SMFG India Credit Interview Questions and Answers

Algorithms are the backbone of machine learning, used to train models and make predictions.

• Algorithms are used to optimize the performance of machine learning models

• Common algorithms include linear regression, decision trees, and neural networks

• The choice of algorithm depends on the type of data and the problem being solved

• Algorithms can be supervised, unsupervised, or semi-supervised

• Hyperparameter tuning is important for optimizing algorithm performance

Model evaluation is the process of assessing the performance of a machine learning model.

• Splitting the data into training and testing sets

• Using metrics such as accuracy, precision, recall, and F1 score to evaluate the model

• Using cross-validation to evaluate the model on multiple splits of the data

• Visualizing the results using confusion matrices and ROC curves

• Comparing the performance of different models using statistical tests

Staging layer is used for data transformation and cleansing before moving to serving layer in Azure Data Factory.

• Staging layer is used for data transformation and cleansing before moving to serving layer

• Serving layer is used for serving the data to end-users or applications

• Staging layer can be used to perform operations like filtering, sorting, and aggregating data

• Serving layer can be used to store data in a format that is optimized for querying and analysis

• Staging layer can ...read more

Action triggers computation and returns results to driver while transformation creates a new RDD from existing one.

• Action is a command that triggers computation and returns results to the driver program.

• Transformation creates a new RDD from an existing one without computing the result immediately.

• Actions are executed immediately while transformations are executed lazily.

• Examples of actions include count(), collect(), and reduce().

• Examples of transformations include map(), fil...read more

SQL elastic pool provides cost savings, resource optimization, and simplified management.

• Cost savings by sharing resources among multiple databases

• Resource optimization by dynamically allocating resources based on demand

• Simplified management by managing multiple databases as a single entity

• Improved performance by reducing contention for resources

• Ability to scale up or down based on changing workload demands

Azure Databricks supports multiple programming languages.

• Azure Databricks supports Python, R, Scala, SQL, and Java.

• Users can also install additional libraries and packages.

• The choice of language depends on the specific use case and user preference.

• Databricks also provides a notebook interface for easy code development and collaboration.