Data Science Technical Interview Questions (MCQ): Test Your Knowledge! What is the purpose of fine-tuning in transfer learning?To make the model slowerTo adapt a pre-trained model to a new taskTo perform clusteringTo increase model complexityWhat is the main purpose of the DBSCAN algorithm?To scan databasesTo cluster data based on densityTo classify data pointsTo reduce data noiseData science involves processing diverse sets of data through:Analysing dataProcessing dataOrganizing dataAll of the aboveWhich technique is commonly used for handling imbalanced datasets?Increasing model complexityOversampling minority classUsing only majority classIgnoring minority classWhat is the main difference between parametric and non-parametric tests?Sample size requirementsAssumptions about population distributionComplexityAccuracyThe primary purpose of cross-validation in machine learning is:To increase model complexityTo assess model performance on unseen dataTo speed up trainingTo create more featuresWhich algorithm is best suited for regression tasks?Linear RegressionK-Means ClusteringDecision TreeNaive BayesWhat is the purpose of a cold start problem in recommender systems?To handle new users or items with no historical dataTo select the best features for a modelTo reduce data sizeTo visualize model performanceWhat is the primary purpose of building a decision tree?To visualize dataTo make sequential decisions based on featuresTo perform clusteringTo reduce dimensionalityWhat is the main purpose of the attention mechanism in deep learning?To reduce model attentionTo allow the model to focus on different parts of the inputTo perform clusteringTo reduce dimensionalityWhat is the purpose of the margin in SVM?To speed up trainingTo maximize separation between classesTo reduce overfittingTo simplify model interpretationIn model evaluation, ROC stands for:Rate of ChangeReceiver Operating CharacteristicRandom Oscillation CurveRecursive Operational CalculationWhat is the primary focus of descriptive analytics?Predicting future trendsSummarizing and visualizing past dataPrescribing optimal actionsUncovering hidden patternsWhat is the main purpose of the Swish activation function?Binary classificationMulti-class classificationTo provide a smooth, non-monotonic functionFeature scalingWhat is the main purpose of the dropout technique?To remove data pointsTo prevent overfitting in neural networksTo classify dropout ratesTo reduce network sizeWhat is the Q-learning alg.? Model-free RL algorithm Learns action-value function Uses Q-table/function All of the aboveWhat is the main purpose of data preprocessing?To analyze data insightsTo visualize data distributionsTo transform raw data into usable formatTo evaluate model performanceExamples of artificial intelligence and machine learning?Image recognition and voice assistantsManual data entry and rule-based systemsFile management and word processingStatic websites and hard-coded logicWhat is Fisher Scoring in log. regression? Optimization algorithm Similar to Newton-Raphson Used for max. likelihood est. All of the aboveWhat does the term "bias" refer to in machine learning?A systematic error in the model's predictionsA random error in data collectionThe process of clustering data pointsA method for data visualizationWhich algorithm is best suited for anomaly detection in streaming data?K-meansIsolation ForestDBSCANHDBSCANWhat is the difference between a convolutional neural network (CNN) and a recurrent neural network (RNN)?CNNs are good for image data, RNNs are good for sequential dataCNNs are used for classification, RNNs are used for regressionCNNs are faster than RNNsCNNs require more data than RNNsWhat is the main advantage of using Transformer models over RNNs in NLP?Faster trainingBetter parallelizationBetter handling of long-term dependenciesAll of the aboveWhat does ROI stand for in time series analysis?Return on InvestmentRegion of InterestRate of IncreaseRecursive Outlier IdentificationThe main idea behind autoencoders is:To classify data pointsTo compress and reconstruct dataTo perform clusteringTo increase dimensionality Score: 0/25 Retake Quiz Next Set of Questions