CN110807060A - Education big data analysis system - Google Patents

Abstract

The invention relates to the field of educational big data analysis, in particular to an educational big data analysis system, comprising an educational data acquisition module, an educational data sorting module, an educational data mining module and an educational data analysis module. The invention takes students' academic performance as a data source, and uses data mining technology to provide an efficient processing method for the rational and effective utilization of educational big data. It not only involves the application of artificial intelligence and statistics, but also involves the management and use of databases. In the analysis of student performance, make the data more time-sensitive and clear It can be deployed in the management centers of various colleges and universities, and has the advantages of complete functions and reliable performance.

Description

An educational big data analysis system

technical field

The invention relates to the technical field of educational big data analysis, in particular to an educational big data analysis system.

Background technique

Colleges and universities evaluate students' academic performance and comprehensive quality based on the students' test scores in various subjects as important indicators. After long-term operation, colleges and universities have accumulated and stored a large amount of student score information, but colleges and universities do not pay much attention to these scores. The analysis and processing of grades are generally still in the age of ancient queries and statistics, such as counting the number of outstanding, good, passing, and failing grades; calculating the average score, standard deviation, calculating the grade point; calculating the grade point. However, there is no in-depth understanding of the relationship between these grades obtained by students and the relationship between courses, and it is not found that these stored grades are an important basis for scheduling courses. The manual course arrangement method still adopted is compiled by the dean of teaching or each department head. They rely on years of teaching experience, combined with relevant regulations, to decide which courses to open to students and the sequence of courses. This inevitably involves a certain degree of subjectivity and ignores the valuable resource of student grades accumulated over the years.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide an educational big data analysis system to solve one or some of the defects proposed in the above background art.

To achieve the above object, the present invention provides the following technical solutions:

An educational big data analysis system, comprising an educational data acquisition module, an educational data sorting module, an educational data mining module and an educational data analysis module, the modules are communicated and connected, and the educational data acquisition module collects educational big data and sends it to the educational data sorting Module, the education data sorting module preprocesses the education big data, cleans the obtained data according to the preset standard format, filters out redundant information, and stores the education big data of different attributes and formats as corresponding templates according to the attributes. The data is stored in the format, and the identified types are labeled with classification labels to obtain classified data. The education data mining module mines various label data in the database, retrieves all the frequent itemsets in the education database, and uses the frequent itemsets to construct a Satisfying the rule of minimum confidence, the educational data analysis module accepts analysis requests of different dimensions submitted by users, extracts the stored data in the educational data fusion storage system for data analysis, and displays the data analysis results visually.

Preferably, the education data collection module includes a network port, a campus port and a manual port. The network port is connected to the cloud server and is used to collect educational big data on the network platform. The campus port is connected to campus management equipment and collects campus management equipment records and records. The generated educational big data, the manual port is used to manually supplement the default data.

Preferably, the collection methods of the education data collection module include network terminal collection, online campus data extraction and manual input.

Preferably, when the classified data is audio data, image data and/or text data, the extraction sub-module is further configured to extract the first data representing student behavior during the class from the audio data, image data and/or text data. The first data includes: learning status, interest questionnaire, interview records, content of responses/questions, times of answering questions and/or homework completion; and extracting second data representing teacher behavior, the second data includes: teaching methods, Interest questionnaire, interview records, content of responses/questions, number of questions asked, teaching progress and/or assignments.

Preferably, the educational data mining module uses association rules and Apriori algorithm to preprocess the classified data.

Preferably, association rules are used to reflect the interdependence and correlation between one data and other data, specifically:

的逻辑蕴含式，规定

在数据集D中支持度是数据集中同时包含X和Y的数据数与所有数据数之比，反应规则的可靠程度记为support Let I={i ₁ , _{i 2} , _{i 3} ,...,in } be the set of data, in is the data, D is the set of database T, T is the unique data number of each data, let X, Y is a set of data in I, and X∩Y=Ф an association rule of the form

The logical implication of , states that

In data set D, the support degree is the ratio of the number of data containing both X and Y to the number of all data in the data set, and the reliability of the reaction rule is recorded as support

＝P(X∪Y)，and support

=P(X∪Y),

If the data set exceeds the minimum support threshold given by the user, the data set is a frequent data set, and the degree of certainty of the response rule is confidence

＝P(Y|X)，and confidence

=P(Y|X),

A rule that satisfies both the minimum support threshold and the minimum confidence threshold at the same time becomes a strong rule. Given a data set D, mining association rules is to find the association rules whose support and confidence are greater than the minimum thresholds given by the user.

As a preference, the Apriori algorithm uses breadth-first iterative search, first finds the frequent 1-itemsets L1, uses L1 to find the frequent 2-itemsets L2, and so on, until all frequent itemsets are found, when a frequent item is found When the number of sets is zero, the calculation stops and finally outputs the frequent set of all items.

Preferably, the educational data analysis module includes a template library, a warehouse and a modeling platform, and the template library, the modeling platform and the warehouse are all connected to the cloud server.

Preferably, the template library includes an identification unit and a matching unit, the identification unit is provided with a fuzzy classifier for performing fuzzy classification on the analysis request, the identification unit is connected with the matching unit, the matching unit includes several matching templates, and the matching unit is connected to the modeling platform. Connect, send the processing information to the modeling platform. The modeling platform includes a modeling template and a data processing unit. The modeling template stores several standard modeling models. The data processing unit is respectively connected to the modeling template and the education data fusion storage system. A data processing unit is connected for storing the generated analysis results.

Preferably, the method for generating the analysis result of the educational data analysis module includes:

S1: Input analysis request;

S2: The template library generates processing information through the identification unit and the matching unit, and sends the processing information to the modeling platform;

S3: The modeling platform extracts the modeling model and generates an analysis request report;

S4: Send the analysis results to the warehouse.

Compared with the prior art, the beneficial effects of the present invention are:

1. This educational big data analysis system takes students' academic performance as the data source, selects multiple courses in the grade database as the research object, finds out whether a certain course has an impact on the opening of other courses, and provides the school's teaching teachers for future course arrangement. Reference, to provide a basis for future students to choose courses.

2. This educational big data analysis system uses data mining technology to provide efficient processing methods for the rational and effective use of educational big data. It not only involves the application of artificial intelligence and statistics, but also involves the management and use of databases, and is applied to student performance. In the analysis, make the data more timely and concise. In addition, integrate the data of network port, campus port and manual port for comprehensive analysis to obtain educational big data, organize and classify the obtained data, organize it into classified data that meets the preset requirements, and store the sorted classified data in the The database corresponding to the classified data is stored and managed in a unified manner, so that the data can be called later. When the data needs to be analyzed and counted, the data stored in the database will be called for in-depth analysis, and after the analysis results are obtained. , and provide targeted services according to the analysis results, so as to improve the quality and efficiency of students' learning and teachers' teaching, and achieve a multiplier effect.

3. As a system that can provide services for the needs of students and/or teachers, this educational big data analysis system can be deployed in the management centers of various colleges and universities, and can analyze the data collected by sensors in the Internet-based architecture. Data and manually imported data undergo in-depth processing including classification, storage, analysis, and service provision, and have the advantages of complete functions and reliable performance.

Description of drawings

1 is a schematic flow diagram of a module of the present invention;

FIG. 2 is a schematic diagram of the composition of the educational data analysis module of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Example 1

An educational big data analysis system, as shown in Figure 1, includes an educational data acquisition module 1, an educational data sorting module 2, an educational data mining module 3 and an educational data analysis module 4, and the communication connection between the modules, the educational data acquisition module 1. Collect educational big data and send it to the educational data sorting module 2. The educational data sorting module 2 preprocesses the educational big data, cleans the obtained data according to the preset standard format, filters out redundant information, and separates different attributes. The educational big data of the format is classified according to the attributes and stored as the storage data of the corresponding template format, and the identified types are labeled with the classification labels to obtain the classified data. The education data mining module 3 mines the various label data in the database, and retrieves the All frequent itemsets in the database are used to construct a rule that satisfies the minimum confidence degree. The educational data analysis module 4 accepts the analysis requests of different dimensions submitted by the user, and extracts the stored data in the educational data fusion storage system for data analysis. Visualize the data analysis results.

Further, the education data collection module 1 includes a network port, a campus port and an artificial port. The network port is connected to the cloud server and is used to collect educational big data on the network platform. The campus port is connected to campus management equipment and collects campus management equipment records. and the generated educational big data. The manual port is used to manually supplement the default data. The collection methods of the educational data collection module 1 include network terminal collection, online campus data extraction and manual input. It can integrate the network port, campus port and manual port. Comprehensive analysis of the data makes the data more comprehensive and the analysis results more accurate.

Specifically, when the classified data is audio data, image data and/or text data, the extraction sub-module is further configured to extract the first data representing student behavior during the class from the audio data, image data and/or text data. The first data includes: learning status, interest questionnaire, interview records, content of responses/questions, times of answering questions and/or homework completion; and extracting second data representing teacher behavior, the second data includes: teaching methods, Interest questionnaire, interview records, content of responses/questions, number of questions asked, teaching progress and/or assignments.

In addition, as shown in Figure 2, the educational data analysis module includes a template library, a warehouse and a modeling platform, and the template library, the modeling platform and the warehouse are all connected to the cloud server. The template library includes an identification unit and a matching unit. The identification unit is provided with a fuzzy classifier for fuzzy classification of the analysis request. The identification unit is connected with the matching unit. The matching unit includes several matching templates. The matching unit is connected with the modeling platform. The processing information is sent to the modeling platform. The modeling platform includes a modeling template and a data processing unit. The modeling template stores several standard modeling models. The data processing unit is respectively connected to the modeling template and the education data fusion storage system, and the warehouse is connected to the data processing. A cell that stores the generated analysis results.

The methods for generating the analysis results of the education data analysis module include:

S1: Input analysis request;

S2: The template library generates processing information through the identification unit and the matching unit, and sends the processing information to the modeling platform;

S3: The modeling platform extracts the modeling model and generates an analysis request report;

S4: Send the analysis results to the warehouse.

The educational big data analysis system of this embodiment utilizes data mining technology to provide an efficient processing method for rational and effective utilization of educational big data. It not only involves the application of artificial intelligence and statistics, but also involves the management and use of databases. It is applied to the analysis of student performance to make the data more timely and concise. Integrate the data of the network port, campus port and manual port for comprehensive analysis to obtain educational big data, organize and classify the obtained data, organize it into classified data that meets the preset requirements, and store the sorted classified data in the The database corresponding to the classification data described above is stored and managed in a unified manner, so that the data can be called later. When the data needs to be analyzed and counted, the data stored in the database will be called for in-depth analysis. After the analysis results are obtained, according to the The analysis results provide targeted services, thereby improving the quality and efficiency of students' learning and teachers' teaching, and achieving a multiplier effect. The educational big data analysis system of this embodiment is a system that can provide services for the needs of students and/or teachers. When in use, it can be deployed in the management centers of various colleges and universities, so that the data collected by sensors in the Internet-based architecture can be analyzed. In-depth processing including classification, storage, analysis, and provision of services is carried out on the data and manually imported data, and has the advantages of complete functions and reliable performance.

Example 2

As the second embodiment of the present invention, the educational data mining module 3 uses association rules and Apriori algorithm to preprocess the classified data.

Specifically, association rules are used to reflect the interdependence and correlation between one data and other data, specifically:

的逻辑蕴含式，规定

在数据集D中支持度是数据集中同时包含X和Y的数据数与所有数据数之比，反应规则的可靠程度记为support

Let I={i ₁ , _{i 2} , _{i 3} ,...,in } be the set of data, in is the data, D is the set of database T, T is the unique data number of each data, let X, Y is a set of data in I, and X∩Y=Ф an association rule of the form

The logical implication of , states that

In data set D, the support degree is the ratio of the number of data containing both X and Y to the number of all data in the data set, and the reliability of the reaction rule is recorded as support

＝PX∪Y，and support

=PX∪Y,

If the data set exceeds the minimum support threshold given by the user, the data set is a frequent data set, and the degree of certainty of the response rule is confidence

＝PY|X，and confidence

=PY|X,

A rule that satisfies both the minimum support threshold and the minimum confidence threshold at the same time becomes a strong rule. Given a data set D, mining association rules is to find the association rules whose support and confidence are greater than the minimum thresholds given by the user.

Apriori algorithm uses breadth-first iterative search, first finds frequent 1-itemsets L1, uses L1 to find frequent 2-itemsets L2, and so on, until all frequent itemsets are found, when the number of frequent itemsets is found If it is zero, the calculation stops, and the frequent set of all items is finally output. The Apriori algorithm is introduced into the field of database mining, and the data, items, basis sets, and association rules in database association mining are defined, and the support and confidence of the association rules are considered. degree to extract similar or similar knowledge points.

The educational big data analysis system of this embodiment takes students' academic achievements as the data source, selects multiple courses in the achievement database as the research objects, finds out whether a certain course has an impact on the opening of other courses, and arranges courses for the school teachers in the future Provide a reference for future students to choose courses.

The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above-mentioned embodiments, and the above-mentioned embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention, without departing from the spirit and scope of the present invention. Under the premise, the present invention will also have various changes and improvements, and these changes and improvements all fall within the scope of the claimed invention. The claimed scope of the present invention is defined by the appended claims and their equivalents.

Claims (10)

1. an educational big data analysis system, it is characterized in that: comprise educational data acquisition module (1), educational data sorting module (2), educational data mining module (3) and educational data analysis module (4), each module The educational data collection module (1) collects educational big data and sends it to the educational data sorting module (2), and the educational data sorting module (2) preprocesses the educational big data and parses the acquired data according to the preset standard format. The data is cleaned, the redundant information is filtered out, the educational big data of different attributes and formats are classified and stored as the storage data of the corresponding template format according to the attributes, and the identified types are labeled with the classification labels to obtain the classified data, and the educational data mining module ( 3) Mining all kinds of label data in the database, retrieving all the frequent itemsets in the education database, and using the frequent itemsets to construct the rules that satisfy the minimum confidence, the education data analysis module (4) accepts different dimensions submitted by users , extract the stored data in the educational data fusion storage system for data analysis, and visualize the data analysis results. 2. education big data analysis system according to claim 1, is characterized in that: education data acquisition module (1) comprises network port, campus port and artificial port, and network port is connected with cloud server, is used for collecting network platform The educational big data on the campus port and the campus management equipment collects the educational big data recorded and generated by the campus management equipment, and the manual port is used to manually supplement the default data. 3 . The educational big data analysis system according to claim 2 , wherein the collection methods of the education data collection module (1) include network terminal collection, online campus data extraction and manual input. 4 . 4. education big data analysis system according to claim 1, is characterized in that: when classification data is audio data, image data and/or text data, extract submodule, also be used for from audio data, image data and/or Extracting first data representing student behavior during class from the text data, the first data includes: learning status, interest questionnaire, interview records, content of responses/questions, times of answering questions and/or homework completion; and extracting representations The second data of the teacher's behavior, the second data includes: the way of teaching, the questionnaire of interest, the interview record, the content of the reply/question, the frequency of the question, the teaching progress and/or the arrangement of the homework. 5. The educational big data analysis system according to claim 1, wherein the educational data mining module (3) uses association rules and Apriori algorithm to preprocess the classified data. 6. The educational big data analysis system according to claim 5, wherein the association rule is used to reflect the interdependence and correlation between one data and other data, specifically: Let I={i ₁ , _{i 2} , _{i 3} ,...,in } be the set of data, in is the data, D is the set of database T, T is the unique data number of each data, let X, Y is a set of data in I, and X∩Y=Ф an association rule of the form

The logical implication of , states that

In data set D, the support degree is the ratio of the number of data containing both X and Y to the number of all data in the data set, and the reliability of the reaction rule is recorded as and If the data set exceeds the minimum support threshold given by the user, the data set is a frequent data set, and the degree of certainty of the reaction rule is

and

A rule that satisfies both the minimum support threshold and the minimum confidence threshold at the same time becomes a strong rule. Given a data set D, mining association rules is to find the association rules whose support and confidence are greater than the minimum thresholds given by the user. 7. The educational big data analysis system according to claim 5, is characterized in that: Apriori algorithm adopts breadth-first iterative search, first finds frequent 1-itemsets L1, uses L1 to find frequent 2-itemsets L2, and sequentially By analogy, until all frequent itemsets are obtained, when the number of frequent itemsets is found to be zero, the calculation stops, and finally the frequent sets of all items are output. 8. The educational big data analysis system according to claim 1, wherein the educational data analysis module (4) comprises a template library, a warehouse and a modeling platform, and the template library, the modeling platform and the warehouse are all connected to a cloud server. 9. The educational big data analysis system according to claim 8, wherein the template library comprises an identification unit and a matching unit, and the identification unit is provided with a fuzzy classifier for performing fuzzy classification on the analysis request, the identification unit and the matching unit The matching unit includes several matching templates. The matching unit is connected to the modeling platform and sends the processing information to the modeling platform. The modeling platform includes modeling templates and data processing units. The modeling template stores several standard modeling models. The data processing unit is respectively connected to the modeling template and the educational data fusion storage system, and the warehouse is connected to the data processing unit for storing the generated analysis results. 10. The education big data analysis system according to claim 9, wherein the method for generating the analysis result of the education data analysis module comprises: S1: Input analysis request; S2: The template library generates processing information through the identification unit and the matching unit, and sends the processing information to the modeling platform; S3: The modeling platform extracts the modeling model and generates an analysis request report; S4: Send the analysis results to the warehouse.

Images