Student Marks Prediction Using Linear Regresion

This research looks at how machine learning applications effect higher education teaching and learning, as well as how to improve the learning environment. Data can be anything related to population, academic data of students, and interests of people. New data appears from time to time, as we can see.. Analysing the data is the difficult task for humans. So here comes the computer, which can analyse data faster than people because it is stored digitally and in a well-formatted manner. This is where the machine learning emerged. Machine learning is the branch of Artificial Intelligence that provides ability to automatically learn from past experiences. Here the machines do get programmed explicitly. It gives the computer the power to make humans and machines look alike in terms of learning, as the name implies. Machine learning is divided into two categories based on the nature of the learning signal: supervised learning and unsupervised learning. This study focuses on supervised learning, more specifically on predictive analysis. When it comes to making predictions about future outcomes, predictive analysis is crucial. Predictive analysis has a wide range of applications. Predicting a student's academic success is critical since it can alert professors to students who may drop out of the course, and it can provide valuable information. Additional help to the scholars who want to enhance their educational performance. This have a look at is on implementation of system mastering in education. The outcome of this study is to predict student‘s academic performance. Students' data is utilised to create a model that can predict a student's academic achievement based on some background information. The dataset created by the students should be used as the study's input data.

1. In this studies paper creator carried out the thesis the usage of SVM technique in java, selection tree, C4.5, Naive Bayes, Lib. SVM, Logistic Regression and Hybrid technique LMT and as compared the accuracy of overall performance prediction most of the hybrid approaches. I have alluded to several research articles that are related to the thesis in order to specify the thesis as a well-structured thought. 2. In this studies paper it's far discovered that the writer used a number of the maximum of Polo, which included 700 courses. The article concludes that decision Trees and SVM produce the best results. The main contribution of this work is to compare the levels of accuracy of several algorithms. 3. The studies is targeted on predicting student‘s overall performance the usage of personalised analytics. This paper presents two different approaches to work on the thesis. The author's initial technique is the Regression Algorithm, which is a data mining function. The root mean square method is also used to calculate the regression algorithm's error rate. In this paper the author worked on how to improve the prediction algorithms which are used to analyze and predict the student‘s performance. The decision trees algorithm is used in this paper's work. 4. This paper proposed the student Academic performance prediction using Support Vector Machine. The author compared SVM to various machine learning approaches such as linear regression, Decision Trees, and KNN and determined that SVM outperformed them.

The principal step within side the implementation is to gather the statistics set required for the studies work. The technique is carried out to the dataset containing the statistics of the students. To simplify our analysis, we can identify the data set's unique attributes and delete them because they can't be used for analysis. The data is collected and then translated into the desired format. This process is called as pre-processing of data. It is the most crucial stage in obtaining the specific necessary info from the raw data. The higher the accuracy rate, the better. The more raw data is preprocessed, the higher the rate accuracy of acceptable data. After pre-processing the data, the following step is to detect and eliminate any incomplete or irrelevant data from the dataset in order to achieve correct findings. Data cleaning is the process of removing unnecessary data. For improved classification, we can use any of the techniques available, such as linear regression, support vector machine, Naive Bayes Standard Classification, and decision tree algorithms. In this research, the linear regression algorithm is used to implement the solution. We must also select a training set from the dataset, determine the Result attributes that determine the output, and begin classification.

[1] Hadzic and D. R. Morgan (2009). "On packet selection criteria for clock recovery," Proceedings of the National Academy of Sciences, vol. IEEE Int. Symp. Precision Clock Synchronization Meas. Contr. Commun. [2] C. S. Turner (2008). ―Slope filtering: an FIR approach to linear regression", IEEE Signal Process. Mag., vol. 25, pp. 159-163. [3] D. Veitch, J. Ridoux and S. B. (2009). ―Robust synchronization of absolute and difference clocks over networks," by Korada in IEEE/ACM Trans. Networking, vol. 17, pp. 417-430. [4] D. R. Morgan and I. Hadžić: Non-uniform linear regression with block wise sample-minimum preprocessing", IEEE Trans. Signal Process [5] Ankitha A Nichat, Dr. Anjali B Raut (2017). ―Predicting and Analysis of student Performance Using Decision Tree Technique‖, International Journal of Innovative Research in Computer and Communication Engineering, Vol.5, Issue 4. [6] S.A. Oloruntoba, J.L. Akinode (2017). ―Student Academic Performance Prediction Using Support Vector Machine‖. [7] Dhanashree Mane, Pranali Namdas, Pooja Gargade, Dnyaneshwari Jagtap, S.S. Rathi (2018). Predicting student Performance Using Machine Learning Approach‖. VJER Vishwakarma Journal of Engineering Research, Volume 2 Issue 4.

TY Student, Department of Computer Engineering, Sahakar Maharshi Shankarrao Mohite Patil Institute of Technology and Research, Akluj, Solapur, Maharashtra, India