Electric power project data processing, analyzing and decision-making system

Abstract

The invention provides an electric power project data processing, analysis and decision-making system, which comprises an information collection module, an information analysis module, a project identification module, an information integration module, a project analysis module and a terminal display module which are connected in sequence, and is used for automatically identifying application of project engineering of an enterprise, tracking the project engineering in real time, and providing a decision-making result. Then, based on information content in the enterprise project engineering, the enterprise project engineering is divided into multiple pieces of one-way information, retrieval is carried out in the intellectual property information base by taking a central word group in the one-way information as retrieval key information, key matching information is obtained, and a property risk coefficient between the key matching information and the one-way information is calculated; and according to the property risk coefficient, a property protection signal is determined, so that innovation achievements in enterprise project engineering are automatically monitored, and the timeliness of intellectual property is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of power information processing, and in particular to a power project data processing, analysis and decision-making system. Background Art

[0002] In enterprises related to power supply companies, multiple project engineering projects will be established, and one power engineering project may involve multiple intellectual property rights.

[0003] Intellectual property rights are the rights of inventors of scientific and technological achievements to possess, use, dispose of and benefit from them according to law, and are proof of the originality and exclusivity of scientific and technological achievements. Research-based enterprises are the supporters of many cutting-edge scientific and technological achievements and the owners of intellectual property rights.

[0004] At present, due to the large amount of information involved, manual identification of intellectual property rights in power project engineering not only has low work processing efficiency, but also low accuracy when manually judging the innovative achievements in power project engineering; this will lead to delays in the protection of innovative achievements. Summary of the invention

[0005] The purpose of the present invention is to solve the problems in the background technology and to propose a power project data processing, analysis and decision-making system.

[0006] The present invention adopts the following technical solution: a power project data processing, analysis and decision-making system, comprising:

[0007] The information analysis module is used to identify the project engineering of the enterprise and take it as the target engineering. According to the project application instruction of the target engineering, the text information in the project application instruction is analyzed to obtain the project information set, and then the word frequency of the information phrases in the project information set is calculated to determine the project theme of the project engineering, wherein the project theme includes the project key carrier and the node phrase. The information analysis module and the engineering identification module are connected in a one-way communication manner;

[0008] The project identification module monitors and identifies the progress of the target project based on the project theme of the project engineering and obtains real-time project information. The project identification module and the information integration module are connected in a one-way manner.

[0009] The information integration module is used to divide the real-time project information into several sub-information according to the title information, and to split the project information set into multiple sub-sets according to the sub-information, and then calculate the phrase feature value of the node phrase in the sub-set, and determine the central phrase in the sub-set based on the phrase feature value, and then merge the sub-sets with the same central phrase to obtain a one-way set and one-way information. The information integration module and the intellectual property analysis module are connected in a one-way communication manner;

[0010] The intellectual property analysis module is used to use the central phrase in the one-way information as the key information for retrieval and to perform retrieval to obtain key matching information. At the same time, the one-way set corresponding to the one-phase information is processed to obtain a dimension set, and the intellectual property risk coefficient of the dimension set and the key matching information is calculated. Based on the intellectual property risk coefficient, a risk warning signal and an intellectual property protection signal are generated. After that, the intellectual property analysis module and the terminal display module are connected in a one-way communication manner.

[0011] The terminal display module is used to receive risk warning signals and intellectual property protection signals and display them on the terminal device.

[0012] As a further solution of the present invention, the method for determining the project theme includes:

[0013] Identify the project application instructions in the collected information and obtain the engineering projects of the enterprise. The project application instructions include the project application form;

[0014] Randomly select a project from the enterprise's engineering projects as the target project;

[0015] Extract the project application instructions of the target project, use natural language processing technology to split the text information in the project application instructions into several information phrases, then identify the parts of speech in the information phrases, extract the content word parts of speech in the parts of speech, and summarize the information phrases corresponding to the content word parts of speech into a project information set;

[0016] Calculate the word frequency FCi of each information phrase in the project information set, CXi represents the number of times information phrase i appears in the project information set, and ZL represents the total number of information phrases in the project information set, where i represents different information phrases in the project information set, and i∈[1,I], indicating that there are I different information phrases in the project information set;

[0017] Arrange the word frequencies of information phrase i in descending order to obtain the first sequence of information phrases;

[0018] Take the information phrase at the first position in the first sequence and mark it as the key carrier of the project;

[0019] A first threshold X1 is set and used as a dividing line to select node phrases in the first sequence, wherein the node phrases refer to information phrases whose frequency is greater than or equal to the first threshold X1.

[0020] As a further solution of the present invention, parts of speech include content word parts of speech and function word parts of speech. Content word parts of speech refer to the components that constitute the main body of a sentence and can independently express a complete meaning. Function word parts of speech refer to words that play a role of modification and connection.

[0021] As a further solution of the present invention, the action of collecting information is completed through an information collection module, and the information collection module is used to collect all information in the enterprise and upload it to the enterprise cloud platform for information storage. There is a two-way communication connection between the enterprise cloud platform and the information analysis module.

[0022] As a further solution of the present invention, a method for calculating a phrase feature value of a node phrase includes:

[0023] SS1: randomly select a project and obtain the real-time project information of the project, and divide the text information in the real-time project information into several sub-information according to the information title, where the information title refers to the title structure layer in the hierarchical relationship of the text information;

[0024] SS2: Split the project information set into several sub-sets according to the information content in the sub-information, wherein the information phrases in each sub-set correspond to the information content in the sub-information;

[0025] Arbitrarily select a subset as the target set, and mark the subsets other than the target set as the control set;

[0026] The node phrases in the target set and the control set are identified respectively, and then the formula is used The phrase feature value Vj of node phrase j in the target set is calculated, where j represents different node phrases, n represents different control sets, and n∈[1,N], N represents the total number of control sets, Lj represents the number of times node phrase j appears in the target set, Mz represents the total number of information phrases in the target set, Djn represents the number of times node phrase j appears in control set n, Hn represents the total number of information phrases in control set n, and A1 is a fixed constant.

[0027] As a further solution of the present invention, a method for obtaining a one-way collection and one-way information includes:

[0028] Set the feature threshold X2, compare the phrase feature value of the node phrase in each subset with the feature threshold X2, and take the node phrase j with Vj≥X2 as the central phrase of this subset;

[0029] The central phrases of each subset are compared. If there are subsets with the same central phrase, the corresponding subsets are merged and marked as unidirectional sets. At the same time, the sub-information corresponding to each subset in the unidirectional set is merged to obtain unidirectional information.

[0030] As a further solution of the present invention, a method for obtaining an intellectual property protection signal includes:

[0031] Randomly select a one-way information as the target information, obtain the central phrase in the target information, use the central phrase as the retrieval key information, search in the intellectual property information database on the Internet, and obtain the key matching information;

[0032] Obtain the one-way set corresponding to the target information, remove the central phrase in the one-way set, integrate the remaining information phrases in the one-way set, and mark them as a dimension set;

[0033] Then, the information phrases and key matching information in the dimension set are converted into vector form respectively, and based on the cosine similarity algorithm, the similarity value between any key matching information and the dimension set is calculated and marked as the intellectual property risk coefficient;

[0034] Compare the intellectual property risk coefficient with the risk threshold Fy, take the maximum value FXmax of the intellectual property risk coefficient in the key matching information, and compare it with the risk threshold Fy. If FXmax≥Fy, generate a risk warning signal; if FXmax<Fy, generate an intellectual property protection signal.

[0035] In operation, the present invention identifies the application of the enterprise's project engineering and tracks the project engineering in real time. Then, based on the information content in the enterprise project engineering, the enterprise project engineering is divided into multiple one-way information, and the central phrase in the one-way information is used as the retrieval key information to search in the intellectual property information database to obtain key matching information, calculate the intellectual property risk coefficient between the key matching information and the one-way information, and determine the intellectual property protection signal based on the intellectual property risk coefficient, thereby automatically monitoring the innovative achievements in the enterprise project engineering, ensuring the timeliness of the intellectual property rights, reducing the need for manual operation and identification, thereby improving the processing speed and accuracy, and enhancing the competitiveness of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0037] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0038] Reference Figure 1 , a power project data processing, analysis and decision-making system, comprising an information collection module, an information analysis module, an engineering identification module, an information integration module, an intellectual property analysis module and a terminal display module connected in sequence;

[0039] The information collection module is used to collect all the information in the enterprise and upload it to the enterprise cloud platform for information storage. There is a two-way communication connection between the enterprise cloud platform and the information analysis module;

[0040] The information analysis module identifies the project application instruction information in the project based on the enterprise's engineering projects, and determines the project theme of the project. In this embodiment, the project refers to the power-related project carried out by the power supply company. The specific project theme determination method includes:

[0041] S1: Based on the information collected by the enterprise by the information collection module, the project application instructions in the collected information are identified to obtain the engineering projects of the enterprise, wherein the project application instructions include application information such as the project application form;

[0042] S2: randomly select an engineering project from the engineering projects of the enterprise as the target project, and take this target project as an example to identify the project theme in the target project, wherein the project theme includes the key carrier and node phrase of the project;

[0043] Furthermore, specific methods for identifying project themes include:

[0044] S21: extracting the project application instructions of the target project, and then using natural language processing technology to split the text information in the project application instructions into a number of information phrases, and at the same time identifying the parts of speech in the information phrases, extracting the content word parts of speech in the parts of speech, and summarizing the information phrases corresponding to the content word parts of speech into a project information set;

[0045] Among them, the word classes include content word classes and function word classes. Content word classes refer to the components that constitute the main body of a sentence and can independently express a complete meaning, including nouns, verbs, adjectives, numerals, quantifiers, adverbs, and pronouns. Function word classes refer to words that play a role in modification and connection, including adverbs, prepositions, conjunctions, auxiliary words, etc. At the same time, natural language processing technology is an existing technology and will not be elaborated here.

[0046] S22: Calculate the word frequency FCi of each information phrase in the project information set, where CXi represents the number of times information phrase i appears in the project information set, and ZL represents the total number of information phrases in the project information set, where i represents different information phrases in the project information set, and i∈[1,I], indicating that there are I different information phrases in the project information set;

[0047] Then, the word frequencies of information phrases i are arranged in descending order to obtain the first sequence of information phrases;

[0048] Take the information phrase at the first position in the first sequence and mark it as the key carrier of the project;

[0049] S23: Then, a first threshold value X1 is set, and the first threshold value X1 is used as a dividing line to select a node phrase in the first sequence;

[0050] Among them, the node phrase refers to an information phrase whose word frequency is greater than or equal to the first threshold value X1, and the specific value of the first threshold value is obtained by those skilled in the art after big data calculation. It should be further explained that when selecting the node phrase, the information phrase in the first position does not participate in the selection of the node phrase;

[0051] There is a one-way communication connection between the information analysis module and the engineering identification module;

[0052] The project identification module monitors and identifies the progress of the enterprise's project engineering based on the key project carriers and node phrases in the project application instructions, and obtains real-time project information. After that, the project identification module and the information integration module are connected in a one-way communication;

[0053] The information integration module is used to integrate the information content in the real-time project information and divide the project into multiple one-way information;

[0054] SS1: randomly select a project and obtain the real-time project information of the project, and divide the text information in the real-time project information into several sub-information according to the information title;

[0055] Among them, the information title refers to the title structure layer in the hierarchical relationship of text information;

[0056] SS2: Split the project information set into several sub-sets according to the information content in the sub-information, wherein the information phrases in each sub-set correspond to the information content in the sub-information;

[0057] Arbitrarily select a subset as the target set, and mark the subsets other than the target set as the control set;

[0058] The node phrases in the target set and the control set are identified respectively, and then the formula is used Calculate the phrase feature value Vj of node phrase j in the target set, where j represents different node phrases, n represents different control sets, and n∈[1,N], N represents the total number of control sets, Lj represents the number of times node phrase j appears in the target set, Mz represents the total number of information phrases in the target set, Djn represents the number of times node phrase j appears in control set n, Hn represents the total number of information phrases in control set n, and A1 is a fixed constant. In this embodiment, A1 is 0.951;

[0059] SS3: Calculate the phrase feature value Vj of the node phrase j in all subsets according to the method in step SS2;

[0060] Then, a feature threshold X2 is set, and the phrase feature value of the node phrase in each subset is compared with the feature threshold X2, and the node phrase j with Vj≥X2 is taken as the central phrase of this subset, wherein the specific value of the feature threshold X2 is obtained by a technician in this field after big data calculation;

[0061] It should be further explained that after the phrase feature value is compared with the feature threshold, there may be a sub-set without the central phrase. At this time, the sub-set without the central phrase is marked as the background set. At the same time, there may be a sub-set containing multiple central phrases.

[0062] SS4: Then compare the central phrases of each subset. If there are subsets with the same central phrase, merge the corresponding subsets and mark them as one-way sets. At the same time, merge the sub-information corresponding to each subset in the one-way set and obtain one-way information.

[0063] There is a one-way communication connection between the information integration module and the intellectual property analysis module;

[0064] The IP analysis module is used to identify and process one-way information and determine the IP risk coefficient. The specific method for determining the IP risk coefficient includes:

[0065] Randomly select a one-way information as the target information, obtain the central phrase in the target information, use the central phrase as the retrieval key information, search in the intellectual property information database on the Internet, and obtain the key matching information;

[0066] Obtain the one-way set corresponding to the target information, remove the central phrase in the one-way set, integrate the remaining information phrases in the one-way set, and mark them as a dimension set;

[0067] Then, the information phrases and key matching information in the dimension set are converted into vector form respectively, and based on the cosine similarity algorithm, the similarity value between any key matching information and the dimension set is calculated and marked as the intellectual property risk coefficient;

[0068] The information phrases and key matching information in the dimension set are converted into vector form. In this embodiment, the bag-of-words model algorithm is selected, and the bag-of-words model algorithm and the cosine similarity algorithm are both existing technologies, which will not be described in detail here;

[0069] Then, the intellectual property risk coefficient is compared with the risk threshold Fy, and the maximum value FXmax of the intellectual property risk coefficient in the key matching information is taken and compared with the risk threshold Fy. If FXmax≥Fy, a risk warning signal is generated; if FXmax<Fy, an intellectual property protection signal is generated, wherein the specific value of the risk threshold Fy is obtained by technicians in this field after big data calculation;

[0070] After that, there is a one-way communication connection between the intellectual property analysis module and the terminal display module;

[0071] The terminal display module is used to receive risk warning signals and intellectual property protection signals and display them on the terminal device. The staff then confirms the intellectual property decisions in the project through the terminal display module.

[0072] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A power project data processing, analysis and decision-making system, characterized in that: It includes an information analysis module, an engineering identification module, an information integration module, an intellectual property analysis module and a terminal display module which are connected in sequence; The information analysis module is used to identify the project engineering of the enterprise and take it as the target engineering. According to the project application instruction of the target engineering, the text information in the project application instruction is analyzed to obtain the project information set, and then the word frequency of the information phrases in the project information set is calculated to determine the project theme of the project engineering, wherein the project theme includes the project key carrier and the node phrase; The project identification module monitors and identifies the progress of target projects based on the project theme of the project engineering and obtains real-time project information; An information integration module is used to divide the real-time project information into several sub-information according to the title information, and to split the project information set into multiple sub-sets according to the sub-information, and then calculate the phrase feature value of the node phrase in the sub-set, and determine the central phrase in the sub-set based on the phrase feature value, and then merge the sub-sets with the same central phrase to obtain a unidirectional set and unidirectional information; The intellectual property analysis module is used to use the central phrase in the one-way information as the key information for retrieval and to perform retrieval to obtain key matching information, and at the same time process the one-way set corresponding to the one-phase information to obtain a dimension set, calculate the intellectual property risk coefficient of the dimension set and the key matching information, and generate a risk warning signal and an intellectual property protection signal based on the intellectual property risk coefficient; The terminal display module is used to receive risk warning signals and intellectual property protection signals and display them on the terminal device.

2. A power project data processing, analysis and decision-making system according to claim 1, characterized in that: Methods for determining project topics include: Identify the project application instructions in the collected information and obtain the engineering projects of the enterprise. The project application instructions include the project application form; Randomly select a project from the enterprise's engineering projects as the target project; Extract the project application instructions of the target project, use natural language processing technology to split the text information in the project application instructions into several information phrases, then identify the parts of speech in the information phrases, extract the content word parts of speech in the parts of speech, and summarize the information phrases corresponding to the content word parts of speech into a project information set; Calculate the word frequency FCi of each information phrase in the project information set, CXi represents the number of times information phrase i appears in the project information set, and ZL represents the total number of information phrases in the project information set, where i represents different information phrases in the project information set, and i∈[1,I], indicating that there are I different information phrases in the project information set; Arrange the word frequencies of information phrases i in descending order to obtain the first sequence of information phrases; Take the information phrase at the first position in the first sequence and mark it as the key carrier of the project; A first threshold X1 is set and used as a dividing line to select node phrases in the first sequence, wherein the node phrases are information phrases whose word frequency is greater than or equal to the first threshold X1.

3. A power project data processing, analysis and decision-making system according to claim 2, characterized in that: The action of collecting information is completed through the information collection module. The information collection module is used to collect all information in the enterprise and upload it to the enterprise cloud platform for information storage. There is a two-way communication connection between the enterprise cloud platform and the information analysis module.

4. A power project data processing, analysis and decision-making system according to claim 1, characterized in that: The calculation method of the phrase feature value of the node phrase includes: SS1: randomly select a project and obtain the real-time project information of the project, and divide the text information in the real-time project information into several sub-information according to the information title, where the information title is the title structure layer in the hierarchical relationship of the text information; SS2: Split the project information set into several sub-sets according to the information content in the sub-information, wherein the information phrases in each sub-set correspond to the information content in the sub-information; Arbitrarily select a subset as the target set, and mark the subsets other than the target set as the control set; The node phrases in the target set and the control set are identified respectively, and then the formula is used The phrase feature value Vj of node phrase j in the target set is calculated, where j represents different node phrases, n represents different control sets, and n∈[1,N], N represents the total number of control sets, Lj represents the number of times node phrase j appears in the target set, Mz represents the total number of information phrases in the target set, Djn represents the number of times node phrase j appears in control set n, Hn represents the total number of information phrases in control set n, and A1 is a fixed constant.

5. A power project data processing, analysis and decision-making system according to claim 4, characterized in that: Methods for obtaining one-way collections and one-way information include: Set the feature threshold X2, compare the phrase feature value of the node phrase in each subset with the feature threshold X2, and take the node phrase j with Vj≥X2 as the central phrase of this subset; The central phrases of each subset are compared. If there are subsets with the same central phrase, the corresponding subsets are merged and marked as unidirectional sets. At the same time, the sub-information corresponding to each subset in the unidirectional set is merged to obtain unidirectional information.

6. The power project data processing, analysis and decision-making system according to claim 1, characterized in that: Methods for obtaining IP protection signals include: Randomly select a one-way information as the target information, obtain the central phrase in the target information, use the central phrase as the retrieval key information, search in the intellectual property information database on the Internet, and obtain the key matching information; Obtain the one-way set corresponding to the target information, remove the central phrase in the one-way set, integrate the remaining information phrases in the one-way set, and mark them as a dimension set; Then, the information phrases and key matching information in the dimension set are converted into vector form respectively, and based on the cosine similarity algorithm, the similarity value between any key matching information and the dimension set is calculated and marked as the intellectual property risk coefficient; Compare the intellectual property risk coefficient with the risk threshold Fy, take the maximum value FXmax of the intellectual property risk coefficient in the key matching information, and compare it with the risk threshold Fy. If FXmax≥Fy, generate a risk warning signal; if FXmax<Fy, generate an intellectual property protection signal.

Images