A business question-answering system and method based on big data and artificial intelligence

Abstract

The present invention discloses a business question-and-answer system and method based on big data and artificial intelligence, which relates to the field of artificial intelligence technology, including: obtaining the historical operation data of the device components of a computer, evaluating the deviation status index of the device components; crawling the computer abnormality report information set based on the Internet; analyzing the correlation between the abnormal event corresponding to the abnormal code of each abnormal report information and the device component, creating an abnormal event-associated device component matrix corresponding to the abnormal code; using the deviation status index of the device component as the attribute of the associated device component in the abnormal event-associated device component matrix corresponding to the abnormal code, constructing an abnormal question-and-answer auxiliary decision-making model for the device component of the abnormal code, and generating an abnormal business question-and-answer task for the computer. The advantages of the present invention are: improving the troubleshooting accuracy of abnormal code question-and-answer and optimizing user experience.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a business question-answering system and method based on big data and artificial intelligence. Background Art

[0002] Business question and answer based on big data and artificial intelligence refers to the use of big data technology and artificial intelligence algorithms to conduct in-depth mining and analysis of massive business data to answer various business-related questions in an automated manner.

[0003] Due to compatibility issues at the computer software and hardware levels, compatibility issues between the computer's device components and drivers will occur when the company's computers are faced with driver push updates. Secondly, due to the reduction in the device life of the local computer, the driver's scheduling resources do not match the hardware resources of the actual device components, resulting in frequent abnormal conditions. Summary of the invention

[0004] In order to solve the above technical problems, a business question and answer system and method based on big data and artificial intelligence are provided. This technical solution solves the above-mentioned compatibility issues between the computer's software and hardware levels, which leads to compatibility issues between the computer's device components and the driver when the company's computers are faced with driver push updates. Secondly, due to the reduction in the device life of the local computer, the driver's scheduling resources do not match the hardware resources of the actual device components, resulting in frequent abnormal conditions.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A business question-answering method based on big data and artificial intelligence, comprising:

[0007] Obtain historical operation data of the equipment components of the computer, analyze the deviation trend of the historical operation data of the equipment components, and evaluate the deviation status indicators of the equipment components;

[0008] Capture computer anomaly report information collection based on the Internet;

[0009] According to the computer abnormality report information set, the correlation between the abnormal event corresponding to the abnormal code of each abnormality report information and the device component is analyzed, and a device component matrix associated with the abnormal event corresponding to the abnormal code is created;

[0010] The deviation status index of the equipment component is used as the attribute of the associated equipment component in the abnormal event-associated equipment component matrix corresponding to the abnormal code, and an abnormal question-answering auxiliary decision-making model of the equipment component of the abnormal code is constructed to generate the abnormal business question-answering task of the computer.

[0011] Preferably, obtaining historical operation data of the device components of the computer, analyzing the deviation trend of the historical operation data of the device components, and evaluating the deviation status index of the device components specifically include:

[0012] Obtain standardized operating parameters of the device components of the computer, and construct a standardized operating parameter array of the device components of the computer;

[0013] Based on the historical operation data of the equipment components of the computer, the historical operation parameters of the equipment components are screened out, and an array of the historical operation parameters of the equipment components of the computer is constructed;

[0014] Linearly mapping the standardized operating parameter array of the computer's device components and the historical operating parameter array of the computer's device components to obtain a standardized operating parameter vector matrix of the computer's device components and a historical operating parameter vector matrix of the computer's device components;

[0015] According to the Euclidean distance formula, the spatial distance between each element in the standardized operation parameter vector matrix of the device components of the computer and each element in the historical operation parameter vector matrix of the device components of the computer is evaluated to obtain the deviation spatial distance of the device components of the computer;

[0016] Using a normalization formula, the deviation space distance of each device component of the computer is normalized to obtain a deviation state index of the device component of the computer;

[0017] The Euclidean distance formula is specifically:

[0018] The Euclidean distance formula is specifically: ;

[0019] In the formula, is the spatial distance of the ith device component of the computer, is the first device component of the computer A vector of standardized operating parameters, is the first device component of the computer A historical running parameter vector, is the total number of equipment components;

[0020] The normalization formula is specifically: ;

[0021] In the formula, is the deviation status indicator of the ith device component of the computer, The i-th device component of the computer deviates from the maximum spatial distance, is the minimum spatial distance deviation of the i-th device component.

[0022] Preferably, according to the computer abnormality report information set, analyzing the correlation between the abnormal event corresponding to the abnormal code of each abnormality report information and the device component, and creating the abnormal event associated device component matrix specifically includes:

[0023] Based on the abnormal report information in the computer abnormal report information set, mark the device component requirements of each abnormal event of the abnormal code for the computer, and build an impact device component association network of the abnormal event of the abnormal code;

[0024] Based on the impact device component association network of the abnormal event of the abnormal code, the average path length between the abnormal event node of the abnormal code and the device component sub-node in the network topology structure is analyzed, and recorded as the association attribute between the abnormal event corresponding to the abnormal code and the device component;

[0025] According to the correlation between the abnormal event corresponding to each abnormal code and the device component, a matrix of the abnormal event associated with the device component corresponding to the abnormal code is created;

[0026] The path length between the abnormal event node of the abnormal code and the device component subnode in the network topology structure is specifically: ;

[0027] In the formula, is the average path length between the kth abnormal event node and the ith device component subnode, is the shortest path length between the kth abnormal event node and the ith device component subnode, The total number of nodes in the network associated with the device component affected by the exception event with the exception code.

[0028] Preferably, the deviation state indicator of the device component is used as the attribute of the associated device component in the abnormal event associated device component matrix corresponding to the abnormal code, and an abnormal question-and-answer auxiliary decision model of the device component of the abnormal code is constructed. The abnormal business question-and-answer task of the computer is generated, which specifically includes:

[0029] The deviation state index of the device component is used as the attribute of the associated device component in the abnormal event associated device component matrix corresponding to the abnormal code, so as to generate the deviation state matrix of the abnormal event associated device component;

[0030] Based on random forest, according to the abnormal event-related device components corresponding to the abnormal code, the deviation state value of the abnormal event-related device components in the abnormal event-related device component deviation state matrix is ​​used as the decision weight of the branch node, and the abnormal event-related device components are used as the division features of the branch sub-leaf nodes. A device component identification decision tree for each abnormal event is constructed, and an abnormal question-and-answer auxiliary decision model for the device component of the abnormal code is established;

[0031] Substitute the computer's exception code into the exception code's device component exception question and answer auxiliary decision-making model, calculate the information gain of the exception code's exception event for the computer's device component requirements for each branch and leaf node, and generate the optimal exception business question and answer for the computer's device component of the exception code.

[0032] Furthermore, a business question-answering system based on big data and artificial intelligence is used to implement the business question-answering method based on big data and artificial intelligence as described above, including:

[0033] A status evaluation module obtains the historical operation data of the equipment components of the computer, analyzes the deviation trend of the historical operation data of the equipment components, and evaluates the deviation status index of the equipment components;

[0034] Abnormal data collection module, which captures computer abnormality report information collection based on the Internet;

[0035] The correlation analysis module is electrically connected to the abnormal data acquisition module, and is used to analyze the correlation between the abnormal event corresponding to the abnormal code of each abnormal report information and the device component according to the computer abnormal report information set, and create an abnormal event-related device component matrix corresponding to the abnormal code;

[0036] The decision analysis module is electrically connected to the state evaluation module and the association analysis module. The decision analysis module is used to use the deviation state index of the equipment component as the attribute of the associated equipment component in the abnormal event-associated equipment component matrix corresponding to the abnormal code, build an abnormal question-and-answer auxiliary decision model for the equipment component of the abnormal code, and generate an abnormal business question-and-answer task for the computer.

[0037] Optional, Condition Assessment module, including:

[0038] A standardization unit, which obtains standardized operating parameters of the device components of the computer and constructs a standardized operating parameter array of the device components of the computer;

[0039] A historical data unit, based on the historical operation data of the device components of the computer, filters out the historical operation parameters of the device components and constructs an array of the historical operation parameters of the device components of the computer;

[0040] A mapping unit, which performs linear mapping on the standardized operation parameter array of the device components of the computer and the historical operation parameter array of the device components of the computer to obtain a standardized operation parameter vector matrix of the device components of the computer and a historical operation parameter vector matrix of the device components of the computer;

[0041] The spatial distance unit evaluates the spatial distance between each element in the standardized operation parameter vector matrix of the device component of the computer and each element in the historical operation parameter vector matrix of the device component of the computer according to the Euclidean distance formula, and obtains the deviation spatial distance of the device component of the computer;

[0042] The deviation state unit uses a normalization formula to normalize the deviation space distance of each device component of the computer to obtain the deviation state index of the device component of the computer.

[0043] Optional, association analysis module, including:

[0044] The association network unit marks the device component requirements of each abnormal event of the abnormal code for the computer based on the abnormal report information in the computer abnormal report information set, and builds an association network of the device components affected by the abnormal event of the abnormal code;

[0045] The association attribute unit analyzes the average path length between the abnormal event node of the abnormal code and the device component sub-node in the network topology structure based on the impact device component association network of the abnormal event of the abnormal code, and records it as the association attribute between the abnormal event corresponding to the abnormal code and the device component;

[0046] The associated device matrix unit creates an abnormal event associated device component matrix corresponding to the abnormal code according to the association between the abnormal event corresponding to each abnormal code and the device component.

[0047] Optional, decision analysis module, including:

[0048] The matrix attribute updating unit uses the deviation state index of the device component as the attribute of the associated device component in the abnormal event associated device component matrix corresponding to the abnormal code to generate the deviation state matrix of the abnormal event associated device component;

[0049] The decision model unit is based on random forests and associates the device components of the abnormal events corresponding to the abnormal codes. The deviation state values ​​of the device components associated with the abnormal events in the deviation state matrix of the abnormal event-associated device components are used as the decision weights of the branch nodes. The device components associated with the abnormal events are used as the division features of the branch sub-leaf nodes to construct a device component identification decision tree for each abnormal event, and to build an abnormal question-and-answer auxiliary decision model for the device components of the abnormal codes.

[0050] The decision generation unit substitutes the computer's exception code into the abnormal question and answer auxiliary decision model of the equipment component of the exception code, calculates the information gain of the abnormal event of the exception code for the computer's equipment component demand for each branch and leaf node, and generates the optimal abnormal business question and answer of the computer's equipment component of the exception code.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present invention proposes a business question-and-answer solution based on big data and artificial intelligence. This solution obtains the historical operation data of computer equipment components, analyzes their deviation trends to evaluate status indicators, and combines computer abnormality report information on the Internet to analyze the correlation between abnormal codes and equipment components, constructs an association matrix, and integrates the deviation status indicators of equipment components as attributes into the association matrix to form an abnormal question-and-answer auxiliary decision-making model to effectively generate abnormal business question-and-answer tasks for computers. The beneficial effects are: improving the troubleshooting accuracy of abnormal code questions and answers and optimizing user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of a business question-answering method based on big data and artificial intelligence of the present invention;

[0054] Figure 2 A flow chart of a method for evaluating a deviation status indicator of a device component in the present invention;

[0055] Figure 3 A flow chart of a method for analyzing the association between an abnormal event corresponding to an abnormal code of each abnormal report information and a device component in the present invention;

[0056] Figure 4 A flow chart of a method for generating an abnormal business question-and-answer task of a computer according to the present invention;

[0057] Figure 5 This is a framework diagram of a business question-and-answer system based on big data and artificial intelligence according to the present invention. DETAILED DESCRIPTION

[0058] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0059] Reference Figure 1 As shown, a business question-answering method based on big data and artificial intelligence includes:

[0060] Obtain historical operation data of the equipment components of the computer, analyze the deviation trend of the historical operation data of the equipment components, and evaluate the deviation status indicators of the equipment components;

[0061] Capture computer anomaly report information collection based on the Internet;

[0062] According to the computer abnormality report information set, the correlation between the abnormal event corresponding to the abnormal code of each abnormality report information and the device component is analyzed, and a device component matrix associated with the abnormal event corresponding to the abnormal code is created;

[0063] The deviation status index of the equipment component is used as the attribute of the associated equipment component in the abnormal event-associated equipment component matrix corresponding to the abnormal code, and an abnormal question-answering auxiliary decision-making model of the equipment component of the abnormal code is constructed to generate the abnormal business question-answering task of the computer.

[0064] This solution obtains the historical operation data of computer equipment components, analyzes their deviation trends to evaluate status indicators, and combines computer abnormality report information on the Internet to analyze the correlation between abnormal codes and equipment components, builds an association matrix, and incorporates the deviation status indicators of equipment components as attributes into the association matrix to form an abnormal question-and-answer auxiliary decision-making model to effectively generate abnormal business question-and-answer tasks for computers. The beneficial effects are: improving the troubleshooting accuracy of abnormal code questions and answers and optimizing user experience.

[0065] Reference Figure 2 As shown, the historical operation data of the device components of the computer are obtained, the deviation trend of the historical operation data of the device components is analyzed, and the deviation status indicators of the device components are evaluated, which specifically include:

[0066] Obtain standardized operating parameters of the device components of the computer, and construct a standardized operating parameter array of the device components of the computer;

[0067] Based on the historical operation data of the equipment components of the computer, the historical operation parameters of the equipment components are screened out, and an array of the historical operation parameters of the equipment components of the computer is constructed;

[0068] Linearly mapping the standardized operating parameter array of the computer's device components and the historical operating parameter array of the computer's device components to obtain a standardized operating parameter vector matrix of the computer's device components and a historical operating parameter vector matrix of the computer's device components;

[0069] According to the Euclidean distance formula, the spatial distance between each element in the standardized operation parameter vector matrix of the device components of the computer and each element in the historical operation parameter vector matrix of the device components of the computer is evaluated to obtain the deviation spatial distance of the device components of the computer;

[0070] Using a normalization formula, the deviation space distance of each device component of the computer is normalized to obtain a deviation state index of the device component of the computer;

[0071] The Euclidean distance formula is specifically: ;

[0072] In the formula, is the spatial distance of the ith device component of the computer, is the first device component of the computer A vector of standardized operating parameters, is the first device component of the computer A historical running parameter vector, is the total number of equipment components;

[0073] The normalization formula is specifically: ;

[0074] In the formula, is the deviation status indicator of the ith device component of the computer, The i-th device component of the computer deviates from the maximum spatial distance, is the minimum spatial distance deviation of the i-th device component.

[0075] This solution can accurately capture the changes in the operating status of computer equipment components, discover the deviation trends of computer equipment components, and provide key decision support for subsequent abnormal question and answer by quantitatively evaluating the deviation status indicators of equipment components.

[0076] Reference Figure 3 As shown, according to the computer abnormality report information set, the correlation between the abnormal event corresponding to the abnormal code of each abnormality report information and the device component is analyzed, and the abnormal event associated device component matrix is ​​created, which specifically includes:

[0077] Based on the abnormal report information in the computer abnormal report information set, mark the device component requirements of each abnormal event of the abnormal code for the computer, and build an impact device component association network of the abnormal event of the abnormal code;

[0078] Based on the impact device component association network of the abnormal event of the abnormal code, the average path length between the abnormal event node of the abnormal code and the device component sub-node in the network topology structure is analyzed, and recorded as the association attribute between the abnormal event corresponding to the abnormal code and the device component;

[0079] According to the correlation between the abnormal event corresponding to each abnormal code and the device component, a matrix of the abnormal event associated with the device component corresponding to the abnormal code is created;

[0080] The path length between the abnormal event node of the abnormal code and the device component subnode in the network topology structure is specifically: ;

[0081] In the formula, is the average path length between the kth abnormal event node and the ith device component subnode, is the shortest path length between the kth abnormal event node and the ith device component subnode, The total number of nodes in the network associated with the device component affected by the exception event with the exception code.

[0082] It can be understood that by analyzing the computer anomaly report information set, marking the impact of the abnormal events corresponding to the abnormal codes on the computer device components, building an association network, and calculating the average path length between the abnormal event nodes and the device component subnodes as the correlation attribute, the abnormal event-associated device component matrix is ​​finally created. The beneficial effect is that it helps to quickly locate the root cause of the problem and improve the efficiency of fault anomaly troubleshooting.

[0083] Reference Figure 4 As shown in the figure, the deviation state index of the equipment component is used as the attribute of the associated equipment component in the abnormal event associated equipment component matrix corresponding to the abnormal code, and the abnormal question-answering auxiliary decision-making model of the equipment component of the abnormal code is constructed. The abnormal business question-answering task generated by the computer specifically includes:

[0084] The deviation state index of the device component is used as the attribute of the associated device component in the abnormal event associated device component matrix corresponding to the abnormal code, so as to generate the deviation state matrix of the abnormal event associated device component;

[0085] Based on random forest, according to the abnormal event-related device components corresponding to the abnormal code, the deviation state value of the abnormal event-related device components in the abnormal event-related device component deviation state matrix is ​​used as the decision weight of the branch node, and the abnormal event-related device components are used as the division features of the branch sub-leaf nodes. A device component identification decision tree for each abnormal event is constructed, and an abnormal question-and-answer auxiliary decision model for the device component of the abnormal code is established;

[0086] Substitute the computer's exception code into the exception code's device component exception question and answer auxiliary decision-making model, calculate the information gain of the exception code's exception event for the computer's device component requirements for each branch and leaf node, and generate the optimal exception business question and answer for the computer's device component of the exception code.

[0087] The implementation principle of this solution is to use the deviation status indicator of the equipment component as a key attribute, and integrate it into the equipment component matrix associated with the abnormal code and abnormal event, thereby forming the deviation status matrix of the equipment component associated with the abnormal event. Based on this matrix, we use the random forest algorithm to build a decision tree model, in which the weight of the branch node of the decision tree is determined by the deviation status value, and the division of the branch cotyledon node is based on the characteristics of the equipment component associated with the abnormal event. For the input computer abnormal code, by calculating the information gain of each branch cotyledon node, it can quickly identify the equipment component related to the abnormal code and generate the optimal abnormal business question and answer. The beneficial effect is: it improves the accuracy and efficiency of abnormal diagnosis, and assists technicians to locate abnormal problems of computer equipment components more quickly.

[0088] Reference Figure 5 As shown, based on the same inventive concept of a business question-answering method based on big data and artificial intelligence, a business question-answering system based on big data and artificial intelligence is proposed, comprising:

[0089] A status evaluation module obtains the historical operation data of the equipment components of the computer, analyzes the deviation trend of the historical operation data of the equipment components, and evaluates the deviation status index of the equipment components;

[0090] Abnormal data collection module, which captures computer abnormality report information collection based on the Internet;

[0091] The correlation analysis module is electrically connected to the abnormal data acquisition module, and is used to analyze the correlation between the abnormal event corresponding to the abnormal code of each abnormal report information and the device component according to the computer abnormal report information set, and create an abnormal event-related device component matrix corresponding to the abnormal code;

[0092] The decision analysis module is electrically connected to the state evaluation module and the association analysis module. The decision analysis module is used to use the deviation state index of the equipment component as the attribute of the associated equipment component in the abnormal event-associated equipment component matrix corresponding to the abnormal code, build an abnormal question-and-answer auxiliary decision model for the equipment component of the abnormal code, and generate an abnormal business question-and-answer task for the computer.

[0093] Among them, the status assessment module includes:

[0094] A standardization unit, which obtains standardized operating parameters of the device components of the computer and constructs an array of standardized operating parameters of the device components of the computer;

[0095] A historical data unit, based on the historical operation data of the device components of the computer, filters out the historical operation parameters of the device components and constructs an array of the historical operation parameters of the device components of the computer;

[0096] A mapping unit, which performs linear mapping on the standardized operation parameter array of the device components of the computer and the historical operation parameter array of the device components of the computer to obtain a standardized operation parameter vector matrix of the device components of the computer and a historical operation parameter vector matrix of the device components of the computer;

[0097] The spatial distance unit evaluates the spatial distance between each element in the standardized operation parameter vector matrix of the device component of the computer and each element in the historical operation parameter vector matrix of the device component of the computer according to the Euclidean distance formula, and obtains the deviation spatial distance of the device component of the computer;

[0098] The deviation state unit uses a normalization formula to normalize the deviation space distance of each device component of the computer to obtain the deviation state index of the device component of the computer.

[0099] The association analysis module includes:

[0100] The association network unit marks the device component requirements of each abnormal event of the abnormal code for the computer based on the abnormal report information in the computer abnormal report information set, and builds an association network of the device components affected by the abnormal event of the abnormal code;

[0101] The association attribute unit analyzes the average path length between the abnormal event node of the abnormal code and the device component sub-node in the network topology structure based on the impact device component association network of the abnormal event of the abnormal code, and records it as the association attribute between the abnormal event corresponding to the abnormal code and the device component;

[0102] The associated device matrix unit creates an abnormal event associated device component matrix corresponding to the abnormal code according to the association between the abnormal event corresponding to each abnormal code and the device component.

[0103] The decision analysis module includes:

[0104] The matrix attribute updating unit uses the deviation state index of the device component as the attribute of the associated device component in the abnormal event associated device component matrix corresponding to the abnormal code to generate the deviation state matrix of the abnormal event associated device component;

[0105] The decision model unit is based on random forests and associates the device components of the abnormal events corresponding to the abnormal codes. The deviation state values ​​of the device components associated with the abnormal events in the deviation state matrix of the abnormal event-associated device components are used as the decision weights of the branch nodes. The device components associated with the abnormal events are used as the division features of the branch sub-leaf nodes to construct a device component identification decision tree for each abnormal event, and to build an abnormal question-and-answer auxiliary decision model for the device components of the abnormal codes.

[0106] The decision generation unit substitutes the computer's exception code into the abnormal question and answer auxiliary decision model of the equipment component of the exception code, calculates the information gain of the abnormal event of the exception code for the computer's equipment component demand for each branch and leaf node, and generates the optimal abnormal business question and answer of the computer's equipment component of the exception code.

[0107] The usage process of a business question-answering system based on big data and artificial intelligence is as follows:

[0108] Step 1: Obtain standardized operating parameters of the computer's device components and construct a standardized operating parameter array of the computer's device components;

[0109] Step 2: Based on the historical operation data of the device components of the computer, filter out the historical operation parameters of the device components, and construct an array of the historical operation parameters of the device components of the computer;

[0110] Step 3: linearly map the standardized operating parameter array of the computer's device components to the historical operating parameter array of the computer's device components to obtain a standardized operating parameter vector matrix of the computer's device components and a historical operating parameter vector matrix of the computer's device components;

[0111] Step 4: According to the Euclidean distance formula, the spatial distance between each element in the standardized operating parameter vector matrix of the computer's device components and each element in the historical operating parameter vector matrix of the computer's device components is evaluated to obtain the deviation spatial distance of the computer's device components;

[0112] Step 5: Use a normalization formula to normalize the deviation spatial distance of each device component of the computer to obtain a deviation state index of the device component of the computer.

[0113] Step 6: Capture computer anomaly report information collection based on the Internet;

[0114] Step 7: Based on the abnormal report information in the computer abnormal report information set, mark the device component requirements of each abnormal event of the abnormal code for the computer, and build an impact device component association network of the abnormal event of the abnormal code;

[0115] Step 8: Based on the impact device component association network of the abnormal event of the abnormal code, the average path length between the abnormal event node of the abnormal code and the device component sub-node in the network topology is analyzed, and recorded as the correlation attribute between the abnormal event corresponding to the abnormal code and the device component;

[0116] Step 9: According to the correlation between the abnormal event corresponding to each abnormal code and the device component, create an abnormal event-related device component matrix corresponding to the abnormal code;

[0117] Step 10: Using the deviation state index of the device component as the attribute of the associated device component in the abnormal event associated device component matrix corresponding to the abnormal code, to generate the abnormal event associated device component deviation state matrix;

[0118] Step 11: Based on random forest, according to the abnormal event-associated device components corresponding to the abnormal code, the deviation state value of the abnormal event-associated device components in the abnormal event-associated device component deviation state matrix is ​​used as the decision weight of the branch node, and the abnormal event-associated device components are used as the division features of the branch sub-leaf nodes. A device component identification decision tree for each abnormal event is constructed, and an abnormal question-and-answer auxiliary decision model for the device component of the abnormal code is constructed;

[0119] Step 12: Substitute the computer's exception code into the exception code's device component exception question and answer auxiliary decision-making model, calculate the information gain of the exception code's exception event for the computer's device component requirements for each branch and leaf node, and generate the optimal exception business question and answer for the computer's device component of the exception code.

[0120] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A business question-answering method based on big data and artificial intelligence, characterized in that: include: Obtain historical operation data of the equipment components of the computer, analyze the deviation trend of the historical operation data of the equipment components, and evaluate the deviation status indicators of the equipment components; Capture computer anomaly report information collection based on the Internet; According to the computer abnormality report information set, the correlation between the abnormal event corresponding to the abnormal code of each abnormality report information and the device component is analyzed, and a device component matrix associated with the abnormal event corresponding to the abnormal code is created; The deviation status index of the equipment component is used as the attribute of the associated equipment component in the abnormal event associated equipment component matrix corresponding to the abnormal code, and an abnormal question-answering auxiliary decision-making model of the equipment component of the abnormal code is constructed to generate the abnormal business question-answering task of the computer; The method of analyzing the correlation between the abnormal event and the device component corresponding to the abnormal code of each abnormal report information according to the computer abnormal report information set and creating the abnormal event-related device component matrix corresponding to the abnormal code specifically includes: Based on the abnormal report information in the computer abnormal report information set, mark the device component requirements of each abnormal event of the abnormal code for the computer, and build an impact device component association network of the abnormal event of the abnormal code; Based on the impact device component association network of the abnormal event of the abnormal code, the average path length between the abnormal event node of the abnormal code and the device component sub-node in the network topology structure is analyzed, and recorded as the association attribute between the abnormal event corresponding to the abnormal code and the device component; According to the correlation between the abnormal event corresponding to each abnormal code and the device component, a matrix of the abnormal event associated with the device component corresponding to the abnormal code is created; The path length between the abnormal event node of the abnormal code and the device component subnode in the network topology structure is specifically: ; In the formula, is the average path length between the kth abnormal event node and the ith device component subnode, is the shortest path length between the kth abnormal event node and the ith device component subnode, The total number of nodes in the network associated with the device component affected by the abnormal event of the abnormal code; The method of using the deviation status indicator of the device component as the attribute of the associated device component in the abnormal event associated device component matrix corresponding to the abnormal code, constructing the device component abnormal question and answer auxiliary decision model of the abnormal code, and generating the abnormal business question and answer task of the computer specifically includes: The deviation state index of the device component is used as the attribute of the associated device component in the abnormal event associated device component matrix corresponding to the abnormal code, so as to generate the deviation state matrix of the abnormal event associated device component; Based on random forest, according to the abnormal event-related device components corresponding to the abnormal code, the deviation state value of the abnormal event-related device components in the abnormal event-related device component deviation state matrix is ​​used as the decision weight of the branch node, and the abnormal event-related device components are used as the division features of the branch sub-leaf nodes. A device component identification decision tree for each abnormal event is constructed, and an abnormal question-and-answer auxiliary decision model for the device component of the abnormal code is established; Substitute the computer's exception code into the exception code's device component exception question and answer auxiliary decision-making model, calculate the information gain of the exception code's exception event for the computer's device component requirements for each branch and leaf node, and generate the optimal exception business question and answer for the computer's device component of the exception code.

2. A business question-answering method based on big data and artificial intelligence according to claim 1, characterized in that: The obtaining of historical operation data of the device components of the computer, analyzing the deviation trend of the historical operation data of the device components, and evaluating the deviation status index of the device components specifically includes: Obtain standardized operating parameters of the device components of the computer, and construct a standardized operating parameter array of the device components of the computer; Based on the historical operation data of the equipment components of the computer, the historical operation parameters of the equipment components are screened out, and an array of the historical operation parameters of the equipment components of the computer is constructed; Linearly mapping the standardized operating parameter array of the computer's device components and the historical operating parameter array of the computer's device components to obtain a standardized operating parameter vector matrix of the computer's device components and a historical operating parameter vector matrix of the computer's device components; According to the Euclidean distance formula, the spatial distance between each element in the standardized operation parameter vector matrix of the device components of the computer and each element in the historical operation parameter vector matrix of the device components of the computer is evaluated to obtain the deviation spatial distance of the device components of the computer; Using a normalization formula, the deviation space distance of each device component of the computer is normalized to obtain a deviation state index of the device component of the computer; The Euclidean distance formula is specifically: ; In the formula, is the spatial distance of the ith device component of the computer, is the first device component of the computer A vector of standardized operating parameters, is the first device component of the computer A historical operation parameter vector, is the total number of equipment components; The normalization formula is specifically: ; In the formula, is the deviation status indicator of the ith device component of the computer, The i-th device component of the computer deviates from the maximum spatial distance, is the minimum spatial distance deviation of the i-th device component.

3. A business question-answering system based on big data and artificial intelligence, characterized by: Used to implement a business question-answering method based on big data and artificial intelligence as described in any one of claims 1-2, comprising: A status evaluation module obtains the historical operation data of the equipment components of the computer, analyzes the deviation trend of the historical operation data of the equipment components, and evaluates the deviation status index of the equipment components; Abnormal data collection module, which captures computer abnormality report information collection based on the Internet; The correlation analysis module is electrically connected to the abnormal data acquisition module, and is used to analyze the correlation between the abnormal event corresponding to the abnormal code of each abnormal report information and the device component according to the computer abnormal report information set, and create an abnormal event-related device component matrix corresponding to the abnormal code; The decision analysis module is electrically connected to the state evaluation module and the association analysis module. The decision analysis module is used to use the deviation state index of the equipment component as the attribute of the associated equipment component in the abnormal event-associated equipment component matrix corresponding to the abnormal code, build an abnormal question-and-answer auxiliary decision model for the equipment component of the abnormal code, and generate an abnormal business question-and-answer task for the computer.

Images