The presented work is a student marks and grade prediction system using supervised machine learning techniques, the system is developed on the historic performance of students. The data used in this research is collected from Federal Board of Intermediate and Secondary Education Islamabad Pakistan, there are 7 regions in FBISE i.e. Punjab, Sindh, Khyber Pakhtunkhwa, Balochistan, Azad Jammu and Kashmir and overseas. The aims of this work is to analyze the education quality which is closely tightened with the sustainable development goals. The implementation of the system has produced an excess of data which must be processed suitably to gain more valuable information that can be more useful for future development and planning. Student marks and grade prediction from their historic academic data is a popular and useful application in educational data mining, so it is becoming a valuable source of information which can be used in different manners to improve the education quality in the country. Related work shows that several method for academic grade prediction are developed for the betterment of teaching and administrative staff of an educational organizational system. In our proposed methodology, the obtained data is preprocessed to improve the quality of data, the labeled student historic data (29 optimal attributes) is used to train decision tree classifier and regression model. The classification system will predict the grade while the regression model will predict the marks, finally the results obtained from both the model are analyzed. The obtain results show the effectiveness and importance of machine learning technology in predicating the students performance.

呈现的作品是使用监督的机器学习技术的学生成绩和成绩预测系统，该系统是根据学生的历史表现开发的。这项研究中使用的数据是从巴基斯坦伊斯兰堡联邦中级和中等教育委员会收集的，FBISE中有7个地区，即旁遮普，信德省，开伯尔·普赫图赫瓦，Bal路支斯坦，阿扎德·查mu和克什米尔以及海外。这项工作的目的是分析与可持续发展目标紧密相关的教育质量。该系统的实施产生了过多的数据，必须对这些数据进行适当的处​​理，以获取更多有价值的信息，这些信息对于将来的开发和计划更有用。从他们的历史学术数据中得出的学生成绩和成绩预测在教育数据挖掘中是一种流行且有用的应用程序，因此它正成为一种有价值的信息来源，可以以各种方式用于提高该国的教育质量。相关工作表明，已经开发了几种用于预测学业成绩的方法，以改善教育组织系统的教学和管理人员。在我们提出的方法中，对获得的数据进行预处理以提高数据质量，使用标记的学生历史数据（29个最佳属性）来训练决策树分类器和回归模型。分类系统将预测等级，而回归模型将预测分数，最后分析从两个模型获得的结果。