Meta Data Science Interview Questions & Answers

Bias occurs when the sample does not represent the population. Reduce it with random or stratified sampling, higher response rates and by checking sample demographics against the population.

The probability of seeing a result at least as extreme as the observed one if the null hypothesis were true. A small p-value is evidence against the null, not proof that the alternative is true.

It normalises layer inputs per mini-batch and then rescales them, stabilising and speeding up training and allowing higher learning rates.

Add a classification head, freeze or gradually unfreeze layers, use a small learning rate, train on labelled data with validation, and track F1 per class.

Pearson measures linear relationships on continuous data; Spearman uses ranks, so it captures monotonic relationships and is less sensitive to outliers.

merge joins on columns, SQL-style (how='inner', 'left', ...). join joins on the index. concat stacks DataFrames along rows or columns.

It predicts from the k nearest labelled points. It is simple but slow at prediction time, sensitive to scaling and irrelevant features, and weak in high dimensions.

This is overfitting. Check for data leakage, simplify the model, add regularization, get more data or augment it, use cross-validation, early stopping and feature selection.

CNNs use convolutions to find spatial patterns (images). RNNs/LSTMs process sequences step by step but struggle with long context. Transformers use self-attention to look at all tokens at once — parallel and better at long-range context; they power modern LLMs.

A measure of asymmetry. Right-skewed data has a long tail on the right (mean greater than median); left-skewed is the opposite. Log transforms often reduce right skew.

It combines gradient boosting with regularisation, handling of missing values, parallel tree construction and speed optimisations, and it performs strongly on tabular data.

Train fits the model, validation tunes hyperparameters and model choice, and the untouched test set gives a final unbiased estimate of performance.

Small anonymous one-line functions, handy as arguments to sort, map, filter or Pandas apply.

RNNs process sequences with a hidden state but struggle with long dependencies. LSTMs add gates that control what to remember or forget, easing the vanishing gradient.

Use grid search, random search or Bayesian optimisation with cross-validation, starting with a coarse random search and narrowing the promising ranges.

Monitor input distributions and live metrics, compare to training baselines, and retrain or recalibrate on fresh data when performance drops.

Package the model behind an API with versioning, track latency and errors, monitor input drift and prediction quality, and automate retraining with rollback.

They build a list in one readable expression, for example [x*x for x in data if x > 0]. They are usually faster than an equivalent for-loop with append.

Raising the threshold improves precision but lowers recall. Fraud detection leans to recall, while spam filters often prefer precision; choose by the cost of each error.

Putting features on comparable ranges using standardisation or min-max scaling. Distance-based and gradient-based models (KNN, SVM, regression, neural nets) need it; tree models do not.

P(A given B) = P(B given A) x P(A) / P(B). In a medical test it converts test accuracy and disease prevalence into the real chance a positive result means disease.

Measure first with cProfile or timeit, then replace loops with NumPy or Pandas vectorised calls, use built-ins, caching and better data structures.

It assigns points to the nearest centroid and updates centroids until stable. Choose k with the elbow method, silhouette score or business context.

It checks whether observed category counts differ from expected counts, used for independence between two categorical variables and for goodness of fit.

Frequentists treat parameters as fixed and use long-run frequencies; Bayesians treat parameters as distributions updated with data via priors and posteriors, giving direct probability statements.

Information from outside the training window, such as future data or test statistics, enters training and inflates performance. Split first, then fit scalers and encoders on training data only.

Reusing a model pretrained on a large dataset and fine-tuning it on a smaller task-specific dataset, saving data, time and compute.

*args collects extra positional arguments into a tuple; **kwargs collects extra keyword arguments into a dict. They make function signatures flexible and are common in wrappers and decorators.

List: ordered and mutable. Tuple: ordered and immutable (hashable, slightly faster). Set: unordered, unique items, fast membership checks. Dict: key–value pairs with fast lookup by key.

Tie metrics to business outcomes such as cost saved, show feature importance or SHAP values, give example predictions and be transparent about limitations.

Use stratified splits, class weights, resampling such as SMOTE, threshold tuning and judge models with precision, recall, F1 or PR-AUC instead of accuracy.

SVM finds the maximum-margin separating hyperplane. Kernels such as RBF compute similarity in a higher-dimensional space implicitly, so non-linear boundaries are possible.

A shallow copy duplicates the outer object but shares nested objects. A deep copy (copy.deepcopy) recursively duplicates everything, so changes to nested items do not leak.

Retrieval-Augmented Generation fetches relevant documents and gives them to an LLM as context, reducing hallucination and keeping answers grounded in current private data.

Modelling the count of events in a fixed interval when events occur independently at a constant average rate, such as calls per hour or defects per batch.

Split the data into k folds, train on k−1 folds and validate on the remaining one, rotate and average the scores. It gives a more reliable estimate than one split and helps tune hyperparameters without touching the test set.

Use filter methods (correlation, chi-square), wrapper methods (RFE), embedded methods (Lasso, tree importance) and domain knowledge, validating each choice with cross-validation.

Regression predicts continuous values such as price; classification predicts discrete categories such as spam or not spam.

The probability of seeing a result at least this extreme if the null hypothesis were true. A small p-value (for example < 0.05) is evidence against the null hypothesis.

Gradients shrink as they pass through many layers (for example with sigmoid), so early layers barely learn. ReLU, residual connections and good initialisation help.

Transformers use self-attention to process all tokens in parallel and capture long-range context, making them faster to train and more scalable than sequential RNNs.

Type I = false positive (rejecting a true null), probability α. Type II = false negative (missing a real effect), probability β. Power = 1 − β.

It projects data onto orthogonal directions of maximum variance to reduce dimensions, remove correlation and speed up models, at some cost of interpretability.

A range built from sample data that would contain the true parameter in a stated percentage (for example 95%) of repeated samples. It is not the probability that the parameter lies inside this one interval.

df.drop_duplicates(subset=[...], keep='first') removes repeated rows; use duplicated() first to inspect what will be dropped.

First understand why data is missing. Then drop rows or columns with too many nulls, impute with mean/median/mode or a model (KNN), add a 'was_missing' flag, and fit imputers on training data only.

Training a meta-model on the predictions of several base models, often using cross-validated predictions to avoid leakage.

== compares values for equality, while is checks whether two names point to the same object in memory. Use is mainly for None checks.

Convolution filters slide over the input to detect local patterns, pooling reduces size, and deeper layers combine edges into shapes and objects. Used mainly for images.

It models the log-odds of a binary outcome as a linear function and passes it through a sigmoid to give probabilities, trained by maximising likelihood.