CN-121256729-B - Student score report data management method and system

CN121256729BCN 121256729 BCN121256729 BCN 121256729BCN-121256729-B

Abstract

The invention relates to the technical field of electronic data processing, and discloses a student score report data management method and system, wherein students are divided into different student groups; the method comprises the steps of converting the achievements of students into achievement grade items, calculating item ordering indexes corresponding to student groups, initializing an FP tree, wherein nodes of the FP tree comprise item names, group counting vectors and learning period sets, arranging the achievement grade items in the student achievement records in a descending order and inserting the item names, the group counting vectors and the learning period sets of all the nodes into the FP tree, obtaining group counting vectors and the learning period sets of end nodes corresponding to each prefix path, judging whether a group dominance threshold is met, judging whether learning period distribution constraint is met, only preserving prefix path generation condition FP trees meeting the conditions, and recursively mining to obtain frequent modes. According to the invention, the student group characteristics, item set relevance and time sequence information are comprehensively analyzed under a unified frame, so that the depth and accuracy of the mining result are improved.

Inventors

ZHU DAHUA
XU SONGCHUAN
CHEN XIAOYU

Assignees

广东光大信息科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20251027

Claims (8)

1. A student score report data management method is characterized by comprising the following steps of dividing students into different student groups according to historical academic data and course selection behaviors, converting the course scores of the students into discrete score grade items, and calculating and obtaining item ordering indexes corresponding to the student groups according to the statistical distribution characteristics of the score grade items in the student groups aiming at the student groups; initializing an FP tree, wherein nodes of the FP tree comprise item names, group count vectors used for recording the group count of each student and a learning period set used for recording learning period information of a path where the nodes are located; Traversing each student score record, descending order the score grade items in the student score records according to item ordering indexes of student groups to which the student score records belong, inserting the item sequences subjected to descending order into the FP tree as paths, and updating group count vectors and school period sets of all nodes on the paths; When a conditional pattern base is constructed for items in a head list, a group count vector and a learning period set of end nodes corresponding to each prefix path are obtained, and whether the group count vector meets a preset group dominance threshold is judged, wherein the method comprises the steps of extracting the count value of a target mining group from the group count vector of the end nodes corresponding to the prefix paths for the target mining group And calculates the sum of the count values of all other groups When the inequality is satisfied When it is determined that the population dominance threshold is satisfied, wherein For the preset dominance coefficient, Judging whether the learning period set meets the preset learning period distribution constraint or not, wherein the learning period distribution constraint comprises the step of counting the number of independent learning periods contained in the learning period set of the end node corresponding to the prefix path When the inequality is satisfied When it is determined that the academic distribution constraint is satisfied, wherein For a preset minimum required learning period number, For integers greater than or equal to 1, only reserving prefix paths meeting the conditions for generating a conditional FP tree, and obtaining frequent patterns comprising group characteristics and time sequence information through recursive mining; and introducing a learning period distribution constraint during mining, so that the mined frequent pattern can reflect rules related to a learning stage or time evolution, and the time sequence information of the data is organically integrated into analysis.
2. The student performance report data management method of claim 1, wherein the dividing students into different student groups according to historical academic data and course selection behavior comprises: And (3) extracting average score points of each student as historical academic data characteristics, counting the number of selected classes of lessons, humanities and social sciences class of lessons as lesson selection behavior characteristics, vectorizing the historical academic data characteristics and the lesson selection behavior characteristics, and dividing the whole students into K student groups by adopting a K-means clustering algorithm, wherein K is a preset integer, and K is more than or equal to 2.
3. The method for managing student performance report data according to claim 1, wherein the calculating and obtaining, for each student group, the item ranking index corresponding to the student group based on the statistical distribution characteristics of the items of each performance level in the student group comprises: And counting the total number of times of each achievement level item in the student group aiming at each student group, and taking the total number of times as an item ordering index of the achievement level item in the student group.
4. The student achievement report data management method of claim 1, wherein the inserting the descending ordered item sequence as a path into the FP-tree and updating the group count vector and the learning period set of each node on the path comprises: When the item sequence is inserted into the FP tree as a path, for each node on the path, adding 1 to the component corresponding to the group to which the student belongs in the group count vector of the node, and meanwhile, adding the learning period information corresponding to the student performance record into the learning period set of the node, and if the learning period information exists, not repeating the adding.
5. A system for applying the student achievement report data management method as set forth in claims 1 to 4, comprising the following modules: The classification module is used for dividing students into different student groups according to historical academic data and course selection behaviors, converting the course achievements of each department of the students into discrete achievement grade items, and calculating and obtaining item ordering indexes corresponding to the student groups according to the statistical distribution characteristics of each achievement grade item in the student groups aiming at each student group; The initialization FP tree module is used for initializing an FP tree, wherein nodes of the FP tree comprise item names, group count vectors used for recording the group count of each student and a learning period set used for recording learning period information of a path where the nodes are located; The updating module is used for traversing each student score record, descending order the score grade items in the student score records according to the item ordering indexes of the student groups to which the student score records belong, inserting the item sequences subjected to descending order into the FP tree as paths, and updating the group count vector and the school period set of each node on the paths; The mining module is used for acquiring a group count vector and a learning period set of the end node corresponding to each prefix path when a condition mode base is constructed for the items in the item head table, judging whether the group count vector meets a preset group dominance threshold value, judging whether the learning period set meets a preset learning period distribution constraint, only reserving the prefix paths meeting the conditions for generating a condition FP tree, and obtaining a frequent mode comprising group characteristics and time sequence information through recursive mining.
6. The student performance report data management system of claim 5, wherein the dividing students into different student groups according to historical academic data and course selection behavior comprises: And (3) extracting average score points of each student as historical academic data characteristics, counting the number of selected classes of lessons, humanities and social sciences class of lessons as lesson selection behavior characteristics, vectorizing the historical academic data characteristics and the lesson selection behavior characteristics, and dividing the whole students into K student groups by adopting a K-means clustering algorithm, wherein K is a preset integer, and K is more than or equal to 2.
7. The student performance report data management system of claim 5, wherein the calculating and obtaining, for each student group, the item ranking index corresponding to the student group based on the statistical distribution characteristics of each performance level item in the student group comprises: And counting the total number of times of each achievement level item in the student group aiming at each student group, and taking the total number of times as an item ordering index of the achievement level item in the student group.
8. The student achievement report data management system of claim 5 wherein the inserting the descending ordered sequence of items as a path into the FP-tree and updating the population count vector and the learning period set for each node on the path comprises: When the item sequence is inserted into the FP tree as a path, for each node on the path, adding 1 to the component corresponding to the group to which the student belongs in the group count vector of the node, and meanwhile, adding the learning period information corresponding to the student performance record into the learning period set of the node, and if the learning period information exists, not repeating the adding.

Description

Student score report data management method and system Technical Field The invention relates to the technical field of electronic data processing, in particular to a student score report data management method and system. Background Student performance data mining plays a key role in the current education field, and can help us to deeply analyze and understand learning behaviors of students, forecast potential academic risks and finally realize personalized teaching. The core technology of performance data mining is association rule mining, which aims at finding those valuable relationships between data items, e.g. associations that exist between combinations of specific courses and student's academic performance. Classical association rule mining algorithms, such as FP-Growth, effectively avoid generating massive candidate sets by constructing a data structure named as an FP tree, thereby remarkably improving mining efficiency. However, such methods typically analyze all students as a homogenous whole, ignoring natural differences between student populations, such as different professional contexts, entrance bases, or learning habits, and their performance patterns and behavior laws may be quite different. To address this problem, some studies have attempted to employ a strategy of "clustering+association rules". The students are first divided into different groups using a clustering algorithm, and then FP-Growth mining is performed independently for each group. However, this approach breaks apart the clustering and mining processes, not only increasing the complexity of the overall computation, but also making it difficult to perform lateral comparisons and associative analysis of patterns of different clusters under a unified framework. In addition, some researches introduce methods such as sequence pattern mining to analyze time attributes of performance data, and although the methods are helpful to reveal time sequence rules in the learning process of students, the methods are often incompatible with an FP-Growth algorithm on a data structure and a mining target, so that it is difficult to comprehensively analyze the group characteristics, item set relevance and time sequence information of students integrally, and the depth of the mining result is insufficient and the accuracy is low. Disclosure of Invention The invention provides a student score report data management method and system, which are used for solving the problems that in the prior art, the group characteristics of students, the relevance of item sets and time sequence information are difficult to comprehensively analyze, so that the depth of an excavation result is insufficient and the accuracy is low. In a first aspect, the student achievement report data management method of the present invention includes the following steps: Dividing students into different student groups according to historical academic data and course selection behaviors, converting the course achievements of the students into discrete achievement grade items, and calculating and obtaining item ordering indexes corresponding to the student groups according to the statistical distribution characteristics of the achievement grade items in the student groups aiming at the student groups; initializing an FP tree, wherein nodes of the FP tree comprise item names, group count vectors used for recording the group count of each student and a learning period set used for recording learning period information of a path where the nodes are located; Traversing each student score record, descending order the score grade items in the student score records according to item ordering indexes of student groups to which the student score records belong, inserting the item sequences subjected to descending order into the FP tree as paths, and updating group count vectors and school period sets of all nodes on the paths; When a condition pattern base is constructed for items in the item header table, a group count vector and a learning period set of an end node corresponding to each prefix path are obtained, whether the group count vector meets a preset group dominance threshold value or not is judged, whether the learning period set meets a preset learning period distribution constraint or not is judged, only prefix paths meeting the conditions are reserved for generating a condition FP tree, and a frequent pattern comprising group characteristics and time sequence information is obtained through recursive mining. Preferably, the dividing students into different student groups according to the historical academic data and the course selection behaviors comprises the steps of extracting average score points of each student as historical academic data characteristics, counting the number of selected classes of course gates of the selected students, taking the historical academic data characteristics and the course selection behavior characteristics as course selection behavior character