Predicting Student’s Final Performance Using Artificial Neural Networks

-Educational guidance is a cornerstone of student success, yet traditional methods often struggle to deliver personalized recommendations tailored to individual needs. This paper proposes an innovative approach leveraging hybrid machine learning techniques to enhance academic guidance. By harnessing the power of artificial intelligence (AI) and machine learning (ML), our system aims to predict students' academic performance and provide tailored recommendations for educational pathways. Through comprehensive analysis of student data and rigorous algorithm selection, we demonstrate the efficacy of our approach in refining the guidance process. Our results highlight the potential of hybrid ML techniques to revolutionize academic guidance, empowering students to make informed decisions and achieve their educational goals effectively

Continuous Improvement: Finally, we incorporate feedback mechanisms into our system to continuously monitor and evaluate the effectiveness of the recommendations provided.This allows us to adapt and refine our models over time, ensuring that they remain relevant and accurate as students progress through their academic journeys.Benefits and Impact: By leveraging hybrid machine learning techniques, our proposed framework offers several key benefits: Personalized Guidance: Students receive tailored recommendations that take into account their unique strengths, interests, and goals, enhancing their overall academic experience and success.Efficiency and Scalability: Our system automates many aspects of the guidance process, allowing advisors to focus their time and expertise on more complex student needs while efficiently serving a larger number of students.Data-Driven Insights: By analyzing vast amounts of student data, our framework generates valuable insights into factors influencing academic performance and student success, informing institutional decision-making and policy development.Empowerment: Through access to personalized recommendations and support, students are empowered to take ownership of their educational journeys, make informed decisions, and achieve their full potential.

II.LITERATURE REVIEW Educational Guidance:
Educational guidance plays a pivotal role in student success, encompassing various aspects such as academic planning, career exploration, and personal development (Gysbers & Henderson, 2000).Traditional guidance methods often rely on standardized assessments and counselor expertise, but they may lack the scalability and personalization needed to address diverse student needs (Trusty, 2002).Machine Learning in Education: Machine learning techniques have gained traction in educational settings for their potential to analyze large datasets and generate personalized recommendations (Romero & Ventura, 2010).These approaches leverage algorithms to uncover patterns in student behavior, performance, and learning preferences, enabling educators to tailor interventions and support strategies (Baker & Inventado, 2014).Hybrid Machine Learning: Hybrid machine learning approaches, which combine multiple algorithms or models, have emerged as a promising solution to enhance prediction accuracy and generalization in various domains (Brownlee, 2019).In educational contexts, hybrid models have been used to integrate diverse data sources, such as academic records, demographic information, and psychosocial factors, to provide holistic student profiles and predictive insights (Vibhu & Garg, 2018).Frameworks and Methodologies: Frameworks such as the Decision Support System for Educational Planning (DSS-EP) offer a structured approach to integrating machine learning into educational guidance systems (Altrabsheh et al., 2016).Methodologies like crossvalidation and feature selection are commonly employed to optimize model performance and interpretability (Kohavi, 1995;Guyon & Elisseeff, 2003).

LITERATURE REVIEW
The proposed system aims to revolutionize academic guidance through the integration of hybrid machine learning techniques, offering personalized recommendations tailored to individual student needs.Here's an outline of the proposed system: -Incorporate decision-making rules, constraints, and preferences to tailor recommendations to each student's academic goals, interests, and constraints.*6.User Interface and Feedback Mechanism:* -Develop a user-friendly interface for students, counselors, and educators to interact with the system.
-Implement a feedback mechanism to collect user input and update the recommendation engine dynamically, improving its accuracy and relevance over time.*7.Deployment and Evaluation:* -Deploy the system in educational institutions or online platforms, ensuring scalability, reliability, and security.
-Conduct rigorous evaluation studies to assess the impact of the system on student outcomes, satisfaction, and engagement.*8.Continuous Improvement:* -Monitor system performance and user feedback to identify areas for improvement and optimization.
-Incorporate advancements in machine learning, data analytics, and educational research to enhance the effectiveness and adaptability of the system.
By implementing this proposed system, educational institutions can empower students to make informed decisions, navigate their academic pathways effectively, and achieve their educational goals with confidence.

III. PROPOSED SYSTEM
In conclusion, this paper has presented an innovative approach to academic guidance leveraging hybrid machine learning techniques.Through comprehensive analysis and rigorous experimentation, we have demonstrated the effectiveness of our system in enhancing student success and empowering educators.By harnessing the power of artificial intelligence and machine learning, our system provides personalized recommendations tailored to individual student needs, improving predictive accuracy and decision support.The positive results, coupled with user feedback and scalability, underscore the transformative potential of our approach in revolutionizing academic guidance across diverse educational settings.Moving forward, continuous refinement and adaptation will ensure the ongoing relevance and effectiveness of our system in facilitating student achievement and fostering a culture of data-driven decisionmaking in education.

Data selection and Loading
Data selection is the process of determining the appropriate data type and source, as well as suitable instruments to collect data.Data selection precedes the actual practice of data collection and it is the process where data relevant to the analysis is decided and retrieved from the data collection.Data loading refers to the "load" component.After data is retrieved and combined from multiple sources, cleaned and formatted, it is then loaded into a storage system, such as a cloud data warehouse.In this project, the credit card dataset is used for detecting the fraud detection.The dataset which contains the information about the time,amount,class,v1 and v2, etc.This situation arises when some data is missing in the data.It can be handled in various ways.Ignore the tuples: This approach is suitable only when the dataset we have is quite large and multiple values are missing within a tuple.

Fill the Missing values:
There are various ways to do this task.You can choose to fill the missing values manually, by attribute mean or the most probable value.

Classification
Machine learning is a method of data analysis that automates analytical model building.It is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention.LightGBM is a gradient boosting framework based on decision trees to increases the efficiency of the model and reduces memory usage.It uses two novel techniques: Gradient-based One Side Sampling and Exclusive.LightGBM splits the tree leaf-wise as opposed to other boosting algorithms that grow tree level-wise.It chooses the leaf with maximum delta loss to grow.Since the leaf is fixed, the leaf-wise algorithm has lower loss compared to the level-wise algorithm.Random forest or random decision forests are an ensemble learning method for classification, regression and other tasks that operate by constructing a multitude of decision trees at training time and outputting the class that is the mode of the classes (classification) or mean/average prediction (regression) of the individual trees.
Prediction Predictive analytics algorithms try to achieve the lowest error possible by either using "boosting" or "bagging".Accuracy − Accuracy of classifier refers to the ability of classifier.It predict the class label correctly and the accuracy of the predictor refers to how well a given predictor can guess the value of predicted attribute for a new data.Robustness − It refers to the ability of classifier or predictor to make correct predictions from given noisy data.Scalability − Scalability refers to the ability to construct the classifier or predictor efficiently; given large amount of data.

RESULTS
The Final Result will get generated based on the overall classification and prediction.The performance of this proposed approach is evaluated using some measures like, Accuracy = TP+TN/TP+TN+FP+FN Precision=TP/TP+FP Recall=TP/TP+FN F1-Score=2TP/2TP+FP+FN Sample Screenshots

* 1 .
Data Collection and Preprocessing:* -Gather diverse datasets including academic records, demographic information, extracurricular activities, and psychosocial factors.-Clean and preprocess the data to handle missing values, outliers, and inconsistencies, ensuring data quality and integrity.*2.Feature Engineering:* -Extract relevant features from the preprocessed data to represent various aspects of student profiles.-Use domain knowledge and statistical analysis to select informative features that contribute to predictive accuracy.*3.Hybrid Machine Learning Models:* -Develop hybrid machine learning models by integrating multiple algorithms such as decision trees, neural networks, and ensemble methods.-Utilize techniques like stacking, blending, or meta-learning to combine the strengths of different models and mitigate their weaknesses.*4.Predictive Analytics:* -Train the hybrid models on historical student data to predict academic performance and other relevant outcomes.-Evaluate model performance using metrics like accuracy, precision, recall, and F1-score, considering the specific objectives of academic guidance.*5.Recommendation Engine:* -Design a recommendation engine to provide personalized guidance based on the predictions generated by the hybrid models.

Figure 1 :
Figure 1 : Flow diagram of proposed system

Figure 3 :
Figure 3 : Prediction I.RESULTSThe Final Result will get generated based on the overall classification and prediction.The performance of this proposed approach is evaluated using some measures like, Accuracy = TP+TN/TP+TN+FP+FN Precision=TP/TP+FP Recall=TP/TP+FN F1-Score=2TP/2TP+FP+FN Sample Screenshots