Master in Data Science (SMS, TU) Fundamentals of Data Science Question Paper 2080 Nepal

Elaborate on agility (agile implementation) of CRISP-DM.

Explain why data preprocessing is a major job of a data scientist.

Why is logistic regression a regression despite its use for classification tasks? Illustrate the sigmoid curve along with its mathematical function.

Define entropy. How is it used with decision tree?

Explain why data ethics is crucial in data science.

Explain the OSEMN framework for data science project implementation with any suitable example of your own.

OR

Discuss the scope and limitations for data analysis of election data. What responsibilities should be fulfilled by a data scientist for such projects. Discuss with one example of your own.

Discuss the problems caused by data quality during the training of machine learning algorithms. Do you recommend data first approach or model first approach for any of the data science problems? Explain with reason.

Apply Naïve bayes algorithm to decide if Married Female with salary of 42000 is probable to have illness or not based on data given below:

OR

Explain the forward propagation and backward propagation of neural networks.

What do you mean by Random Forests? Why is random forest commonly used for feature selection despite being a machine learning model? Is random forest prone to over fitting? Why or why not?

A healthcare organization developed a machine learning model to predict patients' risk of developing certain medical conditions based on their electronic health records (EHR). The model was trained using historical patient data, including diagnoses, treatments, and outcomes. However, it was later discovered that the model exhibited bias against patients from lower socioeconomic backgrounds. The training data disproportionately represented patients from wealthier neighborhoods who had better access to healthcare services and resources. As a result, the model erroneously associated higher socioeconomic status with lower health risks, leading to underestimating the risk of certain conditions for patients from disadvantaged backgrounds.

Explain how you can address the biases in such a situation. Be specific and suggest remedies to avoid biases.