Text processing method, device, equipment and storage medium
By generating and comparing project vectors and combining the similarity of successful application texts, the problem that enterprises find it difficult to identify the reasons for the application failure when applying for the project is solved, and the accurate prediction and optimization of the application success rate is achieved, reducing the risk of application failure.
Patent Information
- Application Number
- CN202210255932.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-15
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2042-03-15
AI Technical Summary
When corporate users apply for support funds for projects, it is difficult for them to accurately identify the specific reasons that cause the project to fail to pass the approval before applying, resulting in uncertainty and waste of resources in the application process.
By generating the target project vector and historical project vector of the project to be applied, the similarity is calculated and similar items are selected, and the declaration success rate is calculated based on the similarity of the successful declaration text, identifying the target paragraph that affects the declaration success rate, so as to predict and Optimize the application process.
It improves the accuracy of project application success rate, reduces the situation of failure after application, and helps enterprises make more informed decisions before applying, and avoid waste of resources.
Smart Images

Figure CN114637836B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a text processing method, device, equipment and storage medium. Background Art
[0002] At present, corporate users who apply for support funds for projects often need to carefully check the corporate information and the reporting requirements in the support policy, and only after completing the relevant application process can they obtain the approval results, and then learn the specific reasons why the final approval was not passed.
[0003] Therefore, before applying for a project, how to accurately identify the reasons why a project cannot be approved has become a technical problem that needs to be solved urgently. Summary of the invention
[0004] In view of the above, it is necessary to provide a text processing method, apparatus, device and storage medium that can accurately determine the project application success rate, and then accurately identify the target paragraphs that affect the project application success rate to be less than a certain threshold.
[0005] In one aspect, the present invention provides a text processing method, the text processing method comprising:
[0006] Obtain the project identification and project information of the project to be applied for;
[0007] Generate a target project vector for the project to be applied for according to the project identifier and the project information, wherein the target project vector is used to characterize the project to be applied for;
[0008] Acquire a historical item vector of a historical item, wherein the historical item vector is used to represent the historical item;
[0009] According to the project similarity between the target project vector and the historical project vector, the number of projects in the historical project that are greater than a first preset threshold is counted;
[0010] Extracting similar projects to the project to be applied for from the historical projects according to the number of projects;
[0011] Obtain the texts successfully submitted in the similar projects as the successfully submitted texts, and obtain the texts to be submitted;
[0012] Generate a success rate of the text to be applied for the project to be applied for based on the text similarity between the text to be applied and each successfully applied text;
[0013] If the declaration success rate is less than a preset success rate threshold, target paragraphs are screened out from the text to be declared based on the preset success rate threshold for marking.
[0014] According to a preferred embodiment of the present invention, generating the target project vector of the project to be applied for according to the project identifier and the project information includes:
[0015] Acquire an identification code corresponding to each identifier in the project identification from a first preset vector mapping table;
[0016] Concatenate the identification codes according to the order of each identifier in the item identification to obtain a first item vector;
[0017] Dividing the project information based on a preset dictionary to obtain multiple division paths and path segmentation of each division path;
[0018] Obtaining the dictionary weight of the path segmentation in the preset dictionary, and calculating the sum of the dictionary weights in each segmented path to obtain a path score;
[0019] Determine the path participle corresponding to the partition path with the largest path score as the project participle;
[0020] Cleaning the project segmentation words based on preset stop words to obtain target segmentation words;
[0021] Encoding the target word segmentation based on a second preset vector mapping table to obtain multiple word segmentation vectors;
[0022] splicing the multiple word segmentation vectors based on the position of the target word in the item information to obtain a second item vector;
[0023] A target item vector is generated according to the first item vector and the second item vector.
[0024] According to a preferred embodiment of the present invention, generating a target item vector according to the first item vector and the second item vector comprises:
[0025] detecting a dimension of the first item vector as a first dimension, and detecting a dimension of the second item vector as a second dimension;
[0026] If the first dimension is smaller than the second dimension, padding the first item vector with zeros until the first dimension is equal to the second dimension, thereby obtaining a processed first item vector;
[0027] The sum of each element in the processed first item vector and each element in the second item vector is calculated to obtain the target item vector.
[0028] According to a preferred embodiment of the present invention, generating the application success rate of the text to be applied for the project to be applied for according to the text similarity between the text to be applied and each successfully applied text comprises:
[0029] Encoding the text to be declared based on the second preset vector mapping table to obtain a first text vector, and encoding each successfully declared text based on the second preset vector mapping table to obtain a second text vector;
[0030] Calculating the similarity between each second text vector and the first text vector to obtain a plurality of text similarities;
[0031] The average value of the multiple text similarities is calculated to obtain the declaration success rate.
[0032] According to a preferred embodiment of the present invention, the step of selecting target paragraphs from the text to be reported based on the preset success rate threshold and marking the target paragraphs includes:
[0033] Calculate the difference between each text similarity and the declaration success rate;
[0034] Selecting the texts whose difference values are within the preset interval from the successfully declared texts as target texts, and extracting the vector corresponding to the target text from the second text vector as the target text vector;
[0035] Performing text segmentation on the text to be reported to obtain a plurality of text paragraphs, and encoding each text paragraph based on the second preset vector mapping table to obtain a paragraph vector;
[0036] Calculating the similarity between the target text vector and each paragraph vector to obtain vector similarity;
[0037] The text paragraph whose vector similarity is less than the preset success rate threshold is determined as the target paragraph, and the target paragraph is marked.
[0038] According to a preferred embodiment of the present invention, extracting similar projects of the project to be applied for from the historical projects according to the number of projects comprises:
[0039] If the number of items is less than a preset number threshold, the historical items corresponding to the item similarities having values greater than a second preset threshold are determined as the similar items, and the second preset threshold is less than the first preset threshold.
[0040] According to a preferred embodiment of the present invention, obtaining the text of the successful application in the similar project as the successful application text includes:
[0041] Obtaining an initial declaration text corresponding to the similar project from a preset text library;
[0042] Obtaining the declaration result of the initial declaration text;
[0043] The initial declaration text whose declaration result is successful declaration is determined as the successful declaration text.
[0044] On the other hand, the present invention further provides a text processing device, the text processing device comprising:
[0045] An acquisition unit, used to acquire the project identification and project information of the project to be applied for;
[0046] A generating unit, configured to generate a target project vector of the project to be applied for according to the project identifier and the project information, wherein the target project vector is used to characterize the project to be applied for;
[0047] The acquisition unit is further used to acquire a historical item vector of a historical item, wherein the historical item vector is used to represent the historical item;
[0048] A calculation unit, configured to count the number of items in the historical items that are greater than a first preset threshold value according to the item similarity between the target item vector and the historical item vector;
[0049] An extraction unit, used for extracting similar projects of the project to be applied for from the historical projects according to the number of projects;
[0050] The acquisition unit is further used to acquire the text of the similar project that has been successfully applied for as the successfully applied text, and acquire the text to be applied for;
[0051] The generating unit is further used to generate a success rate of application of the text to be applied for the project to be applied for based on the text similarity between the text to be applied and each successfully applied text;
[0052] The marking unit is used to select target paragraphs from the text to be reported based on the preset success rate threshold for marking if the reporting success rate is less than the preset success rate threshold.
[0053] On the other hand, the present invention further provides an electronic device, comprising:
[0054] a memory storing computer readable instructions; and
[0055] The processor executes the computer-readable instructions stored in the memory to implement the text processing method.
[0056] On the other hand, the present invention further proposes a computer-readable storage medium, in which computer-readable instructions are stored. The computer-readable instructions are executed by a processor in an electronic device to implement the text processing method.
[0057] It can be seen from the above technical solutions that the present invention can accurately generate a target project vector that represents the project to be applied for through the project identification and the project information, and then combine the target project vector and the historical project vector to screen out the similar projects from the same dimension, thereby improving the accuracy of the similar projects. By calculating the text similarity between the text to be applied for and the successfully applied text in the similar project, the application success rate of the text to be applied for the project to be applied for can be accurately determined, and then when the application success rate is less than the preset success rate threshold, the target paragraphs that affect the project application success rate to be less than a certain threshold can be accurately determined and marked before the project is applied for. In addition, the present invention can reasonably choose whether to apply for the project to be applied for through the application success rate, thereby avoiding the situation where the application fails to pass the approval to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] Figure 1 It is a flow chart of a preferred embodiment of the text processing method of the present invention.
[0059] Figure 2 It is a functional module diagram of a preferred embodiment of the text processing device of the present invention.
[0060] Figure 3 It is a schematic diagram of the structure of an electronic device of a preferred embodiment of the text processing method of the present invention. DETAILED DESCRIPTION
[0061] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.
[0062] like Figure 1 FIG. 1 is a flowchart of a preferred embodiment of the text processing method of the present invention. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted.
[0063] The text processing method can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.
[0064] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.
[0065] The text processing method is applied to one or more electronic devices, which are devices that can automatically perform numerical calculations and / or information processing according to pre-set or stored computer-readable instructions, and whose hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.
[0066] The electronic device may be any electronic product that can perform human-computer interaction with a user, such as a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), a game console, an interactive network television (IPTV), a smart wearable device, etc.
[0067] The electronic device may include a network device and / or a user device. The network device includes, but is not limited to, a single network electronic device, an electronic device group consisting of multiple network electronic devices, or a cloud consisting of a large number of hosts or network electronic devices based on cloud computing.
[0068] The network where the electronic device is located includes, but is not limited to: the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.
[0069] S10, obtaining the project identification and project information of the project to be applied for.
[0070] In at least one embodiment of the present invention, the project to be applied for refers to a project that requires funding application.
[0071] The project identifier refers to a specific label used to identify the project to be applied for.
[0072] The project information refers to the policy information related to the project to be applied for.
[0073] In at least one embodiment of the present invention, the electronic device obtaining the project identification and project information of the project to be applied for includes:
[0074] The project identifier and the project information are obtained from a preset website based on the project to be applied for.
[0075] S11, generating a target project vector of the project to be applied for according to the project identifier and the project information, wherein the target project vector is used to characterize the project to be applied for.
[0076] In at least one embodiment of the present invention, the electronic device generating a target project vector of the project to be applied for according to the project identifier and the project information includes:
[0077] Acquire an identification code corresponding to each identifier in the project identification from a first preset vector mapping table;
[0078] Concatenate the identification codes according to the order of each identifier in the item identification to obtain a first item vector;
[0079] Dividing the project information based on a preset dictionary to obtain multiple division paths and path segmentation of each division path;
[0080] Obtaining the dictionary weight of the path segmentation in the preset dictionary, and calculating the sum of the dictionary weights in each segmented path to obtain a path score;
[0081] Determine the path participle corresponding to the partition path with the largest path score as the project participle;
[0082] Cleaning the project segmentation words based on preset stop words to obtain target segmentation words;
[0083] Encoding the target word segmentation based on a second preset vector mapping table to obtain multiple word segmentation vectors;
[0084] splicing the multiple word segmentation vectors based on the position of the target word in the item information to obtain a second item vector;
[0085] A target item vector is generated according to the first item vector and the second item vector.
[0086] The preset dictionary stores a plurality of mapping relationships between words and weights.
[0087] The first preset vector mapping table stores a mapping relationship between identifiers and vectors, and the second preset vector mapping table stores a mapping relationship between words and vectors.
[0088] The preset stop words can be obtained from stop word lists in major websites.
[0089] The target word segmentation refers to the item word segmentation that does not contain the preset stop words.
[0090] By dividing the project information through the preset dictionary, the accuracy of the project segmentation can be improved. By cleaning the project segmentation, words irrelevant to the project to be applied for can be eliminated, thereby improving the representation ability of the second project vector for the project information. Furthermore, by combining the first project vector and the second project vector, the representation ability of the target project vector for the project to be applied for can be improved.
[0091] Specifically, the electronic device generating a target item vector according to the first item vector and the second item vector includes:
[0092] detecting a dimension of the first item vector as a first dimension, and detecting a dimension of the second item vector as a second dimension;
[0093] If the first dimension is smaller than the second dimension, padding the first item vector with zeros until the first dimension is equal to the second dimension, thereby obtaining a processed first item vector;
[0094] The sum of each element in the processed first item vector and each element in the second item vector is calculated to obtain the target item vector.
[0095] Through the above implementation, it is possible to ensure that the first item vector and the second item vector can be successfully combined, thereby improving the generation accuracy of the target item vector.
[0096] In other embodiments, the processed first item vector and the second item vector are concatenated to obtain the target item vector.
[0097] Through the above implementation, the target item vector can be quickly generated.
[0098] S12, obtaining a historical item vector of the historical item, wherein the historical item vector is used to represent the historical item.
[0099] In at least one embodiment of the present invention, the historical project refers to a policy project different from the project to be applied for.
[0100] The generation method of the historical item vector is similar to the generation method of the target item vector, and the present invention will not elaborate on this.
[0101] In at least one embodiment of the present invention, the electronic device obtains the historical item vector from a preset vector library based on the historical item.
[0102] S13, counting the number of items in the historical items that are greater than a first preset threshold value according to the item similarity between the target item vector and the historical item vector.
[0103] In at least one embodiment of the present invention, the number of items refers to the number of historical items corresponding to the item similarity whose item similarity value is greater than a first preset threshold.
[0104] The item similarity refers to the cosine value of the historical item vector and the target item vector.
[0105] S14, extracting similar projects to the project to be applied for from the historical projects according to the number of projects.
[0106] In at least one embodiment of the present invention, the similar project refers to a historical project similar to the project to be applied for.
[0107] In at least one embodiment of the present invention, the electronic device extracting similar items to the item to be applied for from the historical items according to the number of items comprises:
[0108] If the number of items is less than a preset number threshold, the historical items corresponding to the item similarities having values greater than a second preset threshold are determined as the similar items, and the second preset threshold is less than the first preset threshold.
[0109] The preset quantity threshold may be set according to actual needs. For example, the preset quantity threshold may be 5.
[0110] The first preset threshold and the second preset threshold can be set according to requirements.
[0111] Through the above implementation, more historical projects can be determined as similar projects on the premise of ensuring that the similar projects are sufficiently similar to the projects to be applied for, thereby making the application success rate obtained by subsequent calculations more accurate.
[0112] S15, obtaining the text successfully submitted in the similar project as the successfully submitted text, and obtaining the text to be submitted.
[0113] In at least one embodiment of the present invention, the document to be reported refers to the reporting document corresponding to the project to be applied for.
[0114] In at least one embodiment of the present invention, the electronic device obtaining the text of successful declaration in the similar project as the successful declaration text includes:
[0115] Obtaining an initial declaration text corresponding to the similar project from a preset text library;
[0116] Obtaining the declaration result of the initial declaration text;
[0117] The initial declaration text whose declaration result is successful declaration is determined as the successful declaration text.
[0118] The preset text library stores a plurality of declaration texts related to historical projects and the declaration results of the declaration texts.
[0119] Through the above implementation, the efficiency of obtaining the successful declaration text can be improved.
[0120] In at least one embodiment of the present invention, the electronic device obtains any text corresponding to the project to be applied for from a preset path as the text to be reported.
[0121] S16, generating a success rate of application of the text to be applied for the project to be applied for according to the text similarity between the text to be applied for and each successfully applied text.
[0122] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned declaration success rate, the above-mentioned declaration success rate can also be stored in a blockchain node.
[0123] In at least one embodiment of the present invention, the electronic device generates the application success rate of the text to be applied for the project to be applied for according to the text similarity between the text to be applied and each successfully applied text, including:
[0124] Encoding the text to be declared based on the second preset vector mapping table to obtain a first text vector, and encoding each successfully declared text based on the second preset vector mapping table to obtain a second text vector;
[0125] Calculating the similarity between each second text vector and the first text vector to obtain a plurality of text similarities;
[0126] The average value of the multiple text similarities is calculated to obtain the declaration success rate.
[0127] The text similarity refers to the cosine value between the second text vector and the first text vector.
[0128] By encoding the text to be declared and the successfully declared text through the second preset vector mapping table, the text similarity can be analyzed from the same dimension, thereby improving the accuracy of the text similarity. In addition, by analyzing the declaration success rate based on multiple successfully declared texts, the accuracy of the declaration success rate can be improved.
[0129] S17, if the application success rate is less than a preset success rate threshold, target paragraphs are screened out from the text to be applied for marking based on the preset success rate threshold.
[0130] In at least one embodiment of the present invention, the target paragraph refers to a text paragraph that affects the declaration success rate to be less than a preset success rate threshold.
[0131] In at least one embodiment of the present invention, the step of selecting target paragraphs from the text to be reported based on the preset success rate threshold and marking the target paragraphs includes:
[0132] Calculate the difference between each text similarity and the declaration success rate;
[0133] Selecting the texts whose difference values are within the preset interval from the successfully declared texts as target texts, and extracting the vector corresponding to the target text from the second text vector as the target text vector;
[0134] Performing text segmentation on the text to be reported to obtain a plurality of text paragraphs, and encoding each text paragraph based on the second preset vector mapping table to obtain a paragraph vector;
[0135] Calculating the similarity between the target text vector and each paragraph vector to obtain vector similarity;
[0136] The text paragraphs whose vector similarity is less than the preset success rate threshold are marked.
[0137] The preset success rate threshold and the preset interval may be set according to actual needs. For example, the preset success rate threshold may be 60%, and the preset interval may be [0.25, 0.35].
[0138] Through the above implementation, the text paragraphs that affect the declaration success rate to be less than the preset success rate threshold can be accurately located, and by marking the text paragraphs, the user can quickly learn the specific reasons why the declaration cannot be successful.
[0139] It can be seen from the above technical solutions that the present invention can accurately generate a target project vector that represents the project to be applied for by calculating the project information through the project identification, and then combine the target project vector and the historical project vector to screen out the similar projects from the same dimension, thereby improving the accuracy of the similar projects. By calculating the text similarity between the text to be applied for and the successfully applied text in the similar projects, the application success rate of the text to be applied for the project to be applied for can be accurately determined, and then when the application success rate is less than the preset success rate threshold, the target paragraphs that affect the project application success rate to be less than a certain threshold can be accurately determined and marked before the project is applied for. In addition, the present invention can reasonably choose whether to apply for the project to be applied for through the application success rate, thereby avoiding the situation where the application fails to pass the approval to a certain extent.
[0140] like Figure 2 , which is a functional module diagram of a preferred embodiment of the text processing device of the present invention. The text processing device 11 includes an acquisition unit 110, a generation unit 111, an extraction unit 112, a calculation unit 113 and an identification unit 114. The module / unit referred to in the present invention refers to a series of computer-readable instruction segments that can be acquired by the processor 13 and can perform fixed functions, which are stored in the memory 12. In this embodiment, the functions of each module / unit will be described in detail in subsequent embodiments.
[0141] The acquisition unit 110 acquires the project identification and project information of the project to be applied for.
[0142] In at least one embodiment of the present invention, the project to be applied for refers to a project that requires funding application.
[0143] The project identifier refers to a specific label used to identify the project to be applied for.
[0144] The project information refers to the policy information related to the project to be applied for.
[0145] In at least one embodiment of the present invention, the acquiring unit 110 acquires the project identification and project information of the project to be applied for, including:
[0146] The project identifier and the project information are obtained from a preset website based on the project to be applied for.
[0147] The generating unit 111 generates a target project vector of the project to be applied for according to the project identifier and the project information, wherein the target project vector is used to represent the project to be applied for.
[0148] In at least one embodiment of the present invention, the generating unit 111 generates the target project vector of the project to be applied for according to the project identifier and the project information, including:
[0149] Acquire an identification code corresponding to each identifier in the project identification from a first preset vector mapping table;
[0150] Concatenate the identification codes according to the order of each identifier in the item identification to obtain a first item vector;
[0151] Dividing the project information based on a preset dictionary to obtain multiple division paths and path segmentation of each division path;
[0152] Obtaining the dictionary weight of the path segmentation in the preset dictionary, and calculating the sum of the dictionary weights in each segmented path to obtain a path score;
[0153] Determine the path participle corresponding to the partition path with the largest path score as the project participle;
[0154] Cleaning the project segmentation words based on preset stop words to obtain target segmentation words;
[0155] Encoding the target word segmentation based on a second preset vector mapping table to obtain multiple word segmentation vectors;
[0156] splicing the multiple word segmentation vectors based on the position of the target word in the item information to obtain a second item vector;
[0157] A target item vector is generated according to the first item vector and the second item vector.
[0158] The preset dictionary stores a plurality of mapping relationships between words and weights.
[0159] The first preset vector mapping table stores a mapping relationship between identifiers and vectors, and the second preset vector mapping table stores a mapping relationship between words and vectors.
[0160] The preset stop words can be obtained from stop word lists in major websites.
[0161] The target word segmentation refers to the item word segmentation that does not contain the preset stop words.
[0162] By dividing the project information through the preset dictionary, the accuracy of the project segmentation can be improved. By cleaning the project segmentation, words irrelevant to the project to be applied for can be eliminated, thereby improving the representation ability of the second project vector for the project information. Furthermore, by combining the first project vector and the second project vector, the representation ability of the target project vector for the project to be applied for can be improved.
[0163] Specifically, the generating unit 111 generates a target item vector according to the first item vector and the second item vector, including:
[0164] detecting a dimension of the first item vector as a first dimension, and detecting a dimension of the second item vector as a second dimension;
[0165] If the first dimension is smaller than the second dimension, padding the first item vector with zeros until the first dimension is equal to the second dimension, thereby obtaining a processed first item vector;
[0166] The sum of each element in the processed first item vector and each element in the second item vector is calculated to obtain the target item vector.
[0167] Through the above implementation, it is possible to ensure that the first item vector and the second item vector can be successfully combined, thereby improving the generation accuracy of the target item vector.
[0168] In other embodiments, the processed first item vector and the second item vector are concatenated to obtain the target item vector.
[0169] Through the above implementation, the target item vector can be quickly generated.
[0170] The acquisition unit 110 acquires a historical item vector of a historical item, where the historical item vector is used to represent the historical item.
[0171] In at least one embodiment of the present invention, the historical project refers to a policy project different from the project to be applied for.
[0172] The generation method of the historical item vector is similar to the generation method of the target item vector, and the present invention will not elaborate on this.
[0173] In at least one embodiment of the present invention, the acquisition unit 110 acquires the historical item vector from a preset vector library based on the historical item.
[0174] The calculation unit 113 counts the number of items in the historical items that are greater than a first preset threshold according to the item similarity between the target item vector and the historical item vector.
[0175] In at least one embodiment of the present invention, the number of items refers to the number of historical items corresponding to the item similarity whose item similarity value is greater than a first preset threshold.
[0176] The item similarity refers to the cosine value of the historical item vector and the target item vector.
[0177] The extracting unit 112 extracts similar projects of the project to be applied for from the historical projects according to the number of projects.
[0178] In at least one embodiment of the present invention, the similar project refers to a historical project similar to the project to be applied for.
[0179] In at least one embodiment of the present invention, the extracting unit 112 extracts similar projects of the project to be applied for from the historical projects according to the number of projects, including:
[0180] If the number of items is less than a preset number threshold, the historical items corresponding to the item similarities having values greater than a second preset threshold are determined as the similar items, and the second preset threshold is less than the first preset threshold.
[0181] The preset quantity threshold may be set according to actual needs. For example, the preset quantity threshold may be 5.
[0182] The first preset threshold and the second preset threshold can be set according to requirements.
[0183] Through the above implementation, more historical projects can be determined as similar projects on the premise of ensuring that the similar projects are sufficiently similar to the projects to be applied for, thereby making the application success rate obtained by subsequent calculations more accurate.
[0184] The acquisition unit 110 acquires the successfully declared texts in the similar projects as successfully declared texts, and acquires the texts to be declared.
[0185] In at least one embodiment of the present invention, the document to be reported refers to the reporting document corresponding to the project to be applied for.
[0186] In at least one embodiment of the present invention, the acquiring unit 110 acquires the text of the successful application in the similar project as the successful application text, including:
[0187] Obtaining an initial declaration text corresponding to the similar project from a preset text library;
[0188] Obtaining the declaration result of the initial declaration text;
[0189] The initial declaration text whose declaration result is successful declaration is determined as the successful declaration text.
[0190] The preset text library stores a plurality of declaration texts related to historical projects and the declaration results of the declaration texts.
[0191] Through the above implementation, the efficiency of obtaining the successful declaration text can be improved.
[0192] In at least one embodiment of the present invention, the acquisition unit 110 acquires any text corresponding to the project to be applied for from a preset path as the text to be reported.
[0193] The generating unit 111 generates a success rate of application of the text to be applied for the project to be applied for based on the text similarity between the text to be applied for and each successfully applied text.
[0194] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned declaration success rate, the above-mentioned declaration success rate can also be stored in a blockchain node.
[0195] In at least one embodiment of the present invention, the generating unit 111 generates the application success rate of the to-be-approved text for the to-be-approved project according to the text similarity between the to-be-approved text and each successfully-approved text, including:
[0196] Encoding the text to be declared based on the second preset vector mapping table to obtain a first text vector, and encoding each successfully declared text based on the second preset vector mapping table to obtain a second text vector;
[0197] Calculating the similarity between each second text vector and the first text vector to obtain a plurality of text similarities;
[0198] The average value of the multiple text similarities is calculated to obtain the declaration success rate.
[0199] The text similarity refers to the cosine value between the second text vector and the first text vector.
[0200] By encoding the text to be declared and the successfully declared text through the second preset vector mapping table, the text similarity can be analyzed from the same dimension, thereby improving the accuracy of the text similarity. In addition, by analyzing the declaration success rate based on multiple successfully declared texts, the accuracy of the declaration success rate can be improved.
[0201] If the declaration success rate is less than a preset success rate threshold, the marking unit 114 selects target paragraphs from the text to be declared based on the preset success rate threshold for marking.
[0202] In at least one embodiment of the present invention, the identifying unit 114 selects target paragraphs from the text to be reported based on the preset success rate threshold and identifies the target paragraphs, including:
[0203] Calculate the difference between each text similarity and the declaration success rate;
[0204] Selecting the texts whose difference values are within the preset interval from the successfully declared texts as target texts, and extracting the vector corresponding to the target text from the second text vector as the target text vector;
[0205] Performing text segmentation on the text to be reported to obtain a plurality of text paragraphs, and encoding each text paragraph based on the second preset vector mapping table to obtain a paragraph vector;
[0206] Calculating the similarity between the target text vector and each paragraph vector to obtain vector similarity;
[0207] The text paragraphs whose vector similarity is less than the preset success rate threshold are marked.
[0208] The preset success rate threshold and the preset interval may be set according to actual needs. For example, the preset success rate threshold may be 60%, and the preset interval may be [0.25, 0.35].
[0209] Through the above implementation, the text paragraphs that affect the declaration success rate to be less than the preset success rate threshold can be accurately located, and by marking the text paragraphs, the user can quickly learn the specific reasons why the declaration cannot be successful.
[0210] It can be seen from the above technical solutions that the present invention can accurately generate a target project vector that represents the project to be applied for by calculating the project information through the project identification, and then combine the target project vector and the historical project vector to screen out the similar projects from the same dimension, thereby improving the accuracy of the similar projects. By calculating the text similarity between the text to be applied for and the successfully applied text in the similar projects, the application success rate of the text to be applied for the project to be applied for can be accurately determined, and then when the application success rate is less than the preset success rate threshold, the target paragraphs that affect the project application success rate to be less than a certain threshold can be accurately determined and marked before the project is applied for. In addition, the present invention can reasonably choose whether to apply for the project to be applied for through the application success rate, thereby avoiding the situation where the application fails to pass the approval to a certain extent.
[0211] like Figure 3 FIG. 1 is a schematic diagram of the structure of an electronic device of a preferred embodiment of the text processing method of the present invention.
[0212] In one embodiment of the present invention, the electronic device 1 includes, but is not limited to, a memory 12, a processor 13, and computer-readable instructions stored in the memory 12 and executable on the processor 13, such as a text processing program.
[0213] Those skilled in the art will understand that the schematic diagram is merely an example of the electronic device 1 and does not constitute a limitation on the electronic device 1. The electronic device 1 may include more or fewer components than shown in the diagram, or a combination of certain components, or different components. For example, the electronic device 1 may also include input and output devices, network access devices, buses, etc.
[0214] The processor 13 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 13 is the computing core and control center of the electronic device 1, and uses various interfaces and lines to connect various parts of the entire electronic device 1, and executes the operating system of the electronic device 1 and various installed applications, program codes, etc.
[0215] Exemplarily, the computer-readable instructions may be divided into one or more modules / units, which are stored in the memory 12 and executed by the processor 13 to complete the present invention. The one or more modules / units may be a series of computer-readable instruction segments capable of completing specific functions, which are used to describe the execution process of the computer-readable instructions in the electronic device 1. For example, the computer-readable instructions may be divided into an acquisition unit 110, a generation unit 111, an extraction unit 112, a calculation unit 113, and an identification unit 114.
[0216] The memory 12 can be used to store the computer-readable instructions and / or modules, and the processor 13 implements various functions of the electronic device 1 by running or executing the computer-readable instructions and / or modules stored in the memory 12, and calling the data stored in the memory 12. The memory 12 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the electronic device, etc. The memory 12 can include non-volatile and volatile memories, such as: a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other storage devices.
[0217] The memory 12 may be an external memory and / or an internal memory of the electronic device 1. Furthermore, the memory 12 may be a memory in a physical form, such as a memory stick, a TF card (Trans-flash Card), and the like.
[0218] If the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also instruct the relevant hardware to complete it through computer-readable instructions. The computer-readable instructions can be stored in a computer-readable storage medium. When the computer-readable instructions are executed by the processor, the steps of the above-mentioned various method embodiments can be implemented.
[0219] The computer-readable instructions include computer-readable instruction codes, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer-readable instruction code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM).
[0220] The blockchain referred to in the present invention is a new application model of computer technologies such as distributed text processing, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the blockchain underlying platform, platform product service layer, and application service layer.
[0221] Combination Figure 1 The memory 12 in the electronic device 1 stores computer-readable instructions to implement a text processing method, and the processor 13 can execute the computer-readable instructions to implement:
[0222] Obtain the project identification and project information of the project to be applied for;
[0223] Generate a target project vector for the project to be applied for according to the project identifier and the project information, wherein the target project vector is used to characterize the project to be applied for;
[0224] Acquire a historical item vector of a historical item, wherein the historical item vector is used to represent the historical item;
[0225] According to the project similarity between the target project vector and the historical project vector, the number of projects in the historical project that are greater than a first preset threshold is counted;
[0226] Extracting similar projects to the project to be applied for from the historical projects according to the number of projects;
[0227] Obtain the texts successfully submitted in the similar projects as the successfully submitted texts, and obtain the texts to be submitted;
[0228] Generate a success rate of the text to be applied for the project to be applied for based on the text similarity between the text to be applied and each successfully applied text;
[0229] If the declaration success rate is less than a preset success rate threshold, target paragraphs are screened out from the text to be declared based on the preset success rate threshold for marking.
[0230] Specifically, the specific implementation method of the processor 13 for the above-mentioned computer readable instructions can refer to Figure 1 The description of the relevant steps in the corresponding embodiments will not be repeated here.
[0231] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.
[0232] The computer-readable storage medium stores computer-readable instructions, wherein the computer-readable instructions are used to implement the following steps when executed by the processor 13:
[0233] Obtain the project identification and project information of the project to be applied for;
[0234] Generate a target project vector for the project to be applied for according to the project identifier and the project information, wherein the target project vector is used to characterize the project to be applied for;
[0235] Acquire a historical item vector of a historical item, wherein the historical item vector is used to represent the historical item;
[0236] According to the project similarity between the target project vector and the historical project vector, the number of projects in the historical project that are greater than a first preset threshold is counted;
[0237] Extracting similar projects to the project to be applied for from the historical projects according to the number of projects;
[0238] Obtain the texts successfully submitted in the similar projects as the successfully submitted texts, and obtain the texts to be submitted;
[0239] Generate a success rate of the text to be applied for the project to be applied for based on the text similarity between the text to be applied and each successfully applied text;
[0240] If the declaration success rate is less than a preset success rate threshold, target paragraphs are screened out from the text to be declared based on the preset success rate threshold for marking.
[0241] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0242] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0243] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.
[0244] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or devices mentioned can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.
[0245] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.
Claims
1. A text processing method, characterized in that: The text processing method comprises: Obtain the project identification and project information of the project to be applied for; Generate a target project vector of the project to be applied for according to the project identification and the project information, including: obtaining an identification code corresponding to each identifier in the project identification from a first preset vector mapping table; splicing the identification codes according to the order of each identifier in the project identification to obtain a first project vector; dividing the project information based on a preset dictionary to obtain multiple division paths and path segmentation words for each division path; obtaining the dictionary weight of the path segmentation word in the preset dictionary, and calculating the sum of the dictionary weights in each division path to obtain a path score; determining the path segmentation word corresponding to the division path with the largest path score as the project segmentation word; cleaning the project segmentation word based on preset stop words to obtain a target segmentation word; encoding the target segmentation word based on a second preset vector mapping table to obtain multiple segmentation vectors; splicing the multiple segmentation vectors based on the position of the target segmentation word in the project information to obtain a second project vector; generating a target project vector according to the first project vector and the second project vector, and the target project vector is used to characterize the project to be applied for; Acquire a historical item vector of a historical item, wherein the historical item vector is used to represent the historical item; According to the project similarity between the target project vector and the historical project vector, the number of projects in the historical project that are greater than a first preset threshold is counted; Extracting similar projects to the project to be applied for from the historical projects according to the number of projects; Obtain the texts successfully submitted in the similar projects as the successfully submitted texts, and obtain the texts to be submitted; Generate a success rate of the text to be applied for the project to be applied for based on the text similarity between the text to be applied and each successfully applied text; If the declaration success rate is less than a preset success rate threshold, target paragraphs are screened out from the text to be declared based on the preset success rate threshold for marking.
2. The text processing method according to claim 1, characterized in that: Generating a target item vector according to the first item vector and the second item vector comprises: detecting a dimension of the first item vector as a first dimension, and detecting a dimension of the second item vector as a second dimension; If the first dimension is smaller than the second dimension, padding the first item vector with zeros until the first dimension is equal to the second dimension, thereby obtaining a processed first item vector; The sum of each element in the processed first item vector and each element in the second item vector is calculated to obtain the target item vector.
3. The text processing method according to claim 1, characterized in that: The step of generating the application success rate of the text to be applied for the project to be applied for according to the text similarity between the text to be applied and each successfully applied text comprises: Encoding the text to be declared based on the second preset vector mapping table to obtain a first text vector, and encoding each successfully declared text based on the second preset vector mapping table to obtain a second text vector; Calculating the similarity between each second text vector and the first text vector to obtain a plurality of text similarities; Calculate the average value of the similarities of multiple texts to obtain the declaration success rate.
4. The text processing method according to claim 3, characterized in that: The step of selecting target paragraphs from the text to be reported based on the preset success rate threshold and marking the target paragraphs includes: Calculate the difference between each text similarity and the declaration success rate; Selecting the texts whose difference values are within the preset interval from the successfully declared texts as target texts, and extracting the vector corresponding to the target text from the second text vector as the target text vector; Performing text segmentation on the text to be reported to obtain a plurality of text paragraphs, and encoding each text paragraph based on the second preset vector mapping table to obtain a paragraph vector; Calculating the similarity between the target text vector and each paragraph vector to obtain vector similarity; The text paragraph whose vector similarity is less than the preset success rate threshold is determined as the target paragraph, and the target paragraph is marked.
5. The text processing method according to claim 1, characterized in that: The extracting similar projects of the project to be applied for from the historical projects according to the number of projects comprises: If the number of items is less than a preset number threshold, the historical items corresponding to the item similarities having values greater than a second preset threshold are determined as the similar items, and the second preset threshold is less than the first preset threshold.
6. The text processing method according to claim 1, characterized in that: The obtaining of the successfully declared text in the similar project as the successfully declared text comprises: Obtaining an initial declaration text corresponding to the similar project from a preset text library; Obtaining the declaration result of the initial declaration text; The initial declaration text whose declaration result is successful declaration is determined as the successful declaration text.
7. A text processing device, characterized in that: The text processing device comprises: An acquisition unit, used to acquire the project identification and project information of the project to be applied for; A generating unit, for generating a target project vector of the project to be applied for according to the project identification and the project information, comprising: obtaining an identification code corresponding to each identifier in the project identification from a first preset vector mapping table; splicing the identification codes according to the order of each identifier in the project identification to obtain a first project vector; dividing the project information based on a preset dictionary to obtain a plurality of division paths and path segmentation words of each division path; obtaining a dictionary weight of the path segmentation word in the preset dictionary, and calculating the sum of the dictionary weights in each division path to obtain a path score; determining the path segmentation word corresponding to the division path with the largest path score as the project segmentation word; cleaning the project segmentation word based on preset stop words to obtain a target segmentation word; encoding the target segmentation word based on a second preset vector mapping table to obtain a plurality of segmentation vectors; splicing the plurality of segmentation vectors based on the position of the target segmentation word in the project information to obtain a second project vector; generating a target project vector according to the first project vector and the second project vector, the target project vector being used to characterize the project to be applied for; The acquisition unit is further used to acquire a historical item vector of a historical item, wherein the historical item vector is used to represent the historical item; A calculation unit, configured to count the number of items in the historical items that are greater than a first preset threshold value according to the item similarity between the target item vector and the historical item vector; An extraction unit, used for extracting similar projects of the project to be applied for from the historical projects according to the number of projects; The acquisition unit is further used to acquire the text of the similar project that has been successfully applied for as the successfully applied text, and acquire the text to be applied for; The generating unit is further used to generate a success rate of application of the text to be applied for the project to be applied for based on the text similarity between the text to be applied and each successfully applied text; The marking unit is used to select target paragraphs from the text to be reported based on the preset success rate threshold for marking if the reporting success rate is less than the preset success rate threshold.
8. An electronic device, characterized in that: The electronic device comprises: a memory storing computer-readable instructions; and A processor executes the computer-readable instructions stored in the memory to implement the text processing method according to any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions are executed by a processor in an electronic device to implement the text processing method according to any one of claims 1 to 6.
Citation Information
Patent Citations
Intelligent project recommendation method, electronic device and storage medium
CN110163476A
Archive data-oriented talent and policy intelligent matching system and method
CN110457696A
Project review method, project review device, equipment and storage medium
CN112381458A
Method and apparatus for analysis and matching of resume data information, electronic device, and medium
WO2021174919A1
Knowledge extraction method, apparatus, electronic device, and storage medium
WO2021212682A1
