An AI-based BI real-time decision-making method, system and intelligent terminal

Abstract

The present invention relates to an AI-based BI real-time decision-making method, system and intelligent terminal, and relates to the technical field of data processing. The method includes: obtaining real-time data information of enterprise operation and user input information; determining data requirement information and display requirement information according to the user input information; determining data selection information according to the data requirement information and the real-time data information; determining decision-making display information according to the display requirement information and the data selection information; retrieving user identity information based on the user input information; determining user terminal information according to the user identity information, and outputting the decision-making display information to the terminal corresponding to the user terminal information. The present invention has the effect of facilitating the user to use and make decisions on the BI system in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to an AI-based BI real-time decision-making method, system and intelligent terminal. Background Art

[0002] Data processing is the process of converting data into information and knowledge, covering aspects such as collection, storage, management, analysis, and visualization. Data processing is widely applied in fields such as intelligent transportation, healthcare, and financial risk control.

[0003] Since a vast amount of data is generated at every moment of an enterprise's operation, currently, business intelligence (BI for short) is generally used to process the data generated by operations. Business intelligence is a set of technologies and processes for collecting, managing, and analyzing organizational data to generate insights for business strategies and operations. When using a BI system to process operation data currently, generally, data is first collected from different sources such as an enterprise's internal systems (such as ERP and CRM systems), external data sources (such as market research data), and Internet data, and then these data are cleaned, formatted, and stored to ensure data quality and consistency. Users use the analysis and visualization tools in the BI system to perform multi-dimensional data analysis on the stored data, and present the analysis results in forms such as reports, dashboards, and charts, enabling users to clearly see the trends and correlations behind the data at a glance, thus facilitating users to make decisions based on the analysis results.

[0004] Since BI systems generally involve complex analysis and visualization tools currently, users may need some time to adapt to these tools, and as the BI systems are continuously updated, users also need to continuously learn and adapt to new functions, thus making it inconvenient for users to use and make decisions on the BI system in real time. Summary of the Invention

[0005] In order to facilitate users to use and make decisions on the BI system in real time, the present invention provides an AI-based BI real-time decision-making method, system and intelligent terminal.

[0006] In a first aspect, the present invention provides an AI-based BI real-time decision-making method, adopting the following technical solution:

[0007] An AI-based BI real-time decision-making method includes:

[0008] Obtain real-time data information of enterprise operations and user input information;

[0009] Determine data requirement information and display requirement information according to the user input information;

[0010] Determine data selection information based on data requirement information and real-time data information;

[0011] Determine decision display information based on display requirement information and data selection information;

[0012] Retrieve user identity information based on user input information;

[0013] Determine user terminal information based on user identity information, and output the decision display information to the terminal corresponding to the user terminal information.

[0014] Optionally, the methods for determining data requirement information and display requirement information include:

[0015] Retrieve input type information and input data information based on user input information;

[0016] Based on the correspondence between the input type information and the preset type reference data information, determine the type reference data information corresponding to the input type information;

[0017] Based on the matching result between the input data information and the type reference data information, determine the matching meaning information;

[0018] Determine the meaning requirement information according to the matching meaning information;

[0019] Retrieve meaning data type information and the number of data type values based on the meaning requirement information;

[0020] Determine type display information according to the meaning data type information and the number of data type values, and use the meaning requirement information as the data requirement information and the type display information as the display requirement information.

[0021] Optionally, the method for determining the meaning requirement information includes:

[0022] Based on the correspondence between the matching meaning information and the preset initial meaning requirement information, determine the initial meaning requirement information corresponding to the matching meaning information;

[0023] Determine the user's historical requirement information according to the initial meaning requirement information and the user's identity information;

[0024] Judge whether the user's historical requirement information is consistent with the initial meaning requirement information;

[0025] If it is, use the initial meaning requirement information as the meaning requirement information;

[0026] If it is not, analyze the deviation between the user's historical requirement information and the initial meaning requirement information and use it as the requirement deviation information;

[0027] Determine the requirement adjustment information according to the requirement deviation information;

[0028] Adjust the information based on the requirements to adjust the initial requirement information of the meaning to form the meaning requirement information.

[0029] Optionally, the method for determining the requirement adjustment information includes:

[0030] Retrieve the deviation occurrence time point and the deviation frequency value based on the requirement deviation information;

[0031] Obtain the current time point;

[0032] Calculate the difference between the deviation occurrence time point and the current time point and use it as the current time deviation value;

[0033] According to the correspondence between the deviation frequency value and the preset time deviation reference interval, determine the time deviation reference interval corresponding to the deviation frequency value;

[0034] Judge whether the current time deviation value falls within the time deviation reference interval;

[0035] If it is, then according to the correspondence between the requirement deviation information and the preset deviation adjustment information, determine the deviation adjustment information corresponding to the requirement deviation information, and use the deviation adjustment information as the requirement adjustment information;

[0036] If it is not, then calculate the difference between the current time deviation value and the time deviation reference interval and use it as the time deviation abnormal value;

[0037] According to the correspondence between the time deviation abnormal value, the deviation frequency value and the preset deviation comprehensive influence information, determine the deviation comprehensive influence information corresponding to the time deviation abnormal value and the deviation frequency value;

[0038] According to the correspondence between the deviation comprehensive influence information and the preset comprehensive influence adjustment information, determine the comprehensive influence adjustment information corresponding to the deviation comprehensive influence information, and use the comprehensive influence adjustment information as the requirement adjustment information.

[0039] Optionally, the method for determining the type display information includes:

[0040] According to the correspondence between the meaning data type information and the preset single type display type information, determine the single type display type information corresponding to the meaning data type information;

[0041] Judge whether the value of the number of data types is only one;

[0042] If it is, then use the single type display type information as the type display information;

[0043] If it is not, then select the single type display type information with the same as the same type information, and use the single type display type information other than the same type information as the difference type information;

[0044] Retrieve the same type of numerical values based on the same type of information;

[0045] According to the correspondence between the same type of information and the preset same conversion type information, determine the same conversion type information corresponding to the same type of information;

[0046] Based on the consistency between the same conversion type information and the difference type information, determine the difference conversion numerical values;

[0047] Select the same type of information according to the same type of numerical values and the difference conversion numerical values to obtain the selected type information, and use the selected type information as the category display information.

[0048] Optionally, the selection method of the selected type information includes:

[0049] Sort the same type of numerical values from large to small, and use the sorting result as the same type sorting value;

[0050] According to the correspondence between the same type sorting value and the preset same type sorting influence value, determine the same type sorting influence value corresponding to the same type sorting value;

[0051] Calculate the ratio value between the difference conversion numerical value and the data category numerical value as the conversion ratio value;

[0052] Sort the conversion ratio values from large to small, and use the sorting result as the conversion ratio sorting value;

[0053] According to the correspondence between the conversion ratio sorting value and the preset conversion ratio influence value, determine the conversion ratio influence value corresponding to the conversion ratio sorting value;

[0054] Calculate the sum value between the same type sorting influence value and the conversion ratio influence value as the sorting comprehensive influence value;

[0055] Sort the sorting comprehensive influence values from large to small, and use the same type of information corresponding to the first sorted sorting comprehensive influence value as the selected type information.

[0056] Optionally, the determination method of the decision display information includes:

[0057] Retrieve the selection source information and source data information based on the data selection information;

[0058] According to the correspondence between the source data information and the preset source data type information, determine the source data type information corresponding to the source data information;

[0059] Determine the type conversion control information according to the display requirement information and the source data type information;

[0060] Convert the source data information based on the type conversion control information to form the display initial data information;

[0061] Determine the display adjusted data information according to the selected source information and the display initial data information, and use the display adjusted data information as the decision display information.

[0062] Optionally, the method for determining the display adjusted data information includes:

[0063] Retrieve the source individual value and the source type information based on the selected source information;

[0064] According to the corresponding relationship between the source type information and the preset source type adjustment information, determine the source type adjustment information corresponding to the source type information;

[0065] Judge whether the source individual value is only one;

[0066] If so, adjust the display initial data information based on the source type adjustment information to form the display adjusted data information;

[0067] If not, according to the corresponding relationship between the source individual value and the preset source number adjustment information, determine the source number adjustment information corresponding to the source individual value;

[0068] Combine the source type adjustment information and the source number adjustment information to form the comprehensive adjustment information;

[0069] Adjust the display initial data information based on the comprehensive adjustment information to form the display adjusted data information.

[0070] In a second aspect, the present invention provides an AI-based BI real-time decision-making system, adopting the following technical solution:

[0071] An AI-based BI real-time decision-making system, comprising:

[0072] An acquisition module, configured to acquire real-time data information, user input information, and the current time point;

[0073] A memory, configured to store an AI-based BI real-time decision-making method according to any one of the first aspect;

[0074] A processor, configured to load and execute the program in the memory.

[0075] In a third aspect, the present invention provides an intelligent terminal, adopting the following technical solution:

[0076] An intelligent terminal comprises a memory and a processor, wherein the memory stores a computer program and can be loaded by the processor to execute an AI-based BI real-time decision-making method as described in any one of the first aspects.

[0077] In summary, the present invention includes at least one of the following beneficial technical effects:

[0078] 1. By acquiring real-time data information and user input information, and determining data demand information and display demand information through user input information, and then determining data selection information through data demand information and real-time data information, and determining decision display information through display demand information and data selection information, the user identity information is retrieved through user input information to determine user terminal information, and the decision display information is output to the terminal corresponding to the user terminal information, so as to automatically determine user needs based on the user input information and process and display real-time data, without the need for users to learn new functions, thereby facilitating users to use the BI system in real time and make decisions;

[0079] 2. Retrieve input category information and input data information through user input information, and determine category benchmark data information through query of input category information, and then determine matching meaning information by analyzing the matching results between input data information and category benchmark data information, and then determine meaning demand information, retrieve meaning data category information and data category number value through meaning demand information to determine category display information, and use the meaning demand information as data demand information, and use the category display information as display demand information, so as to improve the accuracy of the acquired data demand information and display demand information;

[0080] 3. Determine the initial meaning demand information by matching the meaning information query, and determine the user's historical demand information through the initial meaning demand information and the user's identity information, and then judge whether the user's historical demand information is consistent with the initial meaning demand information. When they are consistent, the initial meaning demand information is used as the meaning demand information. When they are inconsistent, analyze the demand deviation information to determine the demand adjustment information. Adjust the initial meaning demand information through the demand adjustment information to form the meaning demand information, thereby improving the accuracy of the acquired meaning demand information. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 is a flow chart of a method for real-time BI decision-making based on AI according to an embodiment of the present application;

[0082] Figure 2 is a flow chart of a method for determining data demand information and display demand information in an embodiment of the present application;

[0083] Figure 3 is a flow chart of a method for determining the meaning requirement information of an embodiment of the present application;

[0084] Figure 4 It is a flowchart of a method for determining demand adjustment information according to an embodiment of the present application;

[0085] Figure 5 It is a flowchart of a method for determining category display information according to an embodiment of the present application;

[0086] Figure 6 It is a flowchart of a method for selecting selection type information according to an embodiment of the present application;

[0087] Figure 7 It is a flowchart of a method for determining decision display information according to an embodiment of the present application. Detailed implementation manners

[0088] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0089] An AI-based BI real-time decision-making method, which obtains real-time data information, user input information, and the current time point, and automatically determines user needs by the information input by the user, processes the real-time data, and displays it, without the user having to learn new functions, thereby facilitating the users to use and make decisions on the BI system in real time.

[0090] Refer to Figure 1 , an embodiment of the present invention discloses an AI-based BI real-time decision-making method, which includes:

[0091] Step S100: Obtain real-time data information of enterprise operation and user input information.

[0092] Among them, the real-time data information refers to the data information generated in real time during the operation of the enterprise, and the real-time data information is obtained by reading the data of the system or platform used during the operation of the enterprise. The user input information refers to the required information input by the user to process the data during the operation of the enterprise, and the user input information is obtained after the user inputs it.

[0093] Step S101: Determine data demand information and display demand information according to the user input information.

[0094] Among them, the data demand information refers to the data information that the user needs to use when viewing, and the display demand information refers to the type information that the user needs to use when viewing. By analyzing the natural language corresponding to the user input information, the data demand information and the display demand information are determined, which is convenient for subsequent use. The specific determination steps of the data demand information and the display demand information refer to Step S200 to Step S205.

[0095] Step S102: Determine data selection information according to the data demand information and the real-time data information.

[0096] Among them, the data selection information refers to the data information obtained by selecting the enterprise's operation data according to the user's needs. By selecting the real-time data information based on the data requirement information, the data selection information is determined, which is convenient for subsequent use.

[0097] Step S103: Determine the decision display information according to the display requirement information and the data selection information.

[0098] Among them, the decision display information refers to the data information when displaying the selected data. By adjusting the data selection information according to the display requirement information, the decision display information is determined, which is convenient for subsequent use. The specific steps for determining the decision display information refer to Step S700 to Step S704.

[0099] Step S104: Retrieve the user identity information based on the user input information.

[0100] Among them, the user identity information refers to the identity information corresponding to the user with the current input requirement. The user identity information is obtained by querying the corresponding account information, which is convenient for subsequent use.

[0101] Step S105: Determine the user terminal information according to the user identity information, and output the decision display information to the terminal corresponding to the user terminal information.

[0102] Among them, the user terminal information refers to the terminal information through which the user with the input requirement can receive information. Different user identity information corresponds to different user terminal information. The user terminal information is obtained by matching through querying the preset user identity database. The user identity database pre-stores a comparison table of different user identity information and the corresponding user terminal information, and the user identity database is obtained through pre-input. By querying the user identity information to determine the user terminal information, and outputting the decision display information to the terminal corresponding to the user terminal information, the information input by the user is automatically used to determine the user's needs, process the real-time data and display it, without the user having to learn new functions, thus facilitating the users to use and make decisions on the BI system in real time.

[0103] In Figure 1 In the shown Step S101, in order to further ensure the rationality of the data requirement information and the display requirement information, it is necessary to perform further separate analysis and calculation on the data requirement information and the display requirement information. Specifically, it is described in detail through the Figure 2 shown steps.

[0104] Referring to Figure 2 , the methods for determining the data requirement information and the display requirement information include the following steps:

[0105] Step S200: Retrieving input type information and input data information based on user input information.

[0106] The user input information includes input type information and input data information. The input type information refers to the type information of the data input by the user, and the input data information refers to the specific data information corresponding to the user when inputting. The input type information and input data information are retrieved through the user input information to facilitate subsequent use.

[0107] Step S201: determining the category reference data information corresponding to the input category information according to the correspondence between the input category information and the preset category reference data information.

[0108] Among them, the category benchmark data information refers to the benchmark data information corresponding to different data categories. Different input category information corresponds to different category benchmark data information. The category benchmark data information is obtained by querying the preset input category database and matching, which is convenient for subsequent use. The input category database pre-stores a comparison table of different input category information and the corresponding category benchmark data information and the meaning represented by the category benchmark data information. The input category database is obtained after pre-input. For example, when the input category information is a natural language, the category benchmark data information is the benchmark data containing the requirements in the natural language; when the input category information is the computer-specific C language, the category benchmark data information is the benchmark data containing the requirements in the C language.

[0109] Step S202: Determine matching meaning information based on the matching result between the input data information and the category reference data information.

[0110] Among them, matching meaning information refers to the meaning information obtained by matching after user input. By matching the input data information with the category benchmark data information, the meaning represented by the category benchmark data information consistent with the input data information is used as the matching meaning information, which is convenient for subsequent use.

[0111] Step S203: Determine the meaning requirement information according to the matching meaning information.

[0112] The meaning requirement information refers to the requirement information corresponding to the matched meaning, and the meaning requirement information is determined by analyzing the matched meaning information to facilitate subsequent use. The specific steps for determining the meaning requirement information refer to steps S300 to S306.

[0113] Step S204: Retrieve the meaning data type information and the data type value based on the meaning requirement information.

[0114] Among them, the meaning requirement information includes the meaning data type information and the data type value. The meaning data type information refers to the type information to which the corresponding requirement of the meaning belongs, and the data type value refers to the value of the number of the type to which the corresponding requirement of the meaning belongs. Retrieving the meaning data type information and the data type value through the meaning requirement information facilitates subsequent use.

[0115] Step S205: Determine the type display information according to the meaning data type information and the data type value, use the meaning requirement information as the data requirement information, and use the type display information as the display requirement information.

[0116] Among them, the type display information refers to the type information displayed according to the type to which the corresponding requirement of the meaning belongs. By analyzing the meaning data type information and the data type value, the type display information is determined, and the meaning requirement information is used as the data requirement information, and the type display information is used as the display requirement information, thereby improving the accuracy of the obtained data requirement information and display requirement information. The specific determination steps of the type display information refer to Step S500 to Step S507.

[0117] In Figure 2 In the shown Step S203, in order to further ensure the rationality of the meaning requirement information, it is necessary to perform further separate analysis and calculation on the meaning requirement information, which is specifically described in detail through the Figure 3 shown steps.

[0118] Referring to Figure 3 , the determination method of the meaning requirement information includes the following steps:

[0119] Step S300: Determine the initial meaning requirement information corresponding to the matching meaning information according to the corresponding relationship between the matching meaning information and the preset initial meaning requirement information.

[0120] Among them, the initial meaning requirement information refers to the initial requirement information corresponding to the meaning. Different matching meaning information corresponds to different initial meaning requirement information. The initial meaning requirement information is obtained by querying the preset matching meaning database and is convenient for subsequent use. The matching meaning database pre-stores a database of different matching meaning information and the corresponding initial meaning requirement information, and the matching meaning database is obtained through pre-input.

[0121] Step S301: Determine the user's historical requirement information according to the initial meaning requirement information and the user identity information.

[0122] Among them, the user's historical requirement information refers to the requirement information corresponding to the requirement included in the initial requirement information with meaning at a historical time. By querying the user identity information for the requirement corresponding to the initial requirement information with meaning in the historical time and using it as the user's historical requirement information, it is convenient for subsequent use.

[0123] Step S302: Determine whether the user's historical requirement information is consistent with the initial requirement information with meaning. If it is, execute Step S303; if not, execute Step S304.

[0124] Among them, by judging whether the user's historical requirement information is consistent with the initial requirement information with meaning, it is thus judged whether the initial requirement information with meaning needs to be further processed.

[0125] Step S303: Use the initial requirement information with meaning as the requirement information with meaning.

[0126] Among them, when the user's historical requirement information is consistent with the initial requirement information with meaning, it indicates that at this time, the initial requirement information with meaning does not need to be further processed. Therefore, the initial requirement information with meaning is used as the requirement information with meaning, thereby improving the accuracy of the obtained requirement information with meaning.

[0127] Step S304: Analyze the deviation between the user's historical requirement information and the initial requirement information with meaning and use it as the requirement deviation information.

[0128] Among them, the requirement deviation information refers to the deviation information corresponding to the case where the initial requirement corresponding to the meaning is inconsistent with the requirement at the historical time. When the user's historical requirement information is inconsistent with the initial requirement information with meaning, it indicates that at this time, the initial requirement information with meaning needs to be further processed. Therefore, by analyzing the deviation between the user's historical requirement information and the initial requirement information with meaning and using it as the requirement deviation information, it is convenient for subsequent use.

[0129] Step S305: Determine the requirement adjustment information according to the requirement deviation information.

[0130] Among them, the requirement adjustment information refers to the adjustment information for selecting and adjusting the requirement based on the deviation situation of the requirement. By analyzing the requirement deviation information, the requirement adjustment information is determined, which is convenient for subsequent use. The specific determination steps of the requirement adjustment information refer to Step S400 to Step S408.

[0131] Step S306: Adjust the initial requirement information with meaning based on the requirement adjustment information to form the requirement information with meaning.

[0132] Among them, by selecting and adjusting the requirement corresponding to the initial requirement information with meaning according to the requirement adjustment information, the requirement information with meaning is formed, thereby improving the accuracy of the obtained requirement information with meaning.

[0133] In Figure 3 in step S305 shown, in order to further ensure the rationality of the demand adjustment information, it is necessary to perform a further separate analysis and calculation on the demand adjustment information. Specifically, it is described in detail through the steps shown in Figure 4 shown steps.

[0134] Referring to Figure 4 , the method for determining the demand adjustment information includes the following steps:

[0135] Step S400: Retrieve the deviation occurrence time point and the deviation frequency value based on the demand deviation information.

[0136] Among them, the demand deviation information includes the deviation occurrence time point and the deviation frequency value. The deviation occurrence time point refers to the time point corresponding to the historical time when there is a deviation, and the deviation frequency value refers to the frequency value of the number of times when the demand has a deviation in the total demand. Retrieving the deviation occurrence time point and the deviation frequency value through the demand deviation information facilitates subsequent use.

[0137] Step S401: Obtain the current time point.

[0138] Among them, the current time point refers to the time point of the current time, and the current time point is obtained by querying the database used to time the time in real time.

[0139] Step S402: Calculate the difference between the deviation occurrence time point and the current time point and use it as the current time deviation value.

[0140] Among them, the current time deviation value refers to the time value corresponding to the time when there is a deviation between the current time and the deviation occurrence time. Calculating the difference between the deviation occurrence time point and the current time point and using it as the current time deviation value facilitates subsequent use.

[0141] Step S403: Determine the time deviation reference interval corresponding to the deviation frequency value according to the correspondence between the deviation frequency value and the preset time deviation reference interval.

[0142] Among them, the time deviation reference interval refers to the reference deviation interval that allows the time to exist under normal circumstances according to the deviation frequency value. Different deviation frequency values correspond to different time deviation reference intervals. The time deviation reference interval is obtained by querying the preset deviation frequency database for matching, so as to facilitate subsequent use. The deviation frequency database pre-stores a comparison table of different deviation frequency values and the corresponding time deviation reference intervals, and the deviation frequency database is obtained through pre-input.

[0143] Step S404: Determine whether the current time deviation value falls within the time deviation reference interval. If it is, execute step S405; if not, execute step S406.

[0144] Among them, by judging whether the current time deviation value falls within the time deviation reference interval, it is judged whether to select the demand according to the time.

[0145] Step S405: According to the correspondence between the demand deviation information and the preset deviation adjustment information, determine the deviation adjustment information corresponding to the demand deviation information, and use the deviation adjustment information as the demand adjustment information.

[0146] Among them, the deviation adjustment information refers to the adjustment information for selecting and adjusting the demand according to the deviation generated by the time. Different demand deviation information corresponds to different deviation adjustment information. When the current time deviation value falls within the time deviation reference interval, it means that the demand is selected according to the time at this time. Therefore, after querying the preset deviation adjustment database through the deviation adjustment information, the matching is obtained, and the deviation adjustment information is used as the demand adjustment information, so as to improve the accuracy of the obtained demand adjustment information. The deviation adjustment database pre-stores a comparison table of different demand deviation information and the corresponding deviation adjustment information, and the deviation adjustment database is obtained through pre-input.

[0147] Step S406: Calculate the difference between the current time deviation value and the time deviation reference interval and use it as the time deviation abnormal value.

[0148] Among them, the time deviation abnormal value refers to the deviation value corresponding to the abnormal deviation of the time. When the current time deviation value does not fall within the time deviation reference interval, it means that the demand is not directly selected according to the time at this time. Therefore, the difference between the current time deviation value and the time deviation reference interval is calculated and used as the time deviation abnormal value for subsequent use.

[0149] Step S407: According to the correspondence between the time deviation abnormal value, the deviation frequency value and the preset deviation comprehensive influence information, determine the deviation comprehensive influence information corresponding to the time deviation abnormal value and the deviation frequency value.

[0150] Among them, the deviation comprehensive influence information refers to the influence information generated according to the time abnormal deviation and the deviation frequency. The deviation comprehensive influence information is obtained by querying the preset deviation comprehensive influence database for matching, so as to facilitate subsequent use. The deviation comprehensive influence database pre-stores a comparison table of different time deviation abnormal values and deviation frequency values and the corresponding deviation comprehensive influence information, and the deviation comprehensive influence database is obtained through pre-input.

[0151] Step S408: According to the correspondence between the deviation comprehensive influence information and the preset comprehensive influence adjustment information, determine the comprehensive influence adjustment information corresponding to the deviation comprehensive influence information, and use the comprehensive influence adjustment information as the demand adjustment information.

[0152] Among them, the comprehensive impact adjustment information refers to the adjustment information for selecting and adjusting other deviations based on the impact generated by time anomaly deviation and deviation frequency. The comprehensive impact adjustment information is obtained by matching after querying a preset comprehensive impact adjustment database. The comprehensive impact adjustment database pre-stores a comparison table of different deviation comprehensive impact information and the corresponding comprehensive impact adjustment information, and the comprehensive impact adjustment database is obtained through pre-input. The comprehensive impact adjustment information is determined by querying the deviation comprehensive impact information and used as the demand adjustment information, thereby improving the accuracy of the obtained demand adjustment information.

[0153] In Figure 2 In step S205 shown, in order to further ensure the rationality of the category display information, it is necessary to perform a further separate analysis and calculation on the category display information, specifically through Figure 5 the steps shown for detailed description.

[0154] Referring to Figure 5 , the method for determining the category display information includes the following steps:

[0155] Step S500: According to the correspondence between the meaning data category information and the preset single-category display type information, determine the single-category display type information corresponding to the meaning data category information.

[0156] Among them, the single-category display type information refers to the type information corresponding to the display of a single meaning data category. Different meaning data category information corresponds to different single-category display type information. The single-category display type information is obtained by matching after querying a preset single-category display type database, which is convenient for subsequent use. The single-category display type database pre-stores a comparison table of different meaning data category information and the corresponding single-category display type information, and the single-category display type database is obtained through pre-input.

[0157] Step S501: Determine whether the number value of the data category is only one. If yes, execute step S502; if not, execute step S503.

[0158] Among them, by determining whether the number value of the data category is only one, it is judged whether the single-category display type information can be directly used.

[0159] Step S502: Use the single-category display type information as the category display information.

[0160] Among them, when the number value of the data category is only one, it means that the single-category display type information can be directly used at this time. Therefore, the single-category display type information is used as the category display information, thereby improving the accuracy of the obtained category display information.

[0161] Step S503: Select the single-category display type information that is the same as the same type information, and use the single-category display type information other than the same type information as the different type information.

[0162] Among them, the same type information refers to the type information corresponding to the same type when there is the same type in the single-category display type information. When the value of the data type count is not only one, it means that the single-category display type information cannot be directly used at this time. Therefore, the single-category display type information that is the same is selected and defined as the same type information, and the different type information is defined for convenient subsequent use.

[0163] Step S504: Retrieve the same type count value based on the same type information.

[0164] Among them, the same type count value refers to the count value corresponding to the same type when there is the same type. By counting the types corresponding to the same type information and retrieving the counted result value as the same type count value for convenient subsequent use.

[0165] Step S505: Determine the same conversion type information corresponding to the same type information according to the corresponding relationship between the same type information and the preset same conversion type information.

[0166] Among them, the same conversion type information refers to the type information corresponding to the type that can be converted to the same type. Different same type information corresponds to different same conversion type information. The same conversion type information is obtained by querying and matching the preset same conversion type database for convenient subsequent use. The same conversion type database stores a comparison table of different same type information and the corresponding same conversion type information in advance, and the same conversion type database is obtained through pre-input.

[0167] Step S506: Determine the different conversion count value based on the consistency between the same conversion type information and the different type information.

[0168] Among them, the different conversion count value refers to the count value corresponding to the different type information when it is converted to the same type. By analyzing the consistency between the same conversion type information and the different type information, counting the different type information that is consistent with the same conversion type information, and using the counting result as the different conversion count value for convenient subsequent use.

[0169] Step S507: Select the same type information according to the same type count value and the different conversion count value to obtain the selected type information, and use the selected type information as the category display information.

[0170] Among them, the selected type information refers to the type information corresponding to the selected type. By selecting the same type information based on the same type quantity value and the differential conversion quantity value, the selected type information is obtained, and the selected type information is used as the category display information to improve the accuracy of the obtained category display information. The specific determination steps of the selected type information refer to step S600 to step S606.

[0171] In Figure 5 In step S507 shown, in order to further ensure the rationality of the selected type information, it is necessary to perform a further separate analysis and calculation on the selected type information. Specifically, it is described in detail through Figure 6 the steps shown.

[0172] Referring to Figure 6 , the selection method of the selected type information includes the following steps:

[0173] Step S600: Sort the same type quantity values from large to small, and use the sorting result as the same type sorting value.

[0174] Among them, the same type sorting value refers to the result value obtained by sorting the same type quantity values. By sorting the same type quantity values from large to small and using the sorting result as the same type sorting value, it is convenient for subsequent use.

[0175] Step S601: Determine the same type sorting influence value corresponding to the same type sorting value according to the corresponding relationship between the same type sorting value and the preset same type sorting influence value.

[0176] Among them, the same type sorting influence value refers to the influence value on the selection based on the same type sorting value. Different same type sorting values correspond to different same type sorting influence values. The same type sorting influence value is obtained by querying and matching the preset same type sorting database, which is convenient for subsequent use. The same type sorting database stores in advance a comparison table of different same type sorting values and the corresponding same type sorting influence values, and the same type sorting database is obtained through pre-input.

[0177] Step S602: Calculate the ratio value between the differential conversion quantity value and the data category quantity value and use it as the conversion ratio value.

[0178] Among them, the conversion ratio value refers to the ratio value of the number that can be converted to the total number of categories. By calculating the ratio value between the differential conversion quantity value and the data category quantity value and using it as the conversion ratio value, it is convenient for subsequent use.

[0179] Step S603: Sort the conversion ratio values from large to small, and use the sorting result as the conversion ratio sorting value.

[0180] Among them, the conversion ratio sorting value refers to the result value obtained by sorting according to the conversion ratio value. By sorting the conversion ratio values from large to small and using the sorting result as the conversion ratio sorting value, it is convenient for subsequent use.

[0181] Step S604: Determine the conversion ratio influence value corresponding to the conversion ratio sorting value according to the corresponding relationship between the conversion ratio sorting value and the preset conversion ratio influence value.

[0182] Among them, the conversion ratio influence value refers to the influence value on the selection based on the conversion ratio sorting value. Different conversion ratio sorting values correspond to different conversion ratio influence values. The conversion ratio influence value is obtained by querying and matching the preset conversion ratio sorting database, which is convenient for subsequent use. The conversion ratio sorting database pre-stores a comparison table of different conversion ratio sorting values and the corresponding conversion ratio influence values, and the conversion ratio sorting database is obtained through pre-input.

[0183] Step S605: Calculate the sum value between the same type sorting influence value and the conversion ratio influence value and use it as the sorting comprehensive influence value.

[0184] Among them, the sorting comprehensive influence value refers to the comprehensive influence degree value generated by the sorting results of the same type sorting and the conversion ratio. By calculating the sum value between the same type sorting influence value and the conversion ratio influence value and using it as the sorting comprehensive influence value, it is convenient for subsequent use.

[0185] Step S606: Sort in descending order based on the sorting comprehensive influence value, and use the same type information corresponding to the sorting comprehensive influence value ranked first as the selected type information.

[0186] Among them, by sorting in descending order based on the sorting comprehensive influence value and using the same type information corresponding to the sorting comprehensive influence value ranked first as the selected type information, the accuracy of the obtained selected type information is improved.

[0187] In Figure 1 In step S103 shown, in order to further ensure the rationality of the decision display information, it is necessary to perform a further separate analysis and calculation on the decision display information, which is specifically described in detail through the Figure 7 steps shown.

[0188] Referring to Figure 7 , the method for determining the decision display information includes the following steps:

[0189] Step S700: Retrieve the selection source information and source data information based on the data selection information.

[0190] Among them, the data selection information includes the selection source information and the source data information. The selection source information refers to the source information to which the selected data belongs, and the source data information refers to the data information corresponding to the selection source information. Retrieving the selection source information and the source data information through the data selection information facilitates subsequent use. The selection source information includes sources of different personnel, sources at different times, sources of different departments, sources of different regions, etc.

[0191] Step S701: Determine the source data type information corresponding to the source data information according to the correspondence between the source data information and the preset source data type information.

[0192] Among them, the source data type information refers to the type information to which the source data belongs. Different source data information corresponds to different source data type information. The source data type information is obtained by querying and matching the preset source data type database, which facilitates subsequent use. The source data type database stores a comparison table of different source data information and the corresponding source data type information, and the source data type database is obtained through pre-input.

[0193] Step S702: Determine the type conversion control information according to the display requirement information and the source data type information.

[0194] Among them, the type conversion control information refers to the control information used to control the conversion of the type to which the source data type information belongs into the type to which the display requirement information belongs. The type conversion control information is obtained by querying the preset type conversion database through the display requirement information and the source data type information, which facilitates subsequent use. The type conversion database pre-stores a comparison table of different display requirement information, source data type information, and the corresponding type conversion control information, and the type conversion database is obtained through pre-input.

[0195] Step S703: Convert the source data information based on the type conversion control information to form the display initial data information.

[0196] Among them, the display initial data information refers to the data information obtained after converting the data. The source data information is converted through the type conversion control information to form the display initial data information, which facilitates subsequent use.

[0197] Step S704: Determine the display adjustment data information according to the selection source information and the display initial data information, and use the display adjustment data information as the decision display information.

[0198] Among them, the displayed adjusted data information refers to the data information obtained after adjusting the displayed data. By analyzing the selected source information and the initial displayed data information, the displayed adjusted data information is determined, and the displayed adjusted data information is used as the decision display information to improve the accuracy of the obtained decision display information. The specific determination steps of the displayed adjusted data information refer to step S800 to step S806.

[0199] In Figure 7 In step S704 shown above, in order to further ensure the rationality of the displayed adjusted data information, it is necessary to perform further separate analysis and calculation on the displayed adjusted data information, which will be described in detail through the following steps.

[0200] The method for determining the displayed adjusted data information includes the following steps:

[0201] Step S800: Retrieve the source quantity value and source type information based on the selected source information.

[0202] Among them, the selected source information includes the source quantity value and source type information. The source quantity value refers to the quantity value of the source, and the source type information refers to the type information corresponding to the source. Retrieving the source quantity value and source type information through the selected source information facilitates subsequent use.

[0203] Step S801: Determine the source type adjustment information corresponding to the source type information according to the corresponding relationship between the source type information and the preset source type adjustment information.

[0204] Among them, the source type adjustment information refers to the adjustment information that the source belongs to the type needs to be adjusted. Different source type information corresponds to different source type adjustment information. The source type adjustment information is obtained by querying the preset source type adjustment database and matching, so as to facilitate subsequent use. For example, when the source type information is the source of different people, the source type adjustment information refers to the information adjusted according to the display preferences of this person. When the source type information is the source of different times, the source type adjustment information is the information adjusted according to the display preferences of this time. When the source type information is the source of different departments, the source type adjustment information is the information adjusted according to the display preferences of this department. When the source type information is the source of different regions, the source type adjustment information is the information adjusted according to the display preferences of this region.

[0205] Step S802: Determine whether the source quantity value is only one. If it is, execute step S803; if not, execute step S804.

[0206] Among them, by determining whether the source quantity value is only one, it is determined whether the adjustment can be directly based on the source type adjustment information.

[0207] Step S803: Adjust the information based on the source type to adjust the displayed initial data information to form the displayed adjusted data information.

[0208] Among them, when the source value is only one, it means that the adjustment can be directly made according to the source type adjustment information at this time. Therefore, the displayed initial data information is adjusted by the source type adjustment information to form the displayed adjusted data information, thereby improving the accuracy of the obtained displayed adjusted data information.

[0209] Step S804: Determine the source quantity adjustment information corresponding to the source quantity value according to the correspondence between the source quantity value and the preset source quantity adjustment information.

[0210] Among them, the source quantity adjustment information refers to the adjustment information for adjusting the display according to the source quantity. Different source quantity values correspond to different source quantity adjustment information. The source quantity adjustment information is obtained by querying and matching the preset source quantity adjustment database, which is convenient for subsequent use. The source quantity adjustment database stores a comparison table of different source quantity values and the corresponding source quantity adjustment information, and the source quantity adjustment database is obtained through pre-input. For example, when the source quantity value is two and the sources are from different persons, the source quantity adjustment information refers to the adjustment information required when comparing the sources of the two different persons. When the source quantity value is three, and two of the sources are from different persons and the other source is from a different region, the source quantity adjustment information refers to the adjustment information required when comparing the sources of the two different persons and when comparing the sources of different regions.

[0211] When the source quantity value is only one, it means that the adjustment cannot be directly made according to the source type adjustment information at this time. Therefore, the source quantity adjustment information is determined by querying the source quantity value, which is convenient for subsequent use.

[0212] Step S805: Combine the source type adjustment information and the source quantity adjustment information to form the comprehensive adjustment information.

[0213] Among them, the comprehensive adjustment information refers to the adjustment information corresponding to the comprehensive adjustment of the displayed initial data. By combining the source type adjustment information and the source quantity adjustment information, the comprehensive adjustment information is formed.

[0214] Step S806: Adjust the displayed initial data information based on the comprehensive adjustment information to form the displayed adjusted data information.

[0215] Among them, the initial data information for display is adjusted through comprehensive adjustment information to form the adjusted data information for display, thereby improving the accuracy of the obtained adjusted data information for display.

[0216] Based on the same inventive concept, an embodiment of the present invention provides an AI-based BI real-time decision-making system, including:

[0217] An acquisition module, configured to acquire real-time data information, user input information, and the current time point;

[0218] A memory, configured to store an AI-based BI real-time decision-making method as described above;

[0219] A processor, configured to load and execute the program in the memory.

[0220] Based on the same inventive concept, an embodiment of the present invention provides an intelligent terminal, including a memory and a processor. A computer program is stored on the memory and can be loaded and executed by the processor to implement an AI-based BI real-time decision-making method as described above.

[0221] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here.

[0222] The above is only the preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements made without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A BI real-time decision-making method based on AI, characterized in that: include: Obtain real-time data information of enterprise operations and user input information; Determine data requirement information and display requirement information based on user input information; Determine data selection information based on data demand information and real-time data information; Determine the decision display information based on display demand information and data selection information; Retrieving user identity information based on user input information; Determine user terminal information based on user identity information, and output decision display information to the terminal corresponding to the user terminal information; Methods for determining data requirement information and display requirement information include: Retrieving input type information and input data information based on user input information; According to the correspondence between the input category information and the preset category reference data information, the category reference data information corresponding to the input category information is determined; Based on the matching result between the input data information and the category reference data information, the matching meaning information is determined; Determine the meaning requirement information according to the matching meaning information; Retrieving meaning data type information and data type individual value based on meaning requirement information; The category display information is determined according to the meaning data category information and the data category number value, and the meaning requirement information is used as the data requirement information, and the category display information is used as the display requirement information.

2. The AI-based BI real-time decision-making method according to claim 1, characterized in that: Methods for determining the meaning requirement information include: According to the correspondence between the matching meaning information and the preset meaning initial requirement information, the meaning initial requirement information corresponding to the matching meaning information is determined; Determine the user's historical demand information based on the initial demand information and the user's identity information; Determine whether the user's historical demand information is consistent with the initial demand information; If yes, the meaning initial requirement information is used as the meaning requirement information; If not, then the deviation between the user's historical demand information and the initial demand information of the meaning is analyzed and used as demand deviation information; Determine demand adjustment information based on demand deviation information; The meaning initial requirement information is adjusted based on the requirement adjustment information to form the meaning requirement information.

3. The AI-based BI real-time decision-making method according to claim 2, characterized in that: Methods for determining demand adjustment information include: Based on the demand deviation information, retrieve the deviation occurrence time point and deviation frequency value; Get the current time point; Calculate the difference between the time point when the deviation occurs and the current time point and use it as the current time deviation value; According to the correspondence between the deviation frequency value and the preset time deviation reference interval, the time deviation reference interval corresponding to the deviation frequency value is determined; Determine whether the current time deviation value falls within the time deviation reference interval; If yes, then according to the correspondence between the demand deviation information and the preset deviation adjustment information, the deviation adjustment information corresponding to the demand deviation information is determined, and the deviation adjustment information is used as the demand adjustment information; If not, the difference between the current time deviation value and the time deviation reference interval is calculated and used as the time deviation abnormal value; According to the correspondence between the time deviation abnormal value, the deviation frequency value and the preset deviation comprehensive impact information, the deviation comprehensive impact information corresponding to the time deviation abnormal value and the deviation frequency value is determined; According to the correspondence between the deviation comprehensive impact information and the preset comprehensive impact adjustment information, the comprehensive impact adjustment information corresponding to the deviation comprehensive impact information is determined, and the comprehensive impact adjustment information is used as the demand adjustment information.

4. The AI-based BI real-time decision-making method according to claim 1, characterized in that: Methods for determining the type of display information include: According to the correspondence between the meaning data type information and the preset single-type display type information, the single-type display type information corresponding to the meaning data type information is determined; Determine whether the data type has only one value; If yes, the single category display type information is used as category display information; If not, the same single-category display type information is selected as the same type information, and the single-category display type information other than the same type information is selected as the difference type information; Retrieve values ​​of the same type based on information of the same type; Determine the same conversion type information corresponding to the same type information according to the correspondence between the same type information and the preset same conversion type information; Based on the consistency between the same conversion type information and the difference type information, the difference conversion value is determined; The same type of information is selected according to the same type of values ​​and the difference conversion values ​​to obtain selected type information, and the selected type information is used as the category display information.

5. The AI-based BI real-time decision-making method according to claim 4, characterized in that: Methods for selecting type information include: Sort the values ​​of the same type from large to small, and use the sorting results as the sorting values ​​of the same type; According to the correspondence between the same type ranking value and the preset same type ranking influence value, the same type ranking influence value corresponding to the same type ranking value is determined; Calculate the ratio between the difference conversion value and the data type value and use it as the conversion ratio value; Sort the conversion ratio values ​​from large to small, and use the sorting result as the conversion ratio sorting value; According to the correspondence between the conversion ratio ranking value and the preset conversion ratio influence value, the conversion ratio influence value corresponding to the conversion ratio ranking value is determined; Calculate the sum of the impact value of the same type of ranking and the impact value of the conversion ratio and use it as the comprehensive impact value of the ranking; The ranking is performed from large to small based on the ranking comprehensive impact value, and the same type information corresponding to the ranking comprehensive impact value that ranks first is used as the selected type information.

6. The AI-based BI real-time decision-making method according to claim 1, characterized in that: Methods for determining decision display information include: Retrieve selection source information and source data information based on data selection information; According to the correspondence between the source data information and the preset source data type information, the source data type information corresponding to the source data information is determined; Determine type conversion control information according to display requirement information and source data type information; Based on the type conversion control information, the source data information is converted to form the display initial data information; The display adjustment data information is determined according to the selected source information and the display initial data information, and the display adjustment data information is used as the decision display information.

7. The AI-based BI real-time decision-making method according to claim 6, characterized in that: Methods for determining display adjustment data information include: Retrieve source value and source type information based on selected source information; According to the correspondence between the source type information and the preset source type adjustment information, the source type adjustment information corresponding to the source type information is determined; Determine whether the source value is only one; If yes, adjusting the display initial data information based on the source type adjustment information to form display adjustment data information; If not, then determining the source number adjustment information corresponding to the source number value according to the correspondence between the source number value and the preset source number adjustment information; Combining source type adjustment information with source number adjustment information to form comprehensive adjustment information; The initial presentation data information is adjusted based on the comprehensive adjustment information to form presentation adjustment data information.

8. A BI real-time decision-making system based on AI, characterized in that: include: The acquisition module is used to obtain real-time data information, user input information and the current time point; A memory, used to store an AI-based BI real-time decision-making method according to any one of claims 1 to 7; The processor loads and executes the program in the memory.

9. An intelligent terminal, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a computer program and can be loaded by the processor and execute an AI-based BI real-time decision-making method as described in any one of claims 1 to 7.

Images