CN104133839A - Data processing method and system with intelligent detection function - Google Patents

Abstract

The invention discloses a data processing method and system with an intelligent detection function. Step 1: collect the data of the declared project; Step 2: read the data in the main data table and each sub-data table of the project declaration database, and judge whether the data meets the requirements ;Step 3: Synchronize the project data that meets the requirements in step 2 from the project application database to the review database; screen the projects that meet the requirements three times, and store the data of the final award-winning projects in the third cache area in the review database output. The advantage of data judgment is that the system automatically extracts duplicate checking factors, performs complex matching calculations, reduces human factors, improves the correctness of duplicate checking results, reduces the workload of staff, and greatly improves work efficiency.

Description

A data processing method and system with intelligent detection function

technical field

The invention relates to a data processing method and system with intelligent detection function.

Background technique

The current technology incentive management project has the following shortcomings in data processing:

The amount of data in the science and technology reward management project is large, and a large amount of data needs to be processed every year. During the processing, the data screening is not reasonable enough. In addition, the existing system lacks the functions of automatic duplicate checking and automatic processing.

It is difficult to screen a large amount of cumbersome data, and the processing process is not reasonable enough. The original system screened the data once, with a single screening basis, many human intervention factors, and lack of fairness and rationality. Manual data processing is heavy workload and low efficiency. The original system needs to manually check and compare data, which has low work efficiency and heavy tasks.

When declaring scientific and technological awards, there are many application materials to be filled in. When the application materials are filled with the name of the project or the name of the thesis, the person who completed the project and the patent document, it is necessary to judge whether there is any suspicion of repeated application based on the name. At present, the work They are all identified by humans. Due to the large amount of declared data, the accuracy of human identification is not enough.

Contents of the invention

The purpose of the present invention is to solve the above problems and provide a method and system for project declaration data processing. The data judgment advantage of the present invention lies in that the system automatically extracts plagiarism factors, performs complex matching calculations, reduces human factors, and improves the fairness of plagiarism results. and correctness, reduce the workload of staff, and greatly improve work efficiency.

In order to achieve the above object, the present invention adopts the following technical solutions:

A data processing method with an intelligent detection function, comprising the steps of:

Step 1: Automatically detect the browser version used by the current user through the JS script, give a prompt for non-IE series browsers, and close it; detect that the browser used meets the requirements and then enter the system for data collection;

Step 2: The remote data collection terminal collects the handwriting of the completed person through the tablet and stores it in the handwriting feature database, and stores the data information of the collected application items in the cache area of the network server, and the host computer calls the data information of the application items in the cache area of the network server And stored in the first cache area of the host computer; the collected pictures, Word and PDF document data are stored in the second cache area of the host computer in the form of files, and the relative path of the document is stored in the sub-data table of the project declaration database attachment;

Step 3: The host computer reads the information of the declaration project in the first buffer area, and judges whether the data meets the requirements; for the picture file, use the document path in the sub-data table of the attachment of the project declaration database to call the picture file in the second buffer area of the host computer, Use the image recognition module to read the content of the picture document, identify the handwriting of the person who completed it and compare it with the handwriting feature database, and check whether the seal of the completion unit matches the name of the completion unit in the data sheet of the completion unit. If the handwriting information in the handwriting feature database and the project declaration data table If the information and picture documents of the declared project meet the requirements, go to step 4, and if not, go back to step 2;

Step 4: Synchronize the project data that meets the requirements in step 3 from the project application database to the review database. The synchronization process adopts one-way synchronization; perform three screenings on the application information of the review database, and output the final data on the browsing page of the server .

In the second step, the data directly related to the project are stored in the master data table of the project declaration database, each piece of data corresponds to a unique project number, and the data indirectly related to the project are stored in each sub-data table of the project declaration database, In the sub-data table, the unique item number is set as a foreign key to associate with the main data table; the sub-data table of the declaration database also stores historical project declaration information.

The directly related data includes the basic information of the project; the indirectly related data includes the project introduction, major technological innovations, third-party evaluation, social and economic benefits, scientific and technological awards received, the person who completed it, the unit that completed it, the opinion of the recommending unit, intellectual property rights, papers, etc. treatises and appendices;

The basic information of the project includes: project name, project discipline, technical field, project source and national economic industry, etc.; the intellectual property rights include: patent application number, patent name, inventor and patent authorization date; the completion person includes : Completer's name, ID number and sequence of completers. The data storage forms in the step 1 include: data tables, pictures, Word documents and PDF documents. Stored as data in different formats, the data table is convenient for query and statistics, pictures and PDF documents ensure the authenticity of the data, and Word documents are easy to maintain the original format of the data for easy viewing.

In said step 3, word segmentation or direct matching is carried out to the data stored in each sub-data table of the project declaration database, and the judging module is used to judge whether the information of the current declaration project is the project name or the name of the thesis, and if so, enter the keyword comparison module , if not, judge whether it is the name and ID number or intellectual property number of the person who completes the project, if yes, enter the direct matching module.

Use the keyword comparison module to compare the project name or thesis name of the current application project information with the keyword of another application project’s project name or thesis name in the historical project application information, if the similarity is not lower than the set value, it is determined to be repeated, otherwise it is not repeated;

Use the direct matching module to directly match the completer's name and ID number or intellectual property number of the information item of the current application project with the project completion person's name, ID number or intellectual property number of another application item in the historical project application information Match, if the same judgment is repeated, otherwise not repeated;

Using the storage module, the keyword comparison module and the direct matching module determine the information of non-duplicate declaration items to be stored in the cache area of the host computer, and the information of the duplicate declaration items is stored in the duplicate checking table of the item declaration database in the host computer.

The keyword comparison module compares the project name or thesis name of the information of the current application project with the keywords of the project name or thesis name of another application project in the historical project application information, and if the similarity is not lower than the set Set the value, determine the repetition, otherwise it will not repeat, the specific process is:

Use the keyword extraction module to extract one of the information of the current declared project and the declared information of historical projects, perform word segmentation on the extracted information, and store the decomposed keywords into two arrays corresponding to the project declaration database in the main computer;

Use the keyword matching module to traverse and compare the keywords in the two arrays corresponding to the project declaration database to obtain the same keyword number and the number of keywords in each group;

According to the similarity module, the similarity between the information of the current declared project and the project information in the historical project declaration information is obtained, and the similarity is compared with the set value. If the similarity is not lower than the set value, it is determined to be repeated, otherwise it is not repeated. .

The keyword extraction module extracts one of the information of the current declaration project and the declaration information of historical projects, performs word segmentation on the extracted information, and uses the ShootSearch component to do word segmentation.

The similarity module is specifically used to compare the number of keywords in the two arrays corresponding to the project declaration database, extract the smaller number of keywords, and divide the same number of keywords by the smaller number of keywords to obtain the similarity.

Using the direct matching module to combine the completer’s name and ID number or intellectual property number of the information item of the current application project with the project completion person’s name, ID number or intellectual property number of another application item in the historical project application information Match directly, specifically:

The information of the declaration items collected by the remote data collection terminal and the historical project declaration information in the sub-data table of the declaration database are directly traversed and cyclically matched to determine whether they match. If they match, it is determined to be repeated, otherwise, it is not repeated.

The information of the declared project includes the project name, the name of the thesis, the name and ID number of the person who completed the project, and the intellectual property number. Name, the name of the person who completed the project, ID number and intellectual property number. The intellectual property number is the patent application number or patent publication number.

The similarity calculation method is used to check duplicate items to a greater extent to avoid repeated declarations of recombined items; the word segmentation data includes the project name and the title of the paper; the direct matching data includes: the author's Name and ID number and intellectual property number. Data plagiarism check is an important link in the entire science and technology reward system. The data processing process is complicated and the processing methods are different. Different methods are adopted for different data to check for duplicates, which is to avoid false checks and missed checks. The completer's name, ID number and intellectual property number are complete identifications, and there is no possibility of combination. Project names and thesis titles can be decomposed and reorganized, and word segmentation can be used to calculate the similarity to query duplicate items more accurately.

In the third step, the application information of the review database is screened three times, specifically:

Screen the projects in the review database that meet the requirements for the first time, select the corresponding network review experts according to the project information, and screen the data synchronized to the review database through network review; store the screened declared project data in the review database The third cache area in;

The second screening is performed on the declared project data stored in the third buffer area; senior experts are selected according to the project information, experts vote on the data extracted from the first screening, and the preliminary award-winning projects are taken out from the voting results, and the preliminary award-winning projects are selected The data stored in the fourth cache area in the review database;

The third screening is carried out on the declared project data stored in the fourth buffer area; the experts of the Science and Technology Committee are selected according to the project information, and the data taken out in the second screening are reviewed by experts, and the final award-winning projects are taken out from the review results, and the final Award-winning projects are stored in the fifth cache in the review database.

When screening, set the weight of scoring indicators, select experts, perform expert scoring on the synchronized data, add up the expert scores according to the weights to obtain project scores, sort according to project scores from high to low, and take settings from the sorting results number of items. Voting includes: first prize, second prize, third prize and no award. Reviews include objections and non-objections.

In the above three screenings, the selection process of experts is as follows:

S1: Store the scientific and technological project subject information in the form of data set A in the sub-data table of the review database, and the sub-data table of the review database also stores expert subject information in the form of data set B;

S2: Select the subject information of experts in data set B as a condition factor, judge whether the condition factor of the subject information of the expert is a first-level subject, if so, combine the condition factor with the science and technology project discipline stored in the form of data set A Matching of key factors of information, traversing whether the conditional factors in expert subject information contain key factors, if yes, match and go to step S5, otherwise not match, go to step S3;

S3: Determine whether the condition factor of the expert information is a secondary discipline, if so, match the condition factor with the key factor of the scientific and technological project information stored in the form of data set A, and traverse whether the condition factor in the expert information contains the key Factors, if they are included, they will match and go to step S5, otherwise they will not match and go to step S4;

S4: Determine whether the conditional factor of the expert information is a third-level subject, if so, match the conditional factor with the key factor of the scientific and technological project information stored in the form of data set A, and traverse whether the conditional factor in the expert information contains the key Factors, match if contained, go to S5, otherwise not match;

S5: Randomly select the number of experts matching the scientific and technological project from the stored data set B of expert subject information, and store the retrieved data in the review database;

The step S5 specifically includes selecting data according to random numbers from zero to a specified number returned by the Random (random) function until the specified number is selected.

The three screenings are different, and the evaluation of the data is different. The first screening is scoring, the second screening is voting and grading, and the third screening is voting. But the process of selecting experts in the three screenings is the same. The filtered data is output in the form of a Word format document. The document output in Word format is convenient for users to adjust the data format by themselves.

A data processing system with an intelligent detection function, including a system verification module, which is used to automatically detect the browser version used by the current user through JS scripts, give a prompt to non-IE series browsers, and close them; Enter the system for data collection;

The remote data collection terminal is used to store the data information of the collected declaration items in the cache area of the network server, and store the handwriting of the person who completed the collection by the remote data collection terminal through the handwriting board into the handwriting feature database;

The host computer is used to call the data information of the declared items in the cache area of the network server and store it in the first cache area of the host computer; the collected pictures, Word and PDF document data are stored in the second cache area of the host computer in the form of files, Store the relative path of the document in the attachment sub-data table of the project declaration database;

Duplicate checking and judging module is used to read the information of the declared project in the first cache area according to the main computer, and judge whether the data meets the requirements; For the image file in the cache area, use the image recognition module to read the content of the image file, and compare the handwriting of the person who completed the identification with the handwriting feature database. If the information and picture documents of the declared project in the project declaration data form meet the requirements, it will enter the screening module, and if it does not meet the requirements, the data will be collected again;

The output module is used to synchronize the project data meeting the requirements from the project application database to the review database; perform three screenings on the application information of the review database, and output the final data on the browsing page of the server.

The data processing system also includes a data allocation module, which is specifically used to store data directly related to the project into the master data table of the project declaration database, and store data indirectly related to the project into each sub-data table of the project declaration database, and the sub-data The tables are associated through the primary key of the project; the sub-data table of the declaration database also stores historical project declaration information.

In the described duplicate checking judgment module, specifically also include a selection module, the selection module is used to carry out word segmentation or direct matching to the data stored in each sub-data table of the project declaration library in the first buffer area read by the main computer, and to judge the current declaration by using Whether the project information is the name of the project or the name of the thesis, if it is, enter the keyword comparison module, if not, then judge whether it is the name and ID number or intellectual property number of the person who completed the project, if yes, enter the direct matching module;

The keyword comparison module is used to compare the project name or thesis name of the current application project information with the keywords of the project name or thesis name of another application project in the historical project application information, if the similarity is not less than Set the value to determine the repetition, otherwise it will not repeat;

The direct matching module is used to compare the completer's name and ID number or intellectual property number of the information item of the current application project with the project completion person's name, ID number or intellectual property number of another application item in the historical project application information Match directly, if the same judgment is repeated, otherwise it will not be repeated;

The storage module is used to store the information of non-duplicate declaration items determined by the keyword comparison module and the direct matching module in the cache area of the host computer, and store the information of the duplicate declaration items in the duplicate checking table of the item declaration database in the host computer .

The keyword comparison module specifically includes:

The keyword extraction module is used to extract one of the information of the current declared project and the declared information of historical projects, segment the extracted information, and store the decomposed keywords into two arrays corresponding to the project declaration database in the main computer ;

The similarity module is used to obtain the similarity between the information of the current declared project and the project information in the historical project declaration information, and compare the similarity with the set value. If the similarity is not lower than the set value, it is determined to be repeated, otherwise Not repeating.

The keyword extraction module extracts one of the information of the current declaration project and the declaration information of historical projects, performs word segmentation on the extracted information, and uses the ShootSearch component to do word segmentation.

The similarity module is specifically used to compare the number of keywords in the two arrays corresponding to the project declaration database, extract the smaller number of keywords, and divide the same number of keywords by the smaller number of keywords to obtain the similarity.

The direct matching module specifically includes:

The matching module is used for traversing and cyclically matching the declared project information collected by the remote data collection terminal with the historical project declared information in the sub-data table of the declared database, and judging whether they match.

The screening module specifically includes:

The screening storage module is used to call the scientific and technological project subject information and store it in the sub-data table of the review database in the form of data set A, and the expert subject information stored in the form of data set B is also stored in the sub-data table of the review database;

The first-level subject extraction module is used to select expert subject information in data set B as a condition factor, and judge whether the condition factor of the expert subject information is a first-level subject, and if so, combine the condition factor with the condition factor of data set A The key factor matching of the scientific and technological project subject information stored in the form, traverses whether the conditional factors in the expert subject information contain the key factor, if it is included, it will match, otherwise it will not match;

The second-level subject extraction module is used to judge whether the conditional factor of the expert information is a second-level subject, and if so, match the conditional factor with the key factor of the scientific and technological project information stored in the form of data set A, and traverse the expert information Whether the condition factor of contains the key factor, if it contains, it will match, otherwise it will not match;

The third-level subject extraction module is used to judge whether the conditional factor of the expert information is a third-level subject, and if so, match the conditional factor with the key factor of the scientific and technological project information stored in the form of data set A, and traverse the expert information Whether the condition factor of contains the key factor, if it contains, it will match, otherwise it will not match;

A random data generation module, used to randomly select the number of experts matching the scientific and technological project from the stored data set B of expert subject information, and store the retrieved data in the review database;

The random data generation module specifically selects data according to random numbers from zero to a specified number returned by the Random (random) function until the specified number is selected.

The screening module includes a primary screening module, a secondary screening module, and a tertiary screening module. The primary screening module is used for the first screening of projects that meet the requirements, and selects corresponding network review experts according to project information. Synchronize the data in the review database for screening; store the screened declared project data in the third cache area in the review database;

The secondary screening module is used to perform a second screening of the declared project data stored in the third buffer area; select senior experts according to the project information, conduct expert voting on the data taken out of the primary screening module, and take out the data from the voting results Preliminary award-winning projects, store the data of preliminary award-winning projects in the fourth cache area in the review database;

The third screening module is used for the third screening of the declared project data stored in the fourth buffer area; according to the project information, the experts of the Science and Technology Committee are selected, and the data taken out of the secondary screening module is reviewed by experts. The final award-winning project is taken out, and the final award-winning project is stored in the fifth buffer area in the review database; the data of the final award-winning item stored in the fifth buffer area in the review database is output.

Directly related data include the basic information of the project; indirectly related data include project introduction, major technological innovations, third-party evaluations, social and economic benefits, scientific and technological awards received, completers, completed units, opinions of recommending units, intellectual property rights, papers, and Attachments; collected data include: basic project information, project brief introduction, major technological innovations, third-party evaluations, social and economic benefits, scientific and technological awards received, completers, completed units, recommendations from recommending units, intellectual property rights, papers and annexes; The basic information of the project includes: project name, project discipline, technical field, project source and national economic industry, etc.; the intellectual property rights include: intellectual property number, intellectual property name, intellectual property owner and time when the intellectual property was obtained; the completed person Including: the name of the person who completed it, the ID number and the order of the person who completed it.

The data storage form in the project collection module includes: data sheets, pictures, Word documents and PDF documents, which are stored as data in different formats. Original format for easy viewing.

The primary screening module, the secondary screening module and the tertiary screening module are different, and the evaluation of data is different. The primary screening module is used for scoring, the secondary screening module is used for voting and grading, and the tertiary screening module is used for voting. But the process of selecting experts in the primary screening module, secondary screening module and tertiary screening module is the same.

Beneficial effects of the present invention:

The advantage of data judgment is that the system automatically extracts duplicate checking factors, performs complex matching calculations, reduces human factors, improves the fairness and correctness of duplicate checking results, reduces the workload of staff, and greatly improves work efficiency. Data plagiarism check is an important link in the entire science and technology reward system. The data processing process is complicated and the processing methods are different. Different methods are adopted for different data to check for duplicates, which is to avoid false checks and missed checks. The completer's name, ID number and intellectual property number are complete identifications, and there is no possibility of combination. Project names and thesis titles can be decomposed and reorganized, and word segmentation can be used to calculate the similarity to query duplicate items more accurately.

The advantage of data screening through network review is to reduce work costs, and the screening process reduces the interference of other factors, making the screening more objective. The advantage of expert selection is that experts are randomly selected and related to the subject of the project and the subject of the expert, so that the selection of experts is fair and just, the experts are more targeted, and the quality of the review results is higher.

Description of drawings

Fig. 1 is a schematic diagram of the main process of the present invention;

Fig. 2 is a schematic diagram of a data judgment flow chart of the present invention;

Fig. 3 is a schematic flow chart of data judging keyword comparison of the present invention;

Fig. 4 is a schematic diagram of the data judging direct matching process of the present invention;

Fig. 5 is a schematic diagram of the expert selection process of the present invention.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

As shown in Figure 1, a data processing method with an intelligent detection function comprises the following steps:

Step 1: System detection, use a browser to open the website, automatically detect the browser version used by the current user through JS script, give a prompt for non-IE series browsers, and close it; detect that the browser used meets the requirements and then enter the system for data collection;

Step 2: Collect the data of the declared project; store the handwriting of the person who completed the collection by the remote data collection terminal through the handwriting board into the handwriting feature database, and store the collected data directly related to the project into the master data table of the project declaration database. Each piece of data corresponds to Unique project number, data indirectly related to the project is stored in each sub-data table of the project declaration database, and the project number is set in the sub-data table as a foreign key to associate with the main data table; the collected pictures, Word and PDF document data are stored in the form of files To the server cache area, store the relative path of the document in the attachment sub-data table of the project declaration library;

Step 3: Read the data in the main data table and each sub-data table of the project declaration database, and judge whether the data meets the requirements; for image documents, use the document path in the sub-data table of the project declaration database to call the image file in the server cache area, and use image recognition The module reads the content of the picture document, and compares the handwriting of the person who completes it with the handwriting feature database. Whether the seal of the completion unit matches the name of the completion unit in the data sheet of the completion unit, if the handwriting information in the handwriting feature database and the declaration item in the project declaration If the information and image files meet the requirements, go to step 4; if not, go back to step 2;

Step 4: Synchronize the project data that meets the requirements in step 2 from the project application database to the review database. The synchronization process is one-way synchronization;

Step 5: Screen the projects that meet the requirements multiple times, select the corresponding network review experts according to the project information, and screen the data synchronized to the review database through network review; store the screened declared project data in the review database. The first cache area; perform a second screening of the declared project data stored in the first cache area; select senior experts according to the project information, conduct expert voting on the data taken out in step 4, and take out preliminary award-winning projects from the voting results, Store the data of the preliminary award-winning projects in the second buffer area of the review database; perform the third data screening on the declared project data stored in the second buffer area; select the experts of the Science and Technology Committee according to the project information, The data is reviewed by experts, and the final award-winning projects are taken out from the review results, and the final award-winning projects are stored in the third cache area in the review database;

Step 6: Output the data of the final award-winning project stored in the third cache area in the review database.

In said step 2, the directly related data includes the basic information of the project;

Indirectly related data include project introduction, major technological innovations, third-party evaluations, social and economic benefits, scientific and technological awards received, completers, completed units, opinions of recommending units, intellectual property rights, papers and annexes;

The collected data include: basic project information, project brief introduction, major technological innovations, third-party evaluations, social and economic benefits, scientific and technological awards received, completers, completed units, opinions of recommending units, intellectual property rights, papers and annexes;

The basic information of the project includes: project name, project discipline, technical field, project source and national economic industry to which it belongs, etc.;

The intellectual property rights include: intellectual property number, intellectual property name, intellectual property owner and acquisition time of intellectual property rights;

The completer includes: the completer's name, ID number and sequence of completers.

Data storage form in described step 2 comprises: data table, picture, Word document and PDF document, store as different format data, data table is convenient to query, statistics, picture and PDF document ensure data authenticity, and Word document is in order to keep data original Formatted for easy viewing.

In the third step, word segmentation or direct matching is performed on the data stored in the second step, and the similarity is calculated. If the similarity is lower than the set value, it is judged as meeting the requirements; otherwise, it is judged as not meeting the requirements. The similarity calculation method is used to check duplicate items to a greater extent to avoid repeated declarations of recombined items; the word segmentation data includes the project name and the title of the paper; the direct matching data includes: the author's Name and ID number and intellectual property number.

The specific process in step 5 is: setting the weight of scoring indicators, selecting experts, performing expert scoring on the data synchronized in step 3, summing up the expert scores according to the weights to obtain project scores, and sorting according to the project scores from high to low , fetch the set number of items from the sorted results.

The voting in the step 5 includes: first prize, second prize, third prize and no award.

The review in Step 5 includes objection and non-objection.

In the sixth step, the data filtered out in the sixth step is output, and the output form is a document in Word format. The document output in Word format is convenient for users to adjust the data format by themselves.

As shown in Figure 2, the third step is to obtain the data in the main data table and each sub-data table of the project declaration database, and the steps of judging whether the data meets the requirements include:

Step (3-1): Factors are extracted from the main data table and each sub-data table of the project declaration database; the factors include: project name, project completer name and ID number, intellectual property number, and name of thesis;

Step (3-2): Determine whether the factors are the project name and the name of the thesis, if yes, go to step (3-3), otherwise go to step (3-4);

Step (3-3): Compare the project name and the name of the thesis with the keywords in another project, if they are the same, judge that the project is repeated, otherwise the project is not repeated;

Step (3-4): directly match the name, ID number, and intellectual property number of the person completing the project with another project. If they are the same, determine that the project is duplicated; otherwise, the project is not repeated;

Step (3-5): result storage; the comparison and matching results of step (3-3) and step (3-4) are stored in the project declaration database.

Data plagiarism check is an important link in the entire science and technology reward system. The data processing process is complicated and the processing methods are different. Different methods are adopted for different data to check for duplicates, which is to avoid false checks and missed checks. The completer's name, ID number and intellectual property number are complete identifications, and there is no possibility of combination. Project names and thesis titles can be decomposed and reorganized, and word segmentation can be used to calculate the similarity to query duplicate items more accurately.

As shown in Figure 3, the specific steps of described step (3-3) keyword comparison include:

Step (3-3-1): Take out the project name or thesis title of the current project and another project from the project application database; the other project refers to other projects in the current year or one of all projects in the previous three years;

Step (3-3-2): extract keywords; use the ShootSearch component word segmentation for the data extracted in step (3-3-1), and store the decomposed keywords in an array;

Step (3-3-3): Keyword matching; The two arrays obtained in step (3-3-2) are compared in a traversal loop to obtain the number of identical keywords and the number of keywords in each group;

Step (3-3-4): Calculate the similarity; compare the number of keywords in the two groups obtained in step (3-3-3), take out the smaller number of keywords, and divide the same number of keywords by the smaller keyword Count the similarity;

Step (3-3-5): judge whether to repeat; compare the similarity obtained in step (3-3-4) with the set value, if the similarity is not lower than the set value, determine to repeat, otherwise do not repeat;

Step (3-3-6): data storage; store the weight judgment data in step (3-3-5) into the project declaration database.

As shown in Figure 4, the step of described step (3-4) direct matching comprises:

Step (3-4-1): Take out the current project and another project completer's name and ID number or intellectual property number from the project declaration database; said another item: when the factor is the completer's name or ID number, Refers to one of the other projects in the current year; when the factor is the intellectual property number, it refers to one of the other projects in the current year or one of all the projects in the previous three years.

Step (3-4-2): direct matching; directly compare the data taken out in step (3-4-1);

Step (3-4-3): Determine whether to repeat; repeat the same determination in step (3-4-2), otherwise do not repeat;

Step (3-4-4): data storage, storing the weight judgment data in step (3-4-3) into the project declaration database.

As shown in Figure 5, the selection process of experts in the fifth step is as follows:

Step (I): Take out the key factor from the data set A, which is the subject set of the project, that is, the first subject of the project, and take out factor 1, which is the first subject, factor 2, which is the second subject, and factor 3, which is third subject;

Step (II): select condition factors; the condition factors include: the first subject, the second subject and the third subject;

Step (III): determine whether the conditional factor is factor 1, i.e. the first subject, if so, enter step (IV) and make n=1, otherwise enter step (V);

Step (IV): Match the factor n with the key factor; traverse whether the key factor is included in the factor n, if it is included, match, otherwise it will not match; if it matches, enter step (VI), otherwise return to step (II);

Step (V): determine whether the condition factor is factor 2, i.e. the second subject, if so, enter step (IV) and make n=2, otherwise enter step (IV) and make n=3;

Step (VI): select matching data from data set B; select data according to step (IV) matching result;

Step (VII): randomly select the specified number of data; according to the random number from zero returned by the Random function to the specified bibliography, select data until the specified number is selected;

Step (VIII): data storage; store the data retrieved in step (VII) into the review database.

A data processing system with intelligent detection function, comprising

The system verification module is used to automatically detect the browser version used by the current user through JS scripts, give a prompt for non-IE series browsers, and close them; if the browser meets the requirements, it will enter the system for data collection;

The remote data collection terminal is used to store the data information of the collected declaration items in the cache area of the network server, and store the handwriting of the person who completed the collection by the remote data collection terminal through the handwriting board into the handwriting feature database;

The host computer is used to call the data information of the declared items in the cache area of the network server and store it in the first cache area of the host computer; the collected pictures, Word and PDF document data are stored in the second cache area of the host computer in the form of files, Store the relative path of the document in the attachment sub-data table of the project declaration database;

Duplicate checking and judging module is used to read the information of the declared project in the first cache area according to the main computer, and judge whether the data meets the requirements; For the image file in the cache area, use the image recognition module to read the content of the image file, and compare the handwriting of the person who completed the identification with the handwriting feature database. If the information and picture documents of the declared project in the project declaration data form meet the requirements, it will enter the screening module, and if it does not meet the requirements, the data will be collected again;

The output module is used to synchronize the project data meeting the requirements from the project application database to the review database; perform three screenings on the application information of the review database, and output the final data on the browsing page of the server.

The data processing system also includes a data allocation module, which is specifically used to store data directly related to the project into the master data table of the project declaration database, and store data indirectly related to the project into each sub-data table of the project declaration database, and the sub-data The tables are associated through the primary key of the project; the sub-data table of the declaration database also stores historical project declaration information.

In the described duplicate checking judgment module, specifically also include a selection module, the selection module is used to carry out word segmentation or direct matching to the data stored in each sub-data table of the project declaration library in the first buffer area read by the main computer, and to judge the current declaration by using Whether the project information is the name of the project or the name of the thesis, if it is, enter the keyword comparison module, if not, then judge whether it is the name and ID number or intellectual property number of the person who completed the project, if yes, enter the direct matching module;

The keyword comparison module is used to compare the project name or thesis name of the current application project information with the keywords of the project name or thesis name of another application project in the historical project application information, if the similarity is not less than Set the value to determine the repetition, otherwise it will not repeat;

The direct matching module is used to compare the completer's name and ID number or intellectual property number of the information item of the current application project with the project completion person's name, ID number or intellectual property number of another application item in the historical project application information Match directly, if the same judgment is repeated, otherwise it will not be repeated;

The storage module is used to store the information of non-duplicate declaration items determined by the keyword comparison module and the direct matching module in the cache area of the host computer, and store the information of the duplicate declaration items in the duplicate checking table of the item declaration database in the host computer .

The keyword comparison module specifically includes:

The keyword extraction module is used to extract one of the information of the current declared project and the declared information of historical projects, segment the extracted information, and store the decomposed keywords into two arrays corresponding to the project declaration database in the main computer ;

The similarity module is used to obtain the similarity between the information of the current declared project and the project information in the historical project declaration information, and compare the similarity with the set value. If the similarity is not lower than the set value, it is determined to be repeated, otherwise Not repeating.

The keyword extraction module extracts one of the information of the current declaration project and the declaration information of historical projects, performs word segmentation on the extracted information, and uses the ShootSearch component to do word segmentation.

The similarity module is specifically used to compare the number of keywords in the two arrays corresponding to the project declaration database, extract the smaller number of keywords, and divide the same number of keywords by the smaller number of keywords to obtain the similarity.

The direct matching module specifically includes:

The matching module is used for traversing and cyclically matching the declared project information collected by the remote data collection terminal with the historical project declared information in the sub-data table of the declared database, and judging whether they match.

The screening module specifically includes:

The screening storage module is used to call the scientific and technological project subject information and store it in the sub-data table of the review database in the form of data set A, and the expert subject information stored in the form of data set B is also stored in the sub-data table of the review database;

The first-level subject extraction module is used to select expert subject information in data set B as a condition factor, and judge whether the condition factor of the expert subject information is a first-level subject, and if so, combine the condition factor with the condition factor of data set A The key factor matching of the scientific and technological project subject information stored in the form, traverses whether the conditional factors in the expert subject information contain the key factor, if it is included, it will match, otherwise it will not match;

The second-level subject extraction module is used to judge whether the conditional factor of the expert information is a second-level subject, and if so, match the conditional factor with the key factor of the scientific and technological project information stored in the form of data set A, and traverse the expert information Whether the condition factor of contains the key factor, if it contains, it will match, otherwise it will not match;

The third-level subject extraction module is used to judge whether the conditional factor of the expert information is a third-level subject, and if so, match the conditional factor with the key factor of the scientific and technological project information stored in the form of data set A, and traverse the expert information Whether the condition factor of contains the key factor, if it contains, it will match, otherwise it will not match;

A random data generation module, used to randomly select the number of experts matching the scientific and technological project from the stored data set B of expert subject information, and store the retrieved data in the review database;

The random data generation module specifically selects data according to random numbers from zero to a specified number returned by the Random (random) function until the specified number is selected.

The screening module includes a primary screening module, a secondary screening module, and a tertiary screening module. The primary screening module is used for the first screening of projects that meet the requirements, and selects corresponding network review experts according to project information. Synchronize the data in the review database for screening; store the screened declared project data in the third cache area in the review database;

The secondary screening module is used to perform a second screening of the declared project data stored in the third buffer area; select senior experts according to the project information, conduct expert voting on the data taken out of the primary screening module, and take out the data from the voting results Preliminary award-winning projects, store the data of preliminary award-winning projects in the fourth cache area in the review database;

The third screening module is used for the third screening of the declared project data stored in the fourth buffer area; according to the project information, the experts of the Science and Technology Committee are selected, and the data taken out of the secondary screening module is reviewed by experts. The final award-winning project is taken out, and the final award-winning project is stored in the fifth buffer area in the review database; the data of the final award-winning item stored in the fifth buffer area in the review database is output.

Directly related data include the basic information of the project; indirectly related data include project introduction, major technological innovations, third-party evaluations, social and economic benefits, scientific and technological awards received, completers, completed units, opinions of recommending units, intellectual property rights, papers, and Attachments; collected data include: basic project information, project brief introduction, major technological innovations, third-party evaluations, social and economic benefits, scientific and technological awards received, completers, completed units, recommendations from recommending units, intellectual property rights, papers and annexes; The basic information of the project includes: project name, project discipline, technical field, project source and national economic industry, etc.; the intellectual property rights include: intellectual property number, intellectual property name, intellectual property owner and time when the intellectual property was obtained; the completed person Including: the name of the person who completed it, the ID number and the order of the person who completed it.

The data storage form in the project collection module includes: data sheets, pictures, Word documents and PDF documents, which are stored as data in different formats. Original format for easy viewing.

The primary screening module, the secondary screening module and the tertiary screening module are different, and the evaluation of data is different. The primary screening module is used for scoring, the secondary screening module is used for voting and grading, and the tertiary screening module is used for voting. But the process of selecting experts in the primary screening module, secondary screening module and tertiary screening module is the same.

The advantage of the data judgment of the present invention is that the system automatically extracts duplicate checking factors, performs complex matching calculations, reduces human factors, improves the fairness and correctness of duplicate checking results, reduces the workload of staff, and greatly improves work efficiency. Data plagiarism check is an important link in the entire science and technology reward system. The data processing process is complicated and the processing methods are different. Different methods are adopted for different data to check for duplicates, which is to avoid false checks and missed checks. The completer's name, ID number and intellectual property number are complete identifications, and there is no possibility of combination. Project names and thesis titles can be decomposed and reorganized, and word segmentation can be used to calculate the similarity to query duplicate items more accurately.

The advantage of data screening through network review is to reduce work costs, and the screening process reduces the interference of other factors, making the screening more objective. The advantage of expert selection is that experts are randomly selected and related to the subject of the project and the subject of the expert, so that the selection of experts is fair and just, the experts are more targeted, and the quality of the review results is higher.

Although the specific implementation of the present invention has been described above in conjunction with the accompanying drawings, it does not limit the protection scope of the present invention. Those skilled in the art should understand that on the basis of the technical solution of the present invention, those skilled in the art do not need to pay creative work Various modifications or variations that can be made are still within the protection scope of the present invention.

Claims (10)

1. a data processing method with intelligent detecting function, is characterized in that, comprises the steps:

Step 1: automatically detect active user by JS script and use browser version, point out for non-IE series browser, and close; Detect and use the browser i.e. system of entering that meets the requirements to carry out data acquisition;

Step 2: remote data acquisition terminal is completed to people's person's handwriting by handwriting pad collection and be stored to handwriting characteristic storehouse, the data information memory of the project of declaring gathering is to the buffer area of the webserver, principal computer call the buffer area of the webserver the project of declaring data message and store the first buffer area of principal computer into; Picture, Word and the PDF document data gathering is stored to the second buffer area of principal computer with document form, document relative path is stored in project application database attachments subdatasheet;

Step 3: principal computer reads the information of the project of declaring in the first buffer area, judges whether data meet the requirements; For photo-document, utilize document path in project application database attachments subdatasheet to call the second buffer area photo-document of principal computer, utilize picture recognition module to read photo-document content, identify people's person's handwriting and compare in handwriting characteristic storehouse, complete unit stamp and complete flat data table and complete organization and whether conform to, all meet the requirements and just enter step 4 if declare the information of project and photo-document in handwriting characteristic storehouse in handwriting information, project application tables of data, just return to step 2 if do not met;

Step 4: from project application database synchronization to review data storehouse, this synchronizing process is taked one-way synchronization by satisfactory project data in step 3; The information of declaring in review data storehouse is carried out to three screenings, and final data is exported on the browsing pages of server;

In described step 2, be specially: the data directly related with project are stored in the master data sheet of project application storehouse, every corresponding unique bullets of data, store in the each subdatasheet in project application storehouse with the data of project indirect correlation, in subdatasheet, to be set to external key associated with master data sheet for unique bullets; In the subdatasheet in declaration data storehouse, also store history item and declare information.

2. a kind of data processing method with intelligent detecting function as claimed in claim 1, it is characterized in that, in described step 2, the data of the each subdatasheet storage in project application storehouse are carried out participle or directly mated, utilize judge module to judge whether the information of the current project of declaring is project name or paper treatise title, if just enter keyword comparison module, if not, what determine whether project completes people's name and identification card number or intellecture property number again, if so, enter direct matching module;

Utilize keyword comparison module that the project name of the information of the current project of declaring or paper treatise title and history item are declared to another in information and declare the project name of project or the keyword comparison of paper treatise title, if similarity is not less than setting value, judge and repeat, otherwise do not repeat;

Utilizing direct matching module that people's name and identification card number or intellecture property number and history item another project of declaring project in information of declaring that completes of the information project of the current project of declaring is completed to people's name and identification card number or intellecture property number directly mates, if identical judgement repeats, otherwise does not repeat;

Utilize memory module by keyword comparison module and directly matching module judge that the information that does not repeat the project of declaring stores the buffer area of principal computer into, store the information of judging the project of declaring repeating into project application database in principal computer and look into heavily in table.

3. a kind of data processing method with intelligent detecting function as claimed in claim 2, it is characterized in that, described keyword comparison module is declared another in information by the project name of the information of the current project of declaring or paper treatise title and history item and is declared the project name of project or the keyword comparison of paper treatise title, if similarity is not less than setting value, judge and repeat, otherwise do not repeat, detailed process is:

Utilize keyword extracting module to take out current information and history item of declaring project and declare in information, the information of taking out is carried out to participle, the keyword decompositing is stored into respectively in two arrays corresponding to project declaration data storehouse in principal computer;

Utilize keyword matching module to travel through recycle ratio to the keyword in two arrays corresponding to project application database, draw same keyword number and respectively organize keyword number;

Show that according to similarity module current information of declaring project and history item declare the similarity of the project information in information, by this similarity and setting value comparison, if similarity is not less than setting value, judges and repeat, otherwise do not repeat;

Described current information and the history item of declaring project of keyword extracting module taking-up of utilizing declared in information, and the information of taking out is carried out to participle, adopts ShootSearch assembly participle;

Described similarity module is counted size specifically for keyword in two arrays corresponding to item compared declaration data storehouse, takes out less keyword number, draws similarity with same keyword number divided by less keyword number.

4. a kind of data processing method with intelligent detecting function as claimed in claim 1, is characterized in that, in described step 3, the information of declaring in review data storehouse is carried out to three screenings, is specially:

Project in satisfactory review data storehouse is screened for the first time, select corresponding network evaluation expert according to project information, by network evaluation, the data that are synchronized in review data storehouse are screened; Store the project data of declaring filtering out into the 3rd buffer area in review data storehouse;

The project data of declaring storing in the 3rd buffer area is carried out to programmed screening; Select Senior Expert according to project information, the data of screening for the first time taking-up are carried out to expert's ballot, from voting results, take out preliminary award-winning item, store the data of preliminary award-winning item into the 4th buffer area in review data storehouse;

The project data of declaring storing in the 4th buffer area is screened for the third time; Select the expert of Technology Committee according to project information, the data of taking out in programmed screening are carried out to expert's examination & verification, from auditing result, take out final award-winning item, store final award-winning item into the 5th buffer area in review data storehouse.

5. a kind of data processing method with intelligent detecting function as claimed in claim 1, is characterized in that, in described three screenings, expert's the process of selecting is specially:

S1: science and technology item discipline information is stored in the form of data set A in the subdatasheet in review data storehouse, also stores the expert's discipline information with the form storage of data set B in the subdatasheet in review data storehouse;

S2: choose expert's discipline information and as condition element in data set B, whether the condition element that judges this expert's discipline information is one-level subject, if, this condition element is mated with the key factor of the science and technology item discipline information of the form storage with data set A, whether the condition element in traversal expert discipline information comprises key factor, if comprised, and coupling, and enter step S5, otherwise do not mate, enter step S3;

S3: whether the condition element that judges this expert info is secondary subject, if, this condition element is mated with the key factor of the science and technology item information of the form storage with data set A, whether the condition element in traversal expert info comprises key factor, if comprised, coupling, enter step S5, otherwise do not mate, enter step S4;

S4: whether the condition element that judges this expert info is three grades of subjects, if, this condition element is mated with the key factor of the science and technology item information of the form storage with data set A, whether the condition element in traversal expert info comprises key factor, if comprised, coupling, enter S5, otherwise do not mate;

S5: choose at random the expert's quantity matching with science and technology item from the data set B of expert's discipline information of storage, and the data of taking-up are stored to review data storehouse.

6. a kind of data handling system with intelligent detecting function as claimed in claim 1, is characterized in that, comprising:

System check module, uses browser version for automatically detecting active user by JS script, points out, and close for non-IE series browser; Detect and use the browser i.e. system of entering that meets the requirements to carry out data acquisition;

Remote data acquisition terminal, for the buffer area to the webserver by the data information memory of the project of declaring gathering, completes people's person's handwriting by remote data acquisition terminal by handwriting pad collection and is stored to handwriting characteristic storehouse;

Principal computer, for call the webserver buffer area the project of declaring data message and store the first buffer area of principal computer into; Picture, Word and the PDF document data gathering is stored to the second buffer area of principal computer with document form, document relative path is stored in project application database attachments subdatasheet;

Look into heavy judge module, for read the information of the project of declaring of the first buffer area according to principal computer, judge whether data meet the requirements; For photo-document, utilize document path in project application database attachments subdatasheet to call the second buffer area photo-document of principal computer, utilize picture recognition module to read photo-document content, identify people's person's handwriting and compare in handwriting characteristic storehouse, complete unit stamp and complete flat data table and complete organization and whether conform to, all meet the requirements and just enter screening module if declare the information of project and photo-document in handwriting characteristic storehouse in handwriting information, project application tables of data, if do not met with regard to Resurvey data;

Output module, for by satisfactory project data from project application database synchronization to review data storehouse; The information of declaring in review data storehouse is carried out to three screenings, and final data is exported on the browsing pages of server;

Described data handling system also comprises data allocations module, specifically for the data directly related with project being stored in the master data sheet of project application storehouse, store in the each subdatasheet in project application storehouse with the data of project indirect correlation, between subdatasheet, pass through the association of project major key; In the subdatasheet in declaration data storehouse, also store history item and declare information.

7. a kind of data handling system with intelligent detecting function as claimed in claim 6, it is characterized in that, described looking in heavy judge module, specifically also comprise selection module, select module carry out participle or directly mate for principal computer being read to the data of the each subdatasheet storage in project application storehouse of the first buffer area, utilize and judge whether the information of the current project of declaring is project name or paper treatise title, if just enter keyword comparison module, if not, what determine whether project completes people's name and identification card number or intellecture property number again, if, enter direct matching module,

Keyword comparison module, declare the project name of project or the keyword comparison of paper treatise title for another that the project name of the information of the current project of declaring or paper treatise title and history item are declared to information, if similarity is not less than setting value, judge and repeat, otherwise do not repeat;

Directly matching module, for complete people's name and identification card number or intellecture property number and the history item of the information project of the current project of declaring are declared to another project of declaring project of information and are completed people's name and identification card number or intellecture property number and directly mate, if identical judgement repeats, otherwise does not repeat;

Memory module, for by keyword comparison module and directly matching module judge that the information that does not repeat the project of declaring stores the buffer area of principal computer into, store the information of judging the project of declaring repeating into project application database in principal computer and look into heavily in table.

8. a kind of data handling system with intelligent detecting function as claimed in claim 7, is characterized in that, described keyword comparison module, specifically comprises:

Keyword extracting module, declares of information for taking out current information of declaring project and history item, and the information of taking out is carried out to participle, and the keyword decompositing is stored into respectively in two arrays corresponding to project declaration data storehouse in principal computer;

Similarity module, for showing that current information of declaring project and history item declare the similarity of the project information of information, by this similarity and setting value comparison, if similarity is not less than setting value, judges and repeats, otherwise do not repeat;

Described current information and the history item of declaring project of keyword extracting module taking-up of utilizing declared in information, and the information of taking out is carried out to participle, adopts ShootSearch assembly participle;

Described similarity module is counted size specifically for keyword in two arrays corresponding to item compared declaration data storehouse, takes out less keyword number, draws similarity with same keyword number divided by less keyword number.

9. a kind of data handling system with intelligent detecting function as claimed in claim 6, it is characterized in that, described screening module comprises primary screening module, postsearch screening module and three screening modules, described primary screening module is for screening for the first time satisfactory project, select corresponding network evaluation expert according to project information, by network evaluation, the data that are synchronized in review data storehouse are screened; Store the project data of declaring filtering out into the 3rd buffer area in review data storehouse;

Described postsearch screening module is for carrying out programmed screening to the project data of declaring that stores the 3rd buffer area into; Select Senior Expert according to project information, the data of taking out in primary screening module are carried out to expert's ballot, from voting results, take out preliminary award-winning item, store the data of preliminary award-winning item into the 4th buffer area in review data storehouse;

Described three screening modules are for carrying out data and screen for the third time the project data of declaring that stores the 4th buffer area into; Select the expert of Technology Committee according to project information, the data of taking out in postsearch screening module are carried out to expert's examination & verification, from auditing result, take out final award-winning item, store final award-winning item into the 5th buffer area in review data storehouse; By the data output of the final award-winning item in the 5th buffer area being stored in review data storehouse.

10. a kind of data handling system with intelligent detecting function as claimed in claim 6, is characterized in that, described screening module, specifically comprises:

Screening memory module, for calling science and technology item discipline information and being stored in the subdatasheet in review data storehouse with the form of data set A, also stores the expert's discipline information with the form storage of data set B in the subdatasheet in review data storehouse;

One-level subject extraction module, for choosing expert's discipline information and as condition element at data set B, whether the condition element that judges this expert's discipline information is one-level subject, if, this condition element is mated with the key factor of the science and technology item discipline information of the form storage with data set A, whether the condition element in traversal expert discipline information comprises key factor, if comprised, and coupling, otherwise do not mate;

Secondary subject extraction module, for judging whether the condition element of this expert info is secondary subject, if, this condition element is mated with the key factor of the science and technology item information of the form storage with data set A, whether the condition element in traversal expert info comprises key factor, if comprised, coupling, otherwise do not mate;

Three grades of subject extraction modules, for judging whether the condition element of this expert info is three grades of subjects, if, this condition element is mated with the key factor of the science and technology item information of the form storage with data set A, whether the condition element in traversal expert info comprises key factor, if comprised, coupling, otherwise do not mate;

Random data generation module, chooses at random for the data set B of the expert's discipline information from storage the expert's quantity matching with science and technology item, and the data of taking-up is stored to review data storehouse.