A method for evaluating the quality of academic assets

Abstract

The present invention discloses a method for evaluating the quality of academic assets, including: obtaining basic information of academic assets; determining the use of full sampling or double-blind sampling methods according to the basic information of academic assets, and determining the academic asset sample set to be evaluated in the academic assets; determining the index system for the quality evaluation of academic assets; determining the scoring calculation model; calculating the relevant statistics of the quality evaluation of academic assets; calculating the corresponding quality evaluation indicators according to the relevant statistics; calculating the score of each quality evaluation indicator and the total quality evaluation score according to the quality evaluation indicators and the scoring calculation model; generating an academic asset quality evaluation report. The present invention designs an academic asset quality evaluation index system from multiple dimensions of completeness, accuracy, consistency and accessibility, and can quantitatively and accurately evaluate the quality of unstructured academic data assets, providing solid quality assurance for application scenarios such as academic asset inclusion and academic asset transactions.

Description

Technical Field

[0001] The present application relates to the field of data asset quality evaluation, and in particular to a method for evaluating academic asset quality. Background Art

[0002] With the rapid development of technologies such as big data, cloud computing, artificial intelligence, and blockchain, data has become a new production factor and strategic asset, which has enhanced the efficiency and capacity of technological innovation and industrial upgrading. The "White Paper on Data Asset Management Practice" released by the China Academy of Information and Communications Technology defines data assets as data resources that are owned or controlled by enterprises and can bring future economic benefits to enterprises, recorded in physical or electronic form, such as documents, electronic data, etc. Academic data assets such as journals and conference papers (hereinafter referred to as academic assets) contain huge social and economic value. How to evaluate, use, and give full play to the value of academic assets has become a concern for enterprises and many scholars. The China Asset Appraisal Association pointed out in the "Guidance on Data Asset Evaluation (Draft for Comments)" that data quality evaluation is the basis for data asset value evaluation, and the data quality level directly determines the data value and application effect. Therefore, accurate and reliable academic asset quality evaluation methods are of great significance to the healthy circulation and value realization of academic assets in the data factor market and to bring economic benefits to enterprises.

[0003] Among the related technologies, academic asset quality evaluation technology is mainly oriented to structured data. Structured data usually exists in the form of database tables, has a fixed format and mode, and is easy to query through standardized query languages ​​(such as SQL), and then relies on predefined indicators and rules for evaluation.

[0004] In the process of implementing this application, the applicant found that the related technology has at least the following problems:

[0005] At present, a large number of academic assets are still in an unstructured state (such as journal articles, doctoral and master's theses, conference papers, etc. saved in PDF format). In order to use the existing academic asset quality evaluation methods, data must be structured first, that is, unstructured data must be converted into structured data. This process is not only time-consuming and laborious, but may also lose some important information of the original data, thus affecting the accuracy of the evaluation results. Summary of the invention

[0006] In view of this, this application provides a method for evaluating the quality of academic assets, the main purpose of which is to solve the problem that a large number of academic assets are still in an unstructured state (such as journal articles, doctoral and master's theses, conference papers, etc. saved in PDF format). In order to use the existing academic asset quality evaluation methods, data must be structured first, that is, unstructured data must be converted into structured data. This process is not only time-consuming and laborious, but also may lose some important information of the original data, thereby affecting the accuracy of the evaluation results.

[0007] According to the first aspect of the present application, a method for evaluating the quality of academic assets is provided, the method comprising:

[0008] A. Obtain the academic assets and basic information of academic assets indicated by the current quality evaluation task;

[0009] B. Determine whether to adopt a full sampling method or a double-blind sampling method based on the basic information of the academic assets, and determine a sample set of academic assets to be evaluated from the academic assets;

[0010] C. Determine a quality evaluation index system, wherein the quality evaluation index system is determined according to the quality evaluation target indicated by the current quality evaluation task, the quality evaluation index system includes at least one quality evaluation index, and each of the quality evaluation indicators is associated with at least one relevant statistic;

[0011] D. Determine the scoring calculation model;

[0012] E. Identify the academic asset sample set and calculate the specific value corresponding to each relevant statistic;

[0013] F. Calculate the index value of the corresponding quality evaluation index according to the relevant statistics;

[0014] G. Calculate the score of each quality evaluation indicator and the total quality evaluation score according to the indicator value of each quality evaluation indicator and the score calculation model;

[0015] H. Generate a quality evaluation report corresponding to the current quality evaluation task.

[0016] In the above technical solution, the basic information of academic assets describes the overview of academic assets, including the types of academic data, the fields covered by academic data, the format of academic data and the amount of academic data; the academic data types include journals, doctoral and master's theses and conference papers; the fields covered by academic data refer to the fields covered by academic data content such as social sciences, natural sciences, agricultural sciences, engineering technology, medicine and health, education, culture, language and literature, economics and management, etc.; the format of academic data refers to the format of academic documents, such as pdf, doc, etc.

[0017] In the above technical solution, the double-blind sampling means that the entrusting party and the entrusted party of the academic asset quality evaluation do not know in advance which specific samples of the academic assets to be evaluated will be sampled and selected to form the academic asset sample set for quality evaluation. If the amount of academic asset data to be evaluated is huge, the double-blind sampling method is used; if the amount of academic asset data to be evaluated is small and the quality evaluation takes time to meet the actual requirements, the full quality evaluation is carried out.

[0018] In the above technical scheme, the index system for academic asset quality evaluation includes a completeness index subset, an accuracy index subset, a consistency index subset and an accessibility index subset; the completeness index subset includes three indicators, namely, the catalog content completeness index, the academic article ratio index and the academic article metadata completeness index; the accuracy index subset includes two indicators, namely, metadata format accuracy and article repetition rate; the consistency index subset includes three indicators, namely, the catalog and in-text title consistency index, the catalog and in-text printed page number consistency index and article metadata display consistency index; the accessibility index subset includes three indicators, namely, accessible character density index, accessible character ratio index and accessible image ratio index. Determining the index system for academic asset quality evaluation refers to selecting relevant indicator sets from the above academic asset quality evaluation index system according to the specific goals of a certain academic asset quality evaluation to construct an index system that meets the specific goals of this academic asset quality evaluation.

[0019] In the above technical solution, the scoring calculation model for the academic asset quality evaluation includes an indicator value normalization and normalization calculation sub-model and an indicator weight calculation sub-model. The indicator value normalization and normalization calculation sub-model means that the indicator values ​​are first normalized so that the values ​​of each indicator have the same monotonicity, that is, the larger the indicator value, the higher the quality evaluation, and the smaller the indicator value, the lower the quality evaluation; and then the indicator values ​​are normalized and mapped to the [0,1] interval. The indicator weight calculation sub-model means that a weight value is assigned to each indicator using expert recommendation and hierarchical analysis method. The larger the indicator weight value, the higher the importance of the indicator.

[0020] In the above technical solution, the calculation of the relevant statistics of academic asset quality evaluation means that in order to obtain each indicator value of academic asset quality evaluation, the corresponding statistics need to be calculated. For example, in order to calculate the index of article repetition rate, it is necessary to count the statistic of the number of repeated articles; for another example, in order to calculate the index of accessible character ratio, it is necessary to count the two statistics of the number of directly accessible characters and the number of indirect accessible characters (converted characters, characters represented by pictures).

[0021] In the above technical solution, the academic asset quality evaluation report includes the basic information of the academic assets to be evaluated, the specific method of quality evaluation, and the results and analysis of the quality evaluation.

[0022] Compared with the existing technology, the present invention designs an academic asset quality evaluation index system from multiple dimensions of completeness, accuracy, consistency and accessibility, and adopts expert recommendation and hierarchical analysis method to design an academic asset quality evaluation scoring calculation model. It can accurately evaluate the quality of semi-structured and unstructured academic data assets, and provide solid quality assurance for application scenarios such as academic asset inclusion and academic asset circulation and trading.

[0023] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0025] Figure 1 A schematic diagram of a method flow for evaluating academic asset quality provided in an embodiment of the present application is shown;

[0026] Figure 2 A schematic flow chart of a method for evaluating the quality of academic assets provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0028] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with embodiments and drawings.

[0029] like Figure 1 The following is the process of evaluating the quality of academic assets, including:

[0030] A. Obtain the academic assets and basic information of academic assets indicated by the current quality evaluation task;

[0031] B. Determine whether to adopt a full sampling method or a double-blind sampling method based on the basic information of the academic assets, and determine a sample set of academic assets to be evaluated from the academic assets;

[0032] C. Determine a quality evaluation index system, wherein the quality evaluation index system is determined according to the quality evaluation target indicated by the current quality evaluation task, the quality evaluation index system includes at least one quality evaluation index, and each of the quality evaluation indicators is associated with at least one relevant statistic;

[0033] D. Determine the scoring calculation model;

[0034] E. Identify the academic asset sample set and calculate the specific value corresponding to each relevant statistic;

[0035] F. Calculate the index value of the corresponding quality evaluation index according to the relevant statistics;

[0036] G. Calculate the score of each quality evaluation indicator and the total quality evaluation score according to the indicator value of each quality evaluation indicator and the score calculation model;

[0037] H. Generate a quality evaluation report corresponding to the current quality evaluation task.

[0038] This application embodiment provides a method for evaluating the quality of academic assets, such as Figure 2 As shown, the method includes:

[0039] S10. Obtain the academic assets and basic information of the academic assets indicated by the current quality evaluation task.

[0040] The academic asset quality evaluation method provided in this application is applicable to the academic asset quality evaluation system. Relevant technical personnel can set quality evaluation tasks based on the academic asset quality evaluation system. The quality evaluation task points out the evaluation objectives of this quality evaluation and the academic assets associated with this task. The basic information of academic assets describes the profile of academic assets, including the types of academic data, the fields covered by academic data, the format of academic data and the amount of academic data. The academic data types include journals, doctoral and master's theses and conference papers. The fields covered by the academic data refer to the fields covered by the academic data content, such as social sciences, natural sciences, agricultural sciences, engineering technology, medicine and health, education, culture, language and literature, economics and management, etc. The format of the academic data refers to the format of academic documents, such as pdf, doc, etc.

[0041] S20. Determine, based on the basic information of the academic assets, whether to adopt a full sampling method or a double-blind sampling method, and determine a sample set of academic assets to be evaluated in the academic assets.

[0042] Further, the data quantity of multiple academic data is compared with a preset quantity threshold. If the comparison result indicates that the data quantity is less than or equal to the preset quantity threshold, the multiple academic data in the academic assets are added to the academic asset sample set to obtain the academic asset sample set to be evaluated for full evaluation. If the comparison result indicates that the data quantity is greater than the preset quantity threshold, that is, the amount of academic asset data to be evaluated is huge, then a double-blind sampling method is used to extract a preset number of academic data from multiple academic data and add them to the academic asset sample set to obtain the academic asset sample set to be evaluated. Double-blind sampling means that the entrusting party and the trustee of the academic asset quality evaluation do not know in advance which specific samples in the academic assets to be evaluated will be sampled and selected to form the academic asset sample set for quality evaluation. In addition, the preset quantity threshold can be set by relevant staff according to actual needs, and this application does not specifically limit the value of the preset quantity threshold.

[0043] S30. Determine a quality evaluation index system and a scoring calculation model, wherein the quality evaluation index system is determined according to the quality evaluation target indicated by the current quality evaluation task, and the quality evaluation index system includes at least one quality evaluation index, and each of the quality evaluation indicators is associated with at least one related statistic.

[0044] Determine at least one quality evaluation dimension indicated by the quality evaluation target, determine a subset of quality evaluation indicators corresponding to each of the quality evaluation dimensions in a preset quality evaluation indicator set, and use at least one of the quality evaluation indicator subsets to construct the quality evaluation indicator system, wherein the quality evaluation indicator subset includes a completeness indicator subset, an accuracy indicator subset, a consistency indicator subset, or an accessibility indicator subset, and each of the quality evaluation indicator subsets includes at least one of the quality evaluation indicators. Among them, the completeness indicator subset includes a directory content completeness indicator, an academic article ratio indicator, and an academic article metadata completeness indicator. The accuracy indicator subset includes a metadata format accuracy indicator and an article repetition rate indicator. The consistency indicator subset includes a directory and article title consistency indicator, a directory and article printed page number consistency indicator, and an article metadata display consistency indicator. The accessibility indicator subset includes an accessible character density indicator, an accessible character ratio indicator, and an accessible image ratio indicator.

[0045] In the above technical solution, the scoring calculation model for the academic asset quality evaluation includes an indicator value normalization and normalization calculation sub-model and an indicator weight calculation sub-model. The indicator value normalization and normalization calculation sub-model means that the indicator values ​​are first normalized so that the values ​​of each indicator have the same monotonicity, that is, the larger the indicator value, the higher the quality evaluation, and the smaller the indicator value, the lower the quality evaluation; and then the indicator values ​​are normalized and mapped to the [0,1] interval. The indicator weight calculation sub-model means that a weight value is assigned to each indicator using expert recommendation and hierarchical analysis method. The higher the indicator weight value, the higher the importance of the indicator.

[0046] S40, identifying the academic asset sample set, calculating the specific numerical value corresponding to each of the relevant statistics, and calculating the index value of the corresponding quality evaluation index according to the specific numerical value of the relevant statistics.

[0047] In this step, for each academic data in the academic asset sample set, the academic data is identified, the specific numerical value corresponding to each relevant statistic is determined, and at least one corresponding relevant statistic is calculated according to the indicator calculation formula corresponding to each quality evaluation indicator to obtain the indicator value corresponding to each quality evaluation indicator. For example, in order to calculate the indicator value of the accessible character ratio quality evaluation indicator, it is necessary to determine the specific numerical value corresponding to the statistic related to the number of directly accessible characters and the specific numerical value corresponding to the statistic related to the number of characters that are not directly accessible (curved characters, characters represented by pictures), and then use these two relevant statistics to calculate the indicator value. Among them, the academic data is identified, including the data directory corresponding to the academic data and / or the articles in the academic data and / or the article metadata of the articles. The article metadata is obtained by extracting the articles in the academic data using a specified metadata extraction method. Different academic document formats correspond to different specified metadata extraction methods, that is, the specified metadata extraction method is a metadata extraction method associated with the format of the academic data. The format of the academic data is the format of the academic document, such as PDF format corresponding to a metadata extraction method, WORD format corresponding to another metadata extraction method, and this application does not specifically limit the specified metadata extraction method. The academic documents of academic data include academic articles and other articles, wherein academic articles refer to articles whose article metadata include title, author name, author organization, abstract, keywords, and references.

[0048] The above-mentioned identification of the academic asset sample set, calculation of the specific numerical value corresponding to each relevant statistic, and determination of the index value of each quality evaluation indicator based on the relevant statistics corresponding to each quality evaluation indicator refer to calculation of the full set or subset of the following indicators, including: catalog content integrity index, academic article ratio index, academic article metadata integrity index, metadata format accuracy index, article repetition rate index, catalog and text title consistency index, catalog and text printed page number consistency index, article metadata readability consistency index, accessible character density index, accessible character ratio index and accessible image ratio index. The specific calculation process of the index value of each quality evaluation indicator is as follows:

[0049] The index of completeness of catalog content is defined as the proportion of articles in the catalog of academic documents, which is used to evaluate the completeness of the catalog content. The index calculation formula is shown in the following formula 1:

[0050]

[0051] Among them, n c is the number of articles appearing in the data directory, n t The total number of articles for academic data.

[0052] For the academic article ratio index, the academic article ratio index is defined as the ratio of the number of academic articles to the total number of articles. The larger the academic article ratio, the higher the academic value and the better the quality. The index calculation formula is shown in the following formula 2:

[0053]

[0054] Among them, n a is the number of academic articles in the academic data, n t The total number of articles for academic data.

[0055] Regarding the academic article metadata integrity index, the academic article metadata integrity index is used to evaluate the completeness of academic article metadata. The index calculation formula is shown in the following formula 3:

[0056]

[0057] Among them, n i is the number of times the i-th type of metadata appears in academic articles, n a is the number of academic articles in the academic data, and N is the number of metadata categories. The metadata categories include but are not limited to article title, author name, author institution, abstract, keywords, China Library Classification Number, article number, document identification code, DOI, and references;

[0058] For metadata format accuracy indicators, some metadata such as China Library Classification Number, Article Number, Document Identification Code, and DOI have specific formats. For example, the document identification code is usually represented by a capital English letter, and the format of the China Library Classification Number is a capital English letter followed by three Arabic numerals. The metadata format accuracy indicator is used to evaluate the accuracy of metadata with a specific format. The indicator calculation formula is shown in the following formula 4:

[0059]

[0060] Among them, m i is the number of correct metadata formats for the i-th category, t i is the number of occurrences of the i-th type of specific metadata, and M is the number of metadata categories with a specific format.

[0061] For the article duplication rate indicator, if there is a large amount of duplicate content in the academic assets, the quality of the academic assets will be affected. The article duplication rate indicator is used to evaluate the degree of duplication of academic asset content. The indicator calculation formula is shown in the following formula 5:

[0062]

[0063] Among them, n r is the number of duplicate articles, n t The total number of articles in the academic data. Duplicate articles are articles whose article titles, author names, author institutions, keywords, Chinese Library Classification Numbers, article numbers, document identification codes, and data unique identifiers (DOIs) are consistent in the article metadata of the two articles.

[0064] The consistency index between the catalog and the in-text article title is used to evaluate the degree of consistency between the article title in the data catalog and the in-text article title. The index calculation formula is shown in the following formula 6:

[0065]

[0066] Among them, t consist is the number of articles in the data catalog that have the same title as the corresponding article, n ct The total number of directory entries in the data directory.

[0067] The consistency index between the printed page numbers in the catalog and the text is used to evaluate the consistency between the printed page numbers in the data catalog and the printed page numbers in the text. The calculation formula of the index is shown in the following formula 7:

[0068]

[0069] Among them, pr consistis the number of pages printed in the data directory that match the page numbers printed in the article, n ct The total number of directory entries in the data directory.

[0070] For the article metadata reading consistency index, when reading the article title, author and other metadata from academic documents, the read content may be inconsistent with the displayed content. This is a problem caused by garbled characters. The problem of garbled characters will reduce the quality of academic assets. The article metadata reading consistency index evaluates the degree of garbled characters in academic documents from the perspective of article metadata. The fewer garbled characters, the better the article metadata reading consistency, and the higher the quality of academic assets. The indicator calculation formula is shown in the following formula 8:

[0071]

[0072] Among them, mc i Number of characters of metadata of type i, mcg i Represents the number of garbled characters in the i-th type of metadata, and N is the number of metadata categories.

[0073] For the accessible character ratio indicator, if the characters of the academic document are stored in font format, the characters can be directly accessed, but if the characters are converted (vectorized) or stored in a dot matrix format, they cannot be directly accessed. Whether the characters can be directly accessed affects the convenience of subsequent use of academic documents. For this reason, the accessible character ratio indicator is designed to evaluate the quality of academic assets from the perspective of character accessibility. The indicator calculation formula is shown in the following formula 9:

[0074]

[0075] Among them, C a is the number of characters directly accessible in the academic data, C d is the number of characters converted to vector form, C m The number of characters described for a dot matrix image. The steps to estimate the number of characters in a dot matrix image are as follows:

[0076] Use the document layout analyzer to determine whether there is a text area in the dot map;

[0077] If there is no text area, the bitmap does not contain any characters;

[0078] If there is a text area and the area of ​​the text area reaches the threshold, that is, the area of ​​the text area accounts for more than 50% of the area of ​​the dot matrix, the following formula 10 is used to estimate the number of characters in the dot matrix:

[0079]

[0080] Among them, text h is the height of the text area in the bitmap.w The width of the text area in the bitmap, char h The average font height, char w The average font width and line width of the document where the bitmap is located. h The average line spacing of the document containing the bitmap.

[0081] For the accessible character density index, the index calculation formula is shown in the following formula 11:

[0082]

[0083] Among them, C a is the number of characters directly accessible in the academic data, n page The number of pages in the document for academic data.

[0084] For the accessible image ratio index, if the image of the academic document is stored in the form of a bitmap, the image can be directly accessed; if a picture is described in both bitmap and vector form, the image cannot be directly accessed. The index calculation formula is shown in the following formula 12:

[0085]

[0086] Among them, a is the number of images that can be directly accessed, im t is the total number of images.

[0087] S50: Calculate the score of each quality evaluation indicator and the total quality evaluation score according to the indicator value of each quality evaluation indicator and the score calculation model.

[0088] In the above technical solution, the formula for calculating the total quality evaluation score based on the quality evaluation index and the scoring calculation model is shown in the following formula 13:

[0089]

[0090] Among them, W i is the weight of the ith indicator, P i is the result of the normalization and calculation of the index value of the i-th index, and N is the number of indicators.

[0091] S60: Generate a quality evaluation report corresponding to the current quality evaluation task.

[0092] In the above technical solution, the academic asset quality evaluation report includes the basic information of the academic assets to be evaluated, the specific method of quality evaluation, and the results and analysis of the quality evaluation. In the actual operation process, the identification of the academic asset sample set can be recorded, and the specific numerical value corresponding to each of the relevant statistics can be calculated. According to the relevant statistics, the index value of the corresponding quality evaluation index is calculated, and according to the index value of each quality evaluation index and the scoring calculation model, the score of each quality evaluation index and the quality evaluation total score are calculated. Finally, the basic information of academic assets, the total quality evaluation score and the recorded quality evaluation process are used to update the quality evaluation report corresponding to the current quality evaluation task.

[0093] The method provided in the embodiment of the present application first obtains the basic information of academic assets. Further, according to the basic information of the academic assets, it is determined to adopt a full sampling or double-blind sampling method to determine the academic asset sample set to be evaluated in the academic assets. Next, determine the indicator system for the quality evaluation of academic assets. Determine the scoring calculation model. Subsequently, calculate the relevant statistics of the academic asset quality evaluation. Further, calculate the corresponding quality evaluation indicators based on the relevant statistics. Further, calculate the score of each quality evaluation indicator and the total quality evaluation score based on the quality evaluation indicators and the scoring calculation model. Finally, generate an academic asset quality evaluation report. The quality evaluation method provided in the embodiment of the present application can directly evaluate unstructured data without structuring the data in advance, simplifying the evaluation process, improving the evaluation efficiency, and being able to adapt to different types of academic assets, thereby solving the problems of inefficiency and complex evaluation system in the existing academic asset quality evaluation. Compared with the existing technology, the present invention designs an academic asset quality evaluation index system from multiple dimensions of completeness, accuracy, consistency and accessibility, and adopts expert recommendation and hierarchical analysis method to design an academic asset quality evaluation scoring calculation model. It can accurately evaluate the quality of semi-structured and unstructured academic data assets, and provide solid quality assurance for application scenarios such as academic asset inclusion and academic asset circulation and trading.

[0094] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by hardware, or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0095] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present application.

[0096] Those skilled in the art will appreciate that the modules in the device in the implementation scenario can be distributed in the device in the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more devices different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0097] The above application serial numbers are for description only and do not represent the advantages or disadvantages of the implementation scenarios.

[0098] The above disclosure only discloses several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by technicians in this field should fall within the scope of protection of the present application.

Claims

1. A method for evaluating the quality of academic assets, characterized in that: include: Obtain academic assets and basic information of academic assets indicated by the current quality evaluation task; Determine whether to adopt a full sampling method or a double-blind sampling method according to the basic information of the academic assets, and determine a sample set of academic assets to be evaluated in the academic assets; Determine a quality evaluation index system and a scoring calculation model, wherein the quality evaluation index system is determined according to the quality evaluation target indicated by the current quality evaluation task, the quality evaluation index system includes at least one quality evaluation index, and each of the quality evaluation indicators is associated with at least one relevant statistic; Identify the academic asset sample set, calculate the specific value corresponding to each of the relevant statistics, and calculate the index value of the corresponding quality evaluation index according to the specific value of the relevant statistics; Calculate the score of each quality evaluation indicator and the total quality evaluation score according to the indicator value of each quality evaluation indicator and the score calculation model; Generate a quality evaluation report corresponding to the current quality evaluation task.

2. The method according to claim 1, characterized in that: The method of determining, based on the basic information of the academic assets, using a full sampling method or a double-blind sampling method to determine a sample set of academic assets to be evaluated in the academic assets includes: The academic assets include a plurality of academic data, and the basic information of the academic assets includes but is not limited to the data quantity of the plurality of academic data, and the data type, data content field and data format corresponding to each of the academic data; Comparing the data quantity of the plurality of academic data with a preset quantity threshold; If the comparison result indicates that the amount of data is less than or equal to the preset amount threshold, multiple academic data in the academic assets are added to the academic asset sample set to obtain the academic asset sample set to be evaluated; If the comparison result indicates that the data quantity is greater than the preset quantity threshold, a double-blind sampling method is used to extract a preset number of academic data from the multiple academic data and add them to the academic asset sample set to obtain the academic asset sample set to be evaluated.

3. The method according to claim 1, characterized in that: The quality evaluation index system is determined, including: Determine at least one quality evaluation dimension indicated by the quality evaluation target, determine a quality evaluation indicator subset corresponding to each of the quality evaluation dimensions in a preset quality evaluation indicator set, and adopt at least one of the quality evaluation indicator subsets to construct the quality evaluation indicator system, wherein the quality evaluation indicator subset includes an integrity indicator subset, an accuracy indicator subset, a consistency indicator subset, or an accessibility indicator subset, and each of the quality evaluation indicator subsets includes at least one of the quality evaluation indicators; Among them, the completeness index subset includes the catalog content completeness index, the academic article ratio index and the academic article metadata completeness index; the accuracy index subset includes the metadata format accuracy index and the article repetition rate index; the consistency index subset includes the catalog and article title consistency index, the catalog and article printed page number consistency index and the article metadata display consistency index; the accessibility index subset includes the accessible character density index, the accessible character ratio index and the accessible image ratio index.

4. The method according to claim 3, characterized in that The step of identifying the academic asset sample set, calculating the specific value corresponding to each of the relevant statistics, and calculating the index value of the corresponding quality evaluation index according to the specific value of the relevant statistics includes: For each academic data in the academic asset sample set, the academic data is identified, a specific value corresponding to each relevant statistic is calculated, and at least one corresponding relevant statistic is calculated according to the indicator calculation formula corresponding to each quality evaluation indicator to obtain an indicator value corresponding to each quality evaluation indicator; Among them, the academic data is identified, including identifying the data directory corresponding to the academic data and / or the articles in the academic data and / or the article metadata of the articles, the article metadata is obtained by extracting the articles in the academic data using a specified metadata extraction method, the specified metadata extraction method is a metadata extraction method associated with the format of the academic data, the format of the academic data is the format of academic documents, the articles in the academic data include academic articles and other articles, and the academic articles are articles whose article metadata include title, author name, author organization, abstract, keywords, and references.

5. The method according to claim 4, characterized in that The step of calculating at least one corresponding relevant statistic according to the indicator calculation formula corresponding to each quality evaluation indicator to obtain an indicator value corresponding to each quality evaluation indicator includes: For the directory content integrity index, the index calculation formula is as follows: Among them, n c is the number of articles appearing in the data directory, n t The total number of articles for the academic data; For the academic article ratio indicator, the indicator calculation formula is as follows: Among them, n a is the number of academic articles in the academic data; For the academic article metadata integrity index, the index calculation formula is as follows: Among them, n i is the number of times the i-th type of metadata appears in academic articles, and N is the number of metadata categories; For the metadata format accuracy indicator, the indicator calculation formula is as follows: Among them, m i is the number of correct metadata formats for the i-th category, t i is the number of occurrences of the i-th type of specific metadata, and M is the number of metadata categories with a specific format; For the article duplication rate indicator, the indicator calculation formula is as follows: Among them, n r is the number of duplicate articles, where the article title, author name, author institution, keywords, Chinese Library Classification Number, article number, document identification code and data unique identifier DOI in the article metadata of the two articles are consistent; For the consistency index between the catalog and the article title, the index calculation formula is as follows: Among them, t consist is the number of articles in the data directory whose titles are consistent with the titles in the corresponding articles, n ct is the total number of directory entries in the data directory; For the consistency index between the catalog and the printed page numbers in the document, the index calculation formula is as follows: Among them, pr consist The number of printed page numbers in the data directory that are consistent with the printed page numbers in the corresponding article; For the article metadata read consistency index, the index calculation formula is as follows: Among them, mc i Number of characters of metadata of type i, mcg i Represents the number of garbled characters of the i-th type of metadata; For the accessible character ratio indicator, the indicator calculation formula is as follows: Among them, C a is the number of characters directly accessible in the academic data, C d is the number of characters converted to vector form, C m is the number of characters described by the dot matrix image, wherein, if the document layout analyzer is used to determine that the dot matrix image contains a text area and the area of ​​the text area meets the preset conditions, the following formula is used to calculate C m , the preset condition is that the area of ​​the text area is greater than or equal to half of the area of ​​the dot matrix: Among them, text h is the height of the text area in the bitmap. w The width of the text area in the bitmap, char h The average font height, char w The average font width and line width of the document where the bitmap is located. h is the average line spacing of the document where the dot matrix image is located; For the accessible character density index, the index calculation formula is as follows: Among them, n page The number of pages of the document of the academic data; For the accessible image ratio indicator, the indicator calculation formula is as follows: Among them, a is the number of images that can be directly accessed, im t is the total number of images in the academic data.

6. The method according to claim 1, characterized in that The step of calculating the score of each quality evaluation indicator and the total quality evaluation score according to the indicator value of each quality evaluation indicator and the score calculation model comprises: The indicator value normalization and normalization calculation sub-model is used to normalize the indicator value corresponding to each of the quality evaluation indicators, so as to adjust all the indicator values ​​to have the same monotonicity, so that the larger the indicator value, the higher the quality evaluation, and the smaller the indicator value, the lower the quality evaluation, and the normalized indicator value is normalized, and the normalized indicator value is mapped to the [0, 1] interval to obtain the normalized value corresponding to each indicator value; The indicator weight calculation submodel adopts expert recommendation and hierarchical analysis method to set a weight value for each quality evaluation indicator, and performs weighted calculation on the normalized numerical value corresponding to each indicator value based on the weight value corresponding to each quality evaluation indicator to obtain the total quality evaluation score, wherein the sum of the weight values ​​corresponding to the quality evaluation indicators is 1. The preset scoring calculation model includes the indicator value normalization and normalization calculation submodel and the indicator weight calculation submodel.

7. The method according to claim 1, characterized in that The generating of the quality evaluation report corresponding to the current quality evaluation task includes: Record the quality evaluation process of identifying the academic asset sample set, calculating the specific value corresponding to each of the relevant statistics, calculating the index value of the corresponding quality evaluation index according to the specific value of the relevant statistics, and calculating the score of each of the quality evaluation indicators and the total quality evaluation score according to the index value of each of the quality evaluation indicators and the score calculation model; The academic asset basic information, the quality evaluation total score and the recorded quality evaluation process are used to update the quality evaluation report corresponding to the current quality evaluation task.

Images