt dialog box, prompting which rule sets are being referenced, and failing to delete;

Deleting unreferenced features, namely clicking a 'delete' button, popping up a delete confirmation dialog box, confirming the currently selected feature to be deleted, confirming the deletion, or else canceling the deletion;

When an interface is edited in the interface management module 2, the interface management module comprises an interface name, a request mode, namely a GET/POST and an interface request URL address, as shown in figure 4, the interface request parameters comprise a variable name, a numerical value type (String or numerical Number), wherein the variable value type comprises a characteristic variable and is derived from the characteristic management module 1; the interface variable is derived from the output of other interfaces, the variable value is automatically mapped out of a parameter path such as object1.object11.param112 for the interface variable, the variable value is automatically assigned for the self-defined variable, and the variable value is generally expressed by 1 or 0 for the characteristic variable, wherein 1 represents that the condition is met or met, and 0 represents that the condition is not met or met;

for the characteristic variable and the interface variable, supporting configuration example values, and using the configuration example values as default variable values when the interface is debugged;

after the editing is completed, clicking a save button to save the interface request parameters newly added/edited currently; clicking the delete button deletes the current request parameter.

Clicking a debugging button in the interface management module 2, and if the interface parameter example value meets the interface calling requirement, displaying an interface debugging result in a debugging result frame, automatically generating json format return parameters in a return parameter frame, as shown in figure 5, adding a new interface, wherein the debugging result and the return parameter frame are empty;

The return parameter provides an editing button, defaults to a non-editing state, clicks the editing button, and the return parameter text box enters an editable state to support the return parameter description of the fully selected, copied and edited json format;

for the return parameters, clicking the confirm button, checking the format of the interface return parameters, and checking the content includes:

whether json format is correct or not, code, name, type for each return parameter is not allowed to be null-!

The return parameter format requires that descriptions of field names, types, values be provided in json format, with some return parameters exemplified as follows:

{ "type": "object", "properties": { "code": { "type": "integer" }, "message": { "type": "string" }, "data": { "type": "object", "properties": { "is_support_city": { "type": "integer", "title": " Whether to support the allied city or not and the description is 1, wherein 0 is no, 2 is unknown, 3 is that no address is extracted, the lower two fields have values when returning to 1, "" extract_addr "{" type ": string", "title": extracted address information "," description ": preferentially shows the market, for example, beijing sea lake area can be allied with, beijing" };

Checking interface data:

whether the necessary filling item is filled or not comprises an interface name, a request mode, a request address and a return parameter;

is there the same interface name;

is the request address available? displaying interface verification prompt information under the address;

interface data verification prompts exception type:

The filling items are not filled, and the red fonts are used for prompting which item is not filled under the corresponding filling items;

The request address is inaccessible;

The newly built interface names are the same;

clicking an editing button, opening an interface editing dialog box, and modifying basic information, request parameters and return parameters of the interface, judging whether the interface is being referenced by rules, and judging which rules are being referenced by a referenced popup prompt and not allowing modification;

Clicking a delete button, clicking a popup box to prompt whether to confirm deletion, deleting the interface after confirmation, judging whether the interface is being referenced by rules, and judging which rules are being referenced by the referenced popup box to not allow deletion.

The new rule popup window in the rule management module 3 consists of three panels, namely basic information, rule configuration and rule verification, wherein the basic information comprises basic information such as names, labels and descriptions of rules, the names of the rules are filling items, and a plurality of labels are spaced by "#";

By adding 'and', 'or', configuring a logic judgment expression of a designated field type, and setting the output type and content, configuring a decision output value of a rule, the version supports the output of a character string and a numerical value, as shown in fig. 6;

Clicking an add row button "+", popping up a function menu list, wherein a single row of conditions only support adding ' same row and/or ' conditional expressions ', and when adding rows, the same level only supports adding the same ' and/or ' conditional rows of the relationship, and the upper and lower levels only support mutually exclusive ' and/or ' conditional rows of the relationship;

as shown in fig. 7, the character string type feature condition line:

selecting character string type feature configuration condition lines, dynamically triggering a character string logic operation control group, and supporting equality, inequality, inclusion, non-inclusion, matching regular expressions, non-matching regular expressions, beginning, ending, being empty and not being empty;

Similar to the enumerated type features, when the logical operation is selected to be empty/not empty, automatically deleting the later threshold control;

equal/unequal judging whether the selected characteristic value is equal to the input threshold content;

judging whether the selected characteristic value contains the input threshold content or not;

Judging whether the selected characteristic value is matched or not matched with the input regular expression;

starting with judging whether the selected characteristic value starts with input content;

it is determined whether the selected feature value ends with the input content.

As shown in fig. 8, the numerical type feature condition line:

selecting a numerical type feature configuration condition row, dynamically triggering a numerical logic operation control group, and supporting equal to, unequal to, greater than, less than, greater than or equal to, less than or equal to, empty and not empty;

The method is empty/not empty, similar to the enumeration type feature and the character string type feature, when the logical operation is selected to be empty/not empty, the latter threshold control is automatically deleted;

When a numerical condition is added, a drop-down option control is automatically changed into an input box to support the editing of a complex operation expression of a numerical feature, meanwhile, a selected numerical type feature is automatically changed into a blue font starting with an @ symbol to be expressed as a reference feature element, in order to easily distinguish the reference feature from other contents of the expression, an @ feature name is displayed as a blue highlight, and the other contents are normally black, such as "(@ dispatch distance (blue) - @ car-contact distance (blue))) 1.5 (black) +5 (black)", when the reference feature is needed, the @ symbol is input, a numerical type feature list which is already managed in the system is dynamically pulled, the content of the automatic filling expression is selected to support the starting of a character string content by inputting the feature name, and the associated feature name is fuzzy search.

As shown in fig. 9, model type feature condition line:

And selecting a model type feature configuration condition row, and dynamically triggering a numerical logic operation control group. Support four operations, hit, miss, at least hit, and at most hit;

selecting at least hit and most hit, configuring hit times, defaulting to 1 time;

the hit and miss times need not be configured when selecting hits.

Setting decision output values of the rule under a rule expression configuration frame, supporting character strings and numerical values, setting at least one output value of the two types of rule output values, defaulting to a character string type, and checking whether the input content type is a numerical value type or not when the numerical value type is selected;

When the test data is input, the verification button is clicked to automatically verify that the expression format is correct and pass, and the error is not passed, and the corresponding output value of the set test data is verified and output;

The expression format verification error types comprise four types, namely (1) a numerical operation expression error, (2) a conditional row structure defect comprising two types of defect, namely only a characteristic option box and uncompensated condition judgment threshold content, (3) a decision result is not configured, namely at least one decision result needs to be configured and cannot be set, and (4) the decision result is abnormal in configuration, namely a numerical type is selected, non-numerical content is configured, and two outputs are simultaneously set and are required to be kept consistent in output type.

When rule verification passes the storage, the latest verification data configuration and verification results are stored at the same time;

after the rule expression verification is passed, directly storing the new rule, wherein the expression verification is not passed, and the bullet box prompts that the rule expression verification is not passed and the rule expression can not be stored- "

Clicking the 'rule management' navigation button, entering a rule list page, and rapidly looking up rule information such as rule ID, name, rule label, description, enabling state, creator, creation time and the like;

support and inquire the goal rule fast through rule name, state, rule label, and rule creation date, "inquire" button, according to inquiring the condition set up, inquire the goal rule, "reset" button, reset inquire the condition and resume default value;

The rule name is input in an input box, the rule of the appointed name is inquired, and fuzzy search of the rule name by inputting the inquiry keyword is supported;

the rule state query is that rules supporting the selection of the specified state comprise all, forbidden and enabled, and all are queried by default;

querying by creating a date, supporting querying rules created within a specified date range by selecting the date range;

clicking a drop-down button, selecting rule labels from all rule label lists, inquiring rules of specified labels, supporting one-time selection of a plurality of rule labels, supporting selection by fuzzy inquiry of the rule labels through a search box, and designing the control with the same characteristic labels;

the rule which passes verification and is stored is defaulted to be in a disabled state, the operation bar provides an enable button, the enable button is clicked, and the operation button is changed to be the disable button;

the referenced rule is disabled by clicking a disable button to determine if the rule is being referenced by the policy flow is present, then the popup prompts that the current rule is not allowed to be disabled;

the rule state is changed to be disabled, the operation button is changed to be an enabling button, and the state and the operation button value are directly switched;

Enabling the rule of the state, only allowing to view details of the rule, not allowing to edit and modify rule information, enabling the popup window to be named as 'viewing rule', enabling the bottom to have no cancel and save buttons, enabling rule content in the popup window not to be edited, and only supporting viewing;

The rule of the state is forbidden, the editing rule information is allowed to be edited, the popup window name is an editing rule, and after the content is modified, the verification expression can be saved after passing;

clicking a delete button, clicking a confirm button in a confirm box, judging whether the rule is being referenced by a policy flow is present, and prompting that the current rule does not allow deletion;

clicking a delete button, popup prompting to confirm deletion, clicking a 'confirm' button, and deleting the rule;

clicking a copy button to directly copy all configuration contents of the rule, creating a new rule, and after modifying the contents, verifying that the expression passes and can be stored;

Clicking a new strategy flow button in the flow management module 4, popping up a new popup window from the right side, setting the name and the label of the new strategy flow and describing basic information content;

and the target policy flow is quickly inquired through the policy flow name, the state, the label and the policy flow creation date.

Clicking a 'query' button, querying a target strategy flow according to a set query condition, clicking a 'reset' button, and resetting the query condition to restore a default value;

The strategy process name is input in an input box, the strategy process with the designated name is inquired, and the strategy process name is supported to be searched in a fuzzy way by inputting the inquiry keywords;

The strategy flow supporting the selection of the specified state comprises all, published and unpublished, and the default inquiry is all;

Inquiring through the creation date, namely supporting inquiring the policy flow created in the appointed date range by selecting the date range;

And clicking a drop-down button to select labels from all strategy flow label lists, inquiring the strategy flow of the designated labels, supporting the one-time selection of a plurality of strategy flow labels, supporting the selection of fuzzy inquiry strategy flow labels through a search box, and designing the labels with the same characteristics by a control.

The published strategy flow only allows the strategy flow information to be checked, the strategy flow information is not allowed to be edited and modified, and the popup window name is 'checking strategy flow';

The bottom of the popup window is provided with a closing button, the popup window is closed, and the upper right corner of the popup window is provided with a closing button, so that the popup window is closed;

the strategy flow in the unpublished state allows editing and modifying information, and the popup window name is 'editing strategy flow';

The released state strategy flow 'delete' button is ash-placed and is not operable;

the method comprises the steps of not issuing a state strategy flow, clicking a 'delete' button, opening a delete confirmation prompt popup window, clicking a 'confirm' button, deleting a selected strategy flow, clicking a 'cancel' button, and canceling a deletion operation;

Clicking an undelivered policy flow, clicking an opened policy flow configuration page, and configuring a policy flow:

Starting node:

the starting node is existing by default and does not support deletion;

Each policy flow only supports setting one start node.

Clicking a 'save' button to save modified configuration content of the starting node, clicking a 'cancel' button to cancel modification operation and reserving the content before modification;

Selecting a rule node component on the left side of the canvas, dragging to any position of the canvas, and configuring a strategy flow through connecting lines;

the starting node can only be connected with one rule node;

starting the connection between the node and the rule node, supporting selection and deletion, and not supporting editing;

Starting the node to connect with another rule node, and deleting the connection between the node and the other rule node;

Configuring rule node information (supporting to connect other rule nodes or ending nodes after the rule nodes are configured and stored successfully, otherwise prompting that the rule nodes at the beginning of connecting are not configured when the connecting operation is performed);

The rule node names are dynamically associated with the bound rule names, and can be edited and renamed in a self-defined mode;

End node:

selecting an end node component on the left side of the canvas, dragging the end node component to any position of the canvas, and configuring strategy flow output through connecting rule nodes and end nodes;

and (3) connecting:

The method comprises the steps of enabling a node middle area of a connection to be initiated by clicking, keeping a clicking state, dragging a mouse to a target node middle area, establishing the connection of two nodes, clicking the connection, selecting the connection, inputting delete key information, deleting the selected connection, clicking the connection, selecting the connection, editing the selected connection by double clicking (only the connection initiated by a regular node supports double clicking editing), and obtaining a strategy task after the connection as shown in figure 11.

Association rule node, default rule name of initiating node binding showing association;

the rule output is selected, namely the option of default showing that the association rule is configured with the output value is selected, and if the configuration output value is met or not met, the output is selected by default;

setting branch names, namely setting connection line names displayed by a strategy flow, and dynamically associating output values configured by selected rule nodes by default, wherein the output values can be edited and renamed in a self-defined mode;

connection rule:

The starting node only supports connection with the rule node and only supports connection with one rule node, and when the starting node connects with the rule node, the rule node is allowed to be unconfigured;

the rule nodes support a connection rule node and an end node, the rule nodes initiated by the connection need to be configured and stored successfully, otherwise, the popup window prompts that the rule nodes started by the connection are not configured;

When the rule node is connected, if the bound rule is satisfied and the output values are configured for the two branches, the rule node allows two connecting lines to be initiated and is connected with different post nodes, otherwise, only one connecting line is supported to be initiated;

The end node is not allowed to initiate a connection.

Canvas operation:

The strategy flow nodes and the associated connection lines can support dragging to any position;

after the strategy flow nodes and the connecting lines are selected, deleting the selected nodes or connecting lines by inputting a delete key;

deleting nodes, and synchronously deleting the associated connection lines;

Selecting a plurality of nodes and connecting lines in the support area, and deleting the nodes and the connecting lines by inputting a delete key;

the policy flow supports moving to any position of the canvas;

policy flows support scaling.

Saving configuration:

clicking a 'save' button, and checking whether the strategy flow of configuration is compliant when configuration is saved;

When saving, it is necessary to check whether the current flow has a path from the start node to the end node, at least one path exists, and the saving is allowed.

The published policy flows do not allow configuration/modification of the flow processes and only support viewing;

when the batch test is carried out on the strategy tasks through the batch task test module 5:

no successful policy flow is configured:

after new construction, no strategy flow which is configured and successfully stored is available, the test button is gray, and the operation is unavailable ''

And (3) configuring a successful strategy flow:

After the new construction, configuring and saving a successful strategy flow, and supporting test of both unpublished and issued states, clicking a test button, opening a test window, and automatically generating a field form of a data source to be input according to the configured strategy flow;

Inputting test data, clicking a verification button, starting a test, clicking a refreshing button, refreshing a test process, checking a test result, or closing a popup window, and after the test is finished, clicking the testing button again to check the latest test result of the strategy flow;

The method comprises the steps of creating a new strategy flow, defaulting to be unpublished, setting ash on a release button, not operating the strategy flow which successfully stores configuration information, clicking the release button, switching the button to a release function button, changing the flow state to the released state, and synchronously updating interface information of the strategy flow;

The state of the issued strategy flow is issued, if the strategy flow is edited and configured, the off-shelf button is clicked, the strategy flow is changed into the non-issued state, and basic information editing and strategy flow configuration are carried out;

clicking a new task button in the batch task testing module 5, and configuring new batch testing task information by using a popup window;

task names, namely editing batch test task names;

a strategy flow is selected for batch test;

The method comprises the steps of clicking a 'download import file template', downloading a batch test import file template corresponding to a selected strategy flow, uploading files in a specified format, clicking a 'confirm' button, creating batch test tasks, checking the uploaded batch test data format according to the selected strategy flow when clicking the 'confirm' button, and checking whether all required data source fields of the flow are contained or not;

Clicking the navigation button of the task center to enter a batch test task list page, and rapidly checking information such as task ID, name, strategy flow name, state, task failure prompt information, creator, creation time and the like, prompting failure of execution, prompting failure reasons, prompting success of execution and prompting failure/error of execution of data lines;

And clicking a query button to query the target task according to the set query condition, clicking a reset button to reset the query condition and recovering the default value.

The task name is input in an input box, batch test tasks with specified names are inquired, and fuzzy search of the task names through input of inquiry keywords is supported;

The task of selecting the appointed state comprises all tasks, executing, failing to execute, successful execution, waiting to execute and defaulting to all tasks;

Setting the size of a simultaneous execution queue according to the configuration of system resources, wherein a configuration file is configurable;

the newly created task is completed, determine if the execution queue is idle?

The idle task directly enters an execution queue, and the idle task is in an executing state;

directly joining a waiting queue in a state of waiting for execution if the waiting queue is not idle;

the execution queue is idle, task execution is selected from the waiting queue, and the task state is changed;

by creating date inquiry, supporting inquiry of tasks created in a specified date range by selecting the date range;

The strategy process name is input in an input box, batch test tasks of the designated strategy are inquired, and fuzzy searching of the strategy name by inputting inquiry keywords is supported;

and executing the successful tasks, and exporting batch task results through a download button.

Clicking a deletion button to delete the appointed task;

the executing task ends the process, deletes the stored result data from the execution queue and deletes the task record;

the task waiting for execution is deleted from the waiting queue, and the task record is deleted;

Executing the failed task, deleting the saved result data, and deleting the task record;

And executing the successful task, deleting the stored result data, and deleting the task record.

It is to be understood that the same or similar parts in the above embodiments may be referred to each other, and that in some embodiments, the same or similar parts in other embodiments may be referred to.

It should be noted that in the description of the present invention, the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Furthermore, in the description of the present invention, unless otherwise indicated, the meaning of "plurality" means at least two.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and further implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.

It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of techniques known in the art, discrete logic circuits with logic gates for implementing logic functions on data signals, application specific integrated circuits with appropriate combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.

Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, and where the program, when executed, includes one or a combination of the steps of the method embodiments.

In addition, each functional unit in the embodiments of the present invention may be integrated in one processing module, or each unit may exist alone physically, or two or more units may be integrated in one module. The integrated modules may be implemented in hardware or in software functional modules. The integrated modules may also be stored in a computer readable storage medium if implemented in the form of software functional modules and sold or used as a stand-alone product.

The above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, or the like.

In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.

Claims (7)

1. An intelligent decision making system combining algorithms with business rules, said system comprising:

The feature management module is used for generating data features according to the appointed service field or generating model features according to labels corresponding to the model services built in the platform;

the attributes of the data features comprise a data feature ID, a name, a feature type, a service field of a large data warehouse associated with the data features, a name of an associated service field, a numerical value type of the associated service field and a data feature tag;

the attributes of the model features comprise model feature IDs, names, feature types, model services, mapping tags and model feature tags;

the data characteristic label or the model characteristic label is used for inquiring the data characteristic or the model characteristic;

the feature management module is also used for freely editing all attribute fields of the data features or model features which are not referenced by the regular expression, freely editing the data feature ID and the name and the data feature tag for the data features which are referenced by any regular expression, and freely editing the model feature ID and the name and the model feature tag for the model features which are referenced by any regular expression;

The feature management module is also used for deleting the data features or model features which are not referenced by the rule expression on line;

The feature management module is also used for inquiring any data feature or model feature and acquiring attribute fields and all the rule expressions being referenced;

The interface management module is used for acquiring the characteristic variables of the characteristic management module and configuring variable values of the corresponding characteristic variables, and is also used for acquiring the output interface variables of other interfaces and configuring the variable values of the corresponding interface variables;

The method is also used for outputting characteristic variables containing variable values, interface variables and custom variables as interface characteristics;

The rule management module is used for acquiring the data characteristics, the model characteristics and the interface characteristics of the characteristic management module and the interface management module and classifying the data characteristics, the model characteristics and the interface characteristics into character string type characteristics, numerical value type characteristics and model type characteristics according to data types;

The method comprises the steps of carrying out a character string type feature according to a character string logic operation word, carrying out specific condition judgment on the character string type feature according to the character string logic operation word to generate a character string type feature condition, carrying out specific condition judgment on the numerical value type feature according to a numerical value logic operation word to generate a numerical value type feature condition, and carrying out condition judgment on the model type feature according to a model logic operation word to generate a model type feature condition;

the method is also used for combining the selected character string type characteristic conditions, numerical value type characteristic conditions or model type characteristic conditions, and setting different rule expressions into a list to obtain a rule list by setting output types and output values;

the flow management module is used for selecting one or more rule expressions from the rule list, executing the rule expressions according to a specific sequence in a connecting mode, and configuring an ending condition to obtain a strategy task;

and the batch task testing module is used for respectively testing the strategy tasks in the task list by uploading the data set in the appointed format, issuing the strategy tasks which are successfully executed, and deleting or reconfiguring the strategy tasks which are failed to be executed.

2. The system of claim 1, wherein the system further comprises a controller configured to control the controller,

The variables acquired by the interface management module comprise variable names, numerical value types, variable values, example values and variable descriptions;

The variable value type comprises a characteristic variable, an interface variable and a custom variable;

the example value is used for a default variable value when the interface management module performs interface debugging;

When the interface management module debugs the interface, returning debugging parameters, judging whether the necessary filling items of the variables are filled or not and whether the same interface exists or not and whether the request address is available or not according to the debugging parameters;

The interface management module is also used for editing and modifying interfaces which are not referenced by any rule expression;

The interface management module is further configured to delete interfaces that are not referenced by any of the rule expressions.

3. The system of claim 2, wherein the system further comprises a controller configured to control the controller,

The rule expression comprises a rule name, a rule label and a rule description;

The rule management module further comprises a rule configuration plate which is used for performing configuration editing on the rule expression;

The rule configuration plate is used for generating character string type characteristic conditions by selecting character string type characteristics and judging specific conditions of the character string type characteristics according to the character string logic operation words, wherein the character string logic operation words comprise equal, unequal, containing, not containing, matching regular expressions, not matching regular expressions, beginning, ending, being empty and not being empty;

the rule configuration plate is used for selecting a numerical value type characteristic, and judging a specific condition of the numerical value type characteristic according to the numerical value logic operation words to generate a numerical value type characteristic condition, wherein the numerical value logic operation words comprise equal to, unequal to, greater than, less than, greater than or equal to, less than or equal to, null and not null;

the rule configuration plate is used for generating model type characteristic conditions by selecting model type characteristics and carrying out condition judgment on the model type characteristics according to the model logic operation words, wherein the model logic operation words comprise hit, miss, at least hit and maximum hit;

The rule configuration plate is also used for combining one or more character string type characteristic conditions, numerical value type characteristic conditions and model type characteristic conditions through connecting words, and setting an output type and an output value at the same time, wherein the connecting words comprise or;

The rule management module further comprises a rule verification plate, wherein the rule verification plate is used for verifying the generated rule expression;

The condition that the verification result of the rule expression fails comprises that the operation expression is wrong, any characteristic condition structure in the rule expression is missing, the rule expression is output in an unconfigured mode and the rule expression is output abnormally, and the rule expression which passes the verification result is stored and placed in a rule list.

4. The system of claim 3, wherein the system further comprises a controller configured to control the controller,

The rule management module is also used for inquiring the target rule expression through any one or more of the name, the state, the rule label and the rule creation date of the rule expression;

the state of the rule expression comprises enabling and disabling, and the rule expression passing verification and being saved is defaulted to be a disabled state;

the rule management module is also used for editing rule expressions in a forbidden state, and the edited new rule expressions can be saved and added into the rule list after passing the verification of the rule verification plate;

the rule management module is also used for deleting the rule expressions which are not referenced by any policy tasks.

5. The system of claim 4, wherein the system further comprises a controller configured to control the controller,

The policy tasks comprise basic information and interface information, wherein the basic information comprises a policy ID, a name, a policy tag, a policy description, a policy state, a policy creator and a policy creation time;

for the strategy task in the released state, the flow management module only allows to check the basic information and interface information of the strategy task and does not allow editing and modification;

The interface information comprises interface addresses of all rule expressions referenced by the policy task;

The flow management module is used for editing, modifying and deleting strategy tasks in an unpublished state.

6. The system of claim 5, wherein the system further comprises a controller configured to control the controller,

The strategy tasks comprise a start node, a rule node and an end node;

the method comprises the steps of setting a start node as a flow start and a finish node as a flow finish, wherein the finish node comprises one or more rule nodes as flow intermediate steps, the start node is connected with one rule node, the rule nodes can be simultaneously connected with a plurality of rule nodes, and one rule node is connected with one or more finish nodes;

And when the output values are configured for both the satisfied and unsatisfied branches in the rule expression corresponding to any rule node, the rule node allows two connecting lines to be initiated and is connected with different post-arranged rule nodes or end nodes, otherwise, the rule node only supports the initiation of one connecting line.

7. The system of claim 6, wherein the system further comprises a controller configured to control the controller,

The batch task testing module is used for testing policy tasks which are successfully configured and stored;

When the batch task testing module performs batch testing on the strategy task by acquiring the data set with the specified format, checking the data format in the uploaded data set, and if the data format contains all the data source fields required by the rule expression contained in the strategy task, performing testing if the data source fields pass the checking;

If the test result fails, the batch task test module displays which characteristic condition in which rule expression corresponds to the data execution failure;

The batch task testing module is also used for finishing the process of the strategy task which is being tested, deleting the stored result data from the execution queue and deleting the task record;

the batch task testing module is also used for deleting the strategy task waiting for executing the test from the waiting queue and deleting the task record;

The batch task test module is also used for deleting the stored result data and deleting the task record for the strategy task which fails to execute the test;

the batch task test module is also used for deleting the stored result data and deleting the task record for the task which is successfully tested.