Learning residual workload estimation method based on latent semantic model

Abstract

The invention discloses a learning residual workload estimation method based on a latent semantic model, and belongs to the field of intelligent education, and the method specifically comprises the steps: firstly, dividing test questions of m students according to the mastering conditions of the students for a certain knowledge point K, and forming a two-dimensional matrix S; after converting into a one-dimensional array, sorting elements in the array in an ascending order to obtain a set T, selecting an element with a value of 1 as a subset T0, equally dividing the remaining elements into n-1 subsets, and selecting an interval window; for a certain student u, respectively scanning the knowledge point mastery degree sequence of the student by using n interval windows, and then carrying out Map mapping to obtain the number of test questions completed by the student u in each interval window; and similarly, scanning m students one by one, constructing a completed test question number matrix, inputting an LFM algorithm, and predicting the number of test questions needing to be completed when the mastery degree of the current student u to the knowledge point K reaches 100%. According to the invention, the remaining learning workload to reach the learning target can be estimated in advance, so that the students can reasonably manage the learning time and improve the learning efficiency.

Description

technical field [0001] The invention belongs to the field of intelligent education, in particular to a method for estimating remaining workload of learning based on a latent semantic model. Background technique [0002] The remaining learning workload refers to the number of test questions that need to be practiced when the user learns a knowledge point from the current knowledge point mastery level to the knowledge point mastery level reaching 100%. [0003] The user's mastery of knowledge points is calculated according to the item response theory. Since the traditional item reflection theory assumes that the student's ability value remains unchanged during the problem-solving process, this fundamentally negates the possibility of learning. Therefore, the traditional item response theory Response theory models have been refined and optimized to support that a student's cognitive abilities can change over time. A system that describes dynamic learning must first define what...

Images