Enhancing Student Performance for Low Rank Students using Adaptive Deep Multi-Perception Pattern Learning Classification Model

The tendency of student ability is important for knowledge improvement based on the quality of student learning skills. In recent, the educational services are affected by COVID-19 in continues pandemic period to train the students. There is difficult to process the student features which its leads large collective data forums in unstructured principle. Moreover the large number features creates high dimensions due to improper assets in student information along with grades, performance activities, curriculum etc. so the essential thing is feature based dimension reduction, because the existing prediction model failed to observe the feature weights. To overcome the issues, the novelty creates an adaptive prediction model to classify the results. To propose a Habitual Impact scaling feature selection based student performance prediction via adaptive Deep Multi Perception Pattern Learning Classification (DMP 2 LC) model. Initially this pre-process take place to reduce the noise and non-redundant date reduction to formalize the dataset which is collected from college academic student features dataset. Then features weight scaling factors are verified by Threshold margins to get the structural data. Then the student interest successive rate (SISR) weight are observed by the threshold values by habitual interest of the student learning. Each threshold SISR is scaled into sub-spectral clustering weight (S 2 CW) to group the important features related to academic attribute, then other margins are neglected to reduce the dimension. Then the Subjective Pattern sequence rate (SPSR) is generated to predict the correlation features. Finally the SPSR is trained into Deep Multi Perception Pattern Learning Classification (DMP 2 LC) model to classify the result into categorized classes. From the result low rank predicted students get high preference to motivate to improve the academic performance. This DMP 2 LC high performance in contrast to the alternative existing system additionally in classification accuracy, precision, recall, false rate and time consumption.