Task1: Prediction using Supervised ML as Data Science & Business Analytics Intern at the Sparks Foundation for the batch #GRIPDECEMBER22. In this task is required to predict the student score if the student studies for 9.25hrs/day. Thank you, The Sparks Foundation, for providing me with this opportunity.
- Dataset Link: http://bit.ly/w-data Situation: I was tasked with predicting a student's score in an academic setting using a dataset of hours studied and corresponding scores. This was part of a data science internship program at the Sparks Foundation. The dataset was small and contained student study hours and their corresponding exam scores.
Task: The specific task was to predict the score of a student who studies for 9.25 hours per day, using the provided dataset.
Action: I loaded the dataset into a Python environment using pandas. I explored the data to understand the relationship between hours studied and exam scores. A simple scatter plot showed a positive correlation. Next, I split the data into training and testing sets to evaluate the model's ability to generalize. I chose a linear regression model for prediction because of the apparent linear relationship, trained it on the training data, and used it to predict the score for 9.25 hours. Crucially, I calculated the Mean Squared Error (MSE) on the testing set to quantify the model's accuracy. I evaluated different model types (e.g., linear regression, polynomial regression) to see which best fitted the trend. In this case, linear regression proved appropriate. I then used the trained linear regression model to predict the score for a student studying 9.25 hours. The final step was documenting the process, code, and results in a clear and concise manner.
Result: The predicted score for a student studying 9.25 hours per day was [insert calculated score]. The model performed with an MSE of [insert MSE value]. The result demonstrates a reasonable, [insert, e.g., positive, negative, or appropriate] relationship between study time and exam score based on the data, although further investigation using more data and/or different types of model analysis may yield better results and insight.